New Approach for Quantifying Process Feasibility: Convex and 1-D Quasi-Convex Regions

Size: px
Start display at page:

Download "New Approach for Quantifying Process Feasibility: Convex and 1-D Quasi-Convex Regions"

Transcription

1 New Approach for Quantfyng Process Feasblty: Convex and 1-D Quas-Convex Regons Maranth G. Ierapetrtou Dept. of Chemcal and Bochemcal Engneerng, Rutgers Unversty, Pscataway, NJ Uncertantes n chemcal plants come from numerous sources: nternal lke fluctuated alues of reacton constants and physcal propertes or external such as qualty and flow rates of feedstreams. Accountng for uncertanty n arous stages of plant operatons was dentfed as one of the most mportant problems n chemcal plant desgn and operatons. A new approach proposed descrbes process s feasble regon and a new metrc for e aluatng process flexblty based on the con ex hull that s nscrbed wthn the feasble regon and determnes ts olume based on Delaunay Trangulaton. The two steps n ol ed are: 1. a seres of smple optmzaton problems are sol ed to determne ponts at the boundary of the feasble regon;. g en the set of ponts at the boundary of the feasble regon, the con ex hull nscrbed wthn the feasble regon s determned. Ths s ache ed by mplementng the Quckhull algorthm, an ncremental procedure for e aluatng the con ex hull, and then by computng a Delaunay Trangulaton to determne the olume of the con ex hull pro dng a new metrc for process flexblty. Ths approach not only pro des another feasblty measure, but an accurate descrpton of the feasble space of the process. It was appled to 1-D con ex problems, and work s n progress to extend t to noncon ex systems. Introducton Producton systems typcally nvolve sgnfcant uncertanty n ther operaton. Varablty of process parameters durng operaton and plant model msmatch Žboth parametrc and structural. could gve rse to suboptmalty and even nfeasblty of the determnstc solutons. Consequently, plant flexblty has been recognzed to represent one of the mportant components n the operablty of the producton processes. A bref overvew of the approaches that exst n the lterature to deal wth the problem of feasblty and flexblty quantfcaton follows n ths secton. In a broad sense the area covers Ž. a feasblty test that requres constrant satsfacton over a specfed space of uncertan parameters, Ž. a flexblty ndex assocated wth a gven desgn that represents a quanttatve measure of the range of uncertanty space that satsfes the feasblty temperature, and Ž. the ntegraton of desgn and operatons where trade-offs between desgn cost and plant flexblty are consdered. Halemane and Grossmann Ž proposed a feasblty measure for a gven desgn based on the worst ponts for feasble operaton, whch can be mathematcally formulated as a max-mn-max optmzaton problem as follows Ž d. s max mn g T z max jg x, g I 4 h Ž d, z, x,. s0; g Ž d, z, x,. F0, Ž 1. where the functon Ž d. represents a feasblty measure, d corresponds to the vector of desgn varables, z the vector of control varables, x the vector of state varables, and the vector of uncertan parameters. If Ž d. F0, desgn d s feasble for all gt, whereas f Ž d. 0, the desgn cannot operate for at least some values of gt. The above mass-mn-max problem defnes a nondfferentable global optmzaton problem, whch, however, can be reformulated as the followng two-level optmzaton problem j AIChE Journal June 001 Vol. 47, No

2 Ž d. s max Ž d,. g T s.t. h Ž d,. s mn u z,u Ž d, z, x,. s0; g I gjž d, z, x,. Fu jg J, Ž. where the functon Ž d,. s0 defnes the boundary of the feasble regon n the space of the uncertan parameters. The desgn flexblty ndex problem as ntroduced by Swaney and Grossmann Ž 1985a. can be reformulated to represent the determnaton of the largest hyperrectangle that can be nscrbed wthn the feasble regon of the desgn. Followng ths dea, the mathematcal formulaton of the flexblty problem has the followng form F smn s.t. Ž d,. s0 Ž d,. s mn u z hž d, z, x,. s0, g I gjž d, z, x,. Fu, jg J N y N T Ž. s N y F F y q 4 G0. Ž. 3 Other approaches that exst n the lterature to quantfy the flexblty for a gven desgn nvolve the determnstc measures such as the Reslence Index Ž RI., proposed by Saboo et al. Ž 1983., the flexblty ndex proposed by Swaney and Grossmann Ž 1985a,b. and the stochastc measures such as the desgn relablty proposed by Kubc and Sten Ž 1988., the stochastc flexblty ndex proposed by Pstkopoulos and Mazzuch Ž and Straub and Grossmann Ž For the case where the constrants are one-dmensonal Ž1- D. jontly quas-convex n and quas-convex n z, t was proven by Swaney and Grossmann Ž 1985a. that the pont c that defnes the soluton to Eq. les at one of the vertces of the parameter set T. Based on ths assumpton, the crtcal uncertan parameter ponts correspond to the vertces and the feasblty test problem s reformulated n the followng manner Ž d. s max Ž d, k., Ž 4. k g V where Žd, k. s the evaluaton of the functon Ž d,. at the parameter vertex k and V s the ndex set for the n p vertces for the np uncertan parameters. In a smlar fash- on for the flexblty ndex, problem 3 s reformulated n the followng way F s mn k, Ž. 5 k g V where k s the maxmum devaton along each vertex drecton k, kgv and s determned by the followng problem s.t. g h k s max, z Ž d, z, x,. F0, jg J j Ž d, z, x,. s0, g I s N q k, G0. Ž. 6 Based on the above problem reformulatons, Halemane and Grossmann Ž proposed the drect search method that explctly enumerates all the parameter set vertces. To avod the explct vertex enumeraton, Swaney and Grossmann Ž 1985a,b. proposed two algorthms, a heurstc vertex search, and an mplct enumeraton scheme. These algorthms rely on the assumpton that the crtcal ponts correspond to the vertces of the parameter set T whch s vald f the constrants are jontly 1-D quas-convex n and quas-convex n z. To crcumvent ths lmtaton, Grossmann and Floudas Ž proposed a soluton approach based on the followng deas: Ž. a They replace the nner optmzaton problem by the Karush Kuhn, Tucker optmalty condtons Ž KKT.; Ž b. They utlze the dscrete nature of the selecton of the actve constrants by ntroducng a set of bnary varables to express f a specfc constrant s actve. Based on these deas, the feasblty test and flexblty ndex problems can be reformulated as mxed-nteger optmzaton problems ether lnear or nonlnear dependng on the nature of the constrants. Grossmann and Floudas Ž proposed the actve set strategy for the soluton of the above reformulated problems based on the property that for any combnaton of n q1 bnary varables that s selected Ž z that s, for a gven set of actve constrants., all the other varables can be determned as a functon of. They proposed a procedure of systematcally dentfyng the potental canddates for the actve sets based on the sgns of the gradents g Ž d, z, x,.. Ostrovky et al. Ž z j proposed a branch and bound approach based on the evaluaton of the upper and lower bound of functon Ž d.. Although the suggested boundng problems are smpler than the orgnal feasblty test problem, they correspond to blevel optmzaton problems where global optmalty cannot be guaranteed usng local optmzaton methods Ž Mgdalas et al., Fnally, a global optmzaton approach s proposed recently by Floudas et al. Ž 000. to guarantee the determnaton of the global optmal soluton for both the feasblty and the feasblty ndex problem by generatng a relaxatonrenlargement of the feasble regon based on the convexfcaton of the orgnal problem constrants. Followng ths ntroducton, a small example s studed n the next secton to motvate the need of developng a new approach for quantfyng the range of feasblty for a gven process. The new approach and the detaled procedure that we propose s ntroduced to determne the new metrc that descrbes the feasble regon. The motvatng example s used to llustrate the basc steps of the proposed approach and the results obtaned. Several examples from the lterature are 1408 June 001 Vol. 47, No. 6 AIChE Journal

3 covered and compared wth the results of the prevously used approaches. The extenson of the basc framework to cover 1-D convex regons s dscussed and the basc advantages of the proposed approach are summarzed. Motvatng Example The example presented here s a modfcaton of example used by Pstkopoulos and Ierapetrtou Ž The desgn s descrbed by two parameters Ž d, d. 1 whereas z 1, z are the control varables and 1, are the uncertan parameters. The constrants of the problem are the followng f sy1.3q1.6 y1.6 z q.14 z F f s y.5 q1 z y z F0 1 1 f s.61 z y d F f s3.14 z y d F0 4 f sy z F0 5 1 f sy z F0 6 0F F 1 0F F8. The desgn examned frst corresponds to d1s3.5, ds0.0. For ths problem, frst the actve sets are dentfed based on the fact that three constrants should be actve at a tme snce there are two control varables ŽGrossmann and Floudas, Actve Set 1s f 1, f, f34 Actve Set s f 1, f, f54 Actve Set 3s f 1, f, f64 Actve Set 4s f 1, f, f44 Actve Set 5s f 1, f 3, f64 Actve Set 6s f, f 4, f54 Fgure 1. Feasble regon of Desgn ). ( d s3.5, d s 1 The results presented for the specfc example clearly motvate the need of developng a new approach for determnng the feasblty range for a gven process and a new metrc for evaluatng ts flexblty that can be used for comparson between desgns. Proposed Approach To llustrate the basc concept of the proposed approach, let s assume that the specfc desgn s descrbed by a set of nequalty constrants f Ž d, z,. j F0 assumng that the equal- ty constrants have been elmnated for ease n the presentaton. In these constrants d represents the set of desgn varables, and z the set of control varables that can be adjusted to accommodate varatons on the uncertan parameter vector. Followng the deas of Grossmann and Halemane Ž 198., one can map the feasble regon nto the uncertanty For each one of these actve sets, the feasblty functon s determned and plotted n Fgure 1. The flexblty ndex problem s then solved usng the actve set strategy approach proposed by Grossmann and Floudas Ž that results n a value of F s0.78. Ths value descrbes the rectangular whch s shown dark shaded n Fgure 1. Note that, although the actual feasble regon of the desgn s the lght shaded regon, the flexblty ndex dentfes only a small percentage of t and thus t largely underestmates desgn s feasblty. Moreover, the flexblty ndex does not provde always accurate results for comparng dfferent desgns. For example, we are consderng next the desgn that corresponds to d s.5, d s3.5. For ths desgn, the feasble 1 regon s evaluated based on the determnaton of the dfferent actve sets. The results are llustrated n Fgure. Note that ths desgn has a smaller feasble regon snce actve set 5 constrant moved to the left and thus the desgn s feasble regon s more restrcted. However, the flexblty of ths desgn s F s0.78, the same as for the desgn 1. Fgure. Feasble regon of Desgn 3.5 ). ( d s.5, d s 1 AIChE Journal June 001 Vol. 47, No

4 space by evaluatng the feasblty functon Ž d,. s mn z,u u fjž d, z,. Fu, jg J. Alternatvely, the same result can be obtaned by dentfyng all the actve constrant sets. The queston that arses at ths stage n many dfferent engneerng problems s how to descrbe and then quantfy the feasble range. The frst case that we are gong to address s the convex case, the extenson to the 1-D convex case wll be dscussed later. The man dea n the proposed approach s frst to determne ponts at the boundary of the feasble regon and then to evaluate the convex hull defned by those ponts. For the case of the convex feasble regon, the convex hull determned n ths way s guaranteed to be nscrbed wthn the feasble regon. However, for the nonconvex case, there s no guarantee that ths would be always the case. To overcome ths dffculty, an extenson of the basc proposed approach s presented later whereas the general nonconvex case would be the subject of future publcaton. For the convex case, the followng steps are consdered: Ž. 1 Solve the problems of determnng the maxmum devaton from the nomnal pont towards the vertces Ž k K. of the expected range of uncertan parameter Ž. varablty max k s.t. f Ž d, z,. F0, jg J j s N k Ž. Solve the problems of determnng the maxmum devaton from the nomnal pont by varyng one uncertan parameter Ž. at a tme max s.t. fjž d, z, 1,. F0, jg J s N js j N j Ž. 3 Determne the convex hull based n the ponts obtaned from steps Ž. 1 and Ž. applyng the Quckhull algorthm whch s descrbed below. Ž. 4 Determne the volume of the convex hull obtaned from step Ž. 3 usng Delaunay Trangulaton Žthe bascs of the Delanay Trangulaton are explaned n the sequel.. At the same step and usng the vertces of the expected range of uncertan parameter varablty, the volume of the overall expected regon s determned. The rato between the feasble convex hull and the expected regon volumes s then evaluated to represent the new metrc for comparng desgn s feasblty Feasble Convex Hull Rato Ž FCHR. Volume of the feasble convex hull s Ž. 7 Volume of the overall expected range Note that, at step 3, the lnear functons descrbng the faces of the convex hull are also determned. As wll be llustrated by dfferent examples, ths provdes a very accurate descrpton of the feasble space for a specfc desgn or process whch s of great mportance n many dfferent engneerng problems. An example s the determnaton of the range of valdty of a reduced knetc model as descrbed by Androulaks et al. Ž Quckhull algorthm The convex hull of a set of ponts s the smallest convex set that contans the ponts. There exst a number of dfferent algorthms n the computatonal geometry lterature that are desgned to compute the convex hull for a gven set of ponts. A revew can be found n de Berg et al. Ž Recent work on convex hull has been focused on varatons of a randomzed, ncremental algorthm where ponts are consdered one at a tme. Quckhull algorthm Ž Barber et al., proceeds usng two geometrc operatons: orented hyperplane through Ž n. d ponts and sgned dstance to hyperplane. In partcular the followng steps are followed for each processed pont: Ž. a Locate the vsble facets for the pont. A facet s vsble f the pont s above the facet. The selecton s made by evaluatng the sgned dstance from the facet; Ž b. Construct a cone of new facets from the pont to the vsble facets; Ž. c Delete the vsble facets thus formng the convex hull of the new pont and the prevous processed ponts. The correctness of the algorthm has been proved by Barber et al. Ž They also provde emprcal evdence that the algorthm runs faster when the nput set of ponts contans no extreme ponts. The detaled descrpton of the algorthm, as well as the programmng mplementaton, can be found and downloaded from the Web ste of the geometry center n Mnneapols: Delaunay trangulaton The basc defnton of the Delauany Trangulaton s the graph that has a node for every Vorono cell and a straght lne connectng two nodes f the correspondng cells share an edge. The Vorono dagram of a set of ponts Ž n. s the subdvson of the plate nto Ž n. regons one for each ste such that the regon of a specfc ste contans all the ponts n the plane for whch ths ste s the closest. The regon of ths ste s called the Vorono cell. The computaton of the Delaunay Trangulaton s based on the same teratve procedure as the convex hull estmaton. In partcular, a Delaunay Trangulaton n R d can be computed from a convex hull n R dq1,as descrbed by Barber et al. Ž The same software descrbed above has the capablty of evaluatng Delaunay Trangulaton. Example re sted The example presented n the prevous secton s used here agan to llustrate the basc steps of the proposed approach and the results obtaned. The problem nvolves convex constrants so there s no need for functon convexfcaton. The 1410 June 001 Vol. 47, No. 6 AIChE Journal

5 Fgure 3. Feasble regon of Desgn 1 ( d1s3.5, ds 0.0 ). Fgure 4. Step 1: fndng ponts at the boundary tces ). The solutons to the above optmzaton problems gve rse to the square pont shown n Fgure 4. Note that these problems correspond to the vertex enumeraton subproblems ŽSwaney and Grossmann, 1985a. proposed to evaluate the flexblty ndex. Consequently, at ths step, one can utlze the results to evaluate the flexblty ndex that corresponds to the mnmum of the results Ž., where denotes dfferent vertces. For ths example, t s found that F s0.78 lmted by the flexblty pont, as shown n Fgure 4. Step. Solve the problems of maxmzng the devatons from the nomnal pont by varyng one uncertan parameter at a tme. The optmzaton problems solved at ths step have the followng form ( ver- actve sets of the problem have been dentfed and the feasble regon s descrbed by the shaded regon shown n Fgure 3. The steps of the proposed analyss are then followed. Step 1. Solve the optmzaton problems of maxmzng the devatons from the nomnal pont towards the edges. The problems solved have the followng form max s.t. fjž d, z, 1,. F0, js1,..., 6 s N s N. and max s.t. f Ž d, z,,. F0, js1,..., 6 j 1 s N 1 1 s N. In ths way the ponts llustrated wth the crcles n Fgure 5 are determned. Step 3. Quckhull algorthm s appled to dentfy the convex hull based on the ponts obtaned from steps 1 and. The output of the algorthm s the set of lnear constrants descrbng the convex hull. The constrants are gven and llus- max s.t. fjž d, z, 1,. F0, js1,..., 6 s N s N Fgure 5. Step : Fndng ponts at the boundary ( varyng one parameter at a tme ). AIChE Journal June 001 Vol. 47, No

6 Fgure 6. Step 3: Determne the feasble convex hull. trated n Fgure 6 g s y8f0 1 g sy F0 1 g s y1.3 F0 3 1 g s y7.5 q7.0f0 4 1 g s y5.71 q4.4f0 5 1 g s y5.88 q4.63f0 6 1 Step 4. The volume Žarea n ths case, snce the problem nvolves two uncertan parameters. s evaluated at ths step usng the Quckhull algorthm that utlzes Delaunay Trangulaton to determne the volume of the convex hull. The volume of the overall expected range s also calculated. For ths example, t s found that the total volume s 16 unts whereas the feasble convex hull has a volume of 10.9 unts. Ths results n a rato of 0.68 that corresponds to the percentage of desgn feasblty. The same procedure s appled for Desgn Žd1s.5, d s3.5.. As ponted out n the prevous secton, the flexblty ndex for ths desgn has the same value as the flexblty for Desgn 1 equals to F s0.78. Ths s due to the fact that the lmtng actve set s the same n both cases. However, as shown n Fgure 7, Desgn has a smaller feasble regon than Desgn 1. Ths s correctly dentfed by the proposed approach whch determnes a rato of feasblty equal to 0.64 for Desgn compared to 0.68 for Desgn 1. It s mportant to pont out that the convex hull offers a very accurate descrpton of the feasble regon for both desgns. Remark 1. It should be ponted out that the proposed rato Ž FCHR. as defned by Eq. 7 corresponds to the percentage of feasblty based on the overall expected range of uncertanty and not the true feasblty regon. As can be observed from Fgures 6 and 7, the percentage that corresponds to the feasble convex hull s much hgher n terms of the actual feasble regon. Fgure 7. Desgn : Determnaton of feasble convex hull. Remark. It should be also noted that the proposed methodology for feasble regon evaluaton and the proposed feasblty metrc s not a probablstc measure and thus does not take nto account the probablty of uncertanty realzatons. Extensons of the proposed work along the lnes proposed by Straub and Grossmann Ž and Pstkopoulos and Mazzuch Ž s a subject of present research. Remark 3. The computatonal requrements of the proposed approach nvolve the soluton of n optmzaton problems at step 1, and n optmzaton problems at step, where n s the number of uncertan parameters nvolved. However, t should be noted that one can choose to consder a subset of these problems, dentfy the correspondng ponts at the boundary of the feasble regon and proceed n the next step of evaluatng the convex hull. Examples Example 1 The frst example consdered n ths secton also nvolves a set of convex constrants that restrct the feasble regon f s q y y40f f s q y yf0 1 1 f s y4 y30f0 3 1 The feasble regon s shown shaded n Fgure 8. We assume that the nomnal pont s Ž,. sž.5, 0. 1 wth expected devatons of y 1 s7.5, q 1 s.5, q s0 and y s60. The steps of the proposed analyss are then appled to determne frst the ponts at the boundary of the feasble regon and then the feasble convex hull. At step of the procedure, the flexblty ndex s also evaluated to be equal to F s The resultng feasble convex hull has a total volume of unts, whle the expected regon corresponds to a volume of 141 June 001 Vol. 47, No. 6 AIChE Journal

7 followng m 1.84 y f sp q Hy yk m D yp y F0.0 c m 1.84 y f syp y Hq qk m D qp y F0.0 c ˆ f s m Hy W F0.0 3 f sc yc max F0.0 4 f syc q r c max F Fgure 8. Example 1: feasble convex hull. 800 unts, that results n a feasble rato of FCHRs0.19. Note that the nscrbed convex hull descrbes the feasble regon much more accurately than the flexblty ndex that corresponds to only 3% of the expected range of varablty compared to 19% of the feasble convex hull, as shown n Fgure 8. Note that the percentage n terms of the actual feasble regon s much hgher snce only the lght shaded areas ŽFg- ure 8. are excluded from the feasble convex hull. The value of FCHR s, however, stll small because t reflects the rato of the feasble convex hull regon when compared not to the actual feasble regon that s unknown, but the expected range of uncertanty whch can be much larger. The exact lnear constrants descrbng the convex hull are g s q9.98 q4.94s g s y3. q1.31s0.0 1 ˆ The desgn varables are the drver power W, the pump head H, the ppe dameter D, and the control valve sze c max, the control varable s the valve coeffcent c, and the uncer- tan parameters are the desred pressure s P, and the 1 flow rate s m. The expected devatons consdered n ths example are q s, y s6, q s00, y s Two desgns are examned and evaluated n terms of ther feasblty Desgn 1 that corresponds to the followng set of desgn varables ˆ max W s31., H s1.3, Ds0.076, c s and Desgn that corresponds to the followng desgn values, ˆ max W s33.74, H s1.406, Ds0.0764, c s The proposed approach s appled to both desgns n order to evaluate the feasble convex hull. Fgures 9 and 10 show the results. It s found that the feasblty rato s FCHRs0.97 for desgn 1 and 0.9 for desgn compared to flexblty ndex that g s y4.0 y30.0s g s y.75 y9.38s g s q0.1 y104.74s g s q5.3 y49.3 s g s y8.4 q15.4 s g s y5.9 q8.06s Example The second example corresponds to the pump and ppe example descrbed by Swaney and Grossmann Ž 1985a.. The set of constrants represent the energy balance, the outlet pressure tolerance, the pump drver power lmt, and the constrant valve range. By elmnatng the outlet pressure P usng the energy balance, the reduced set of nequaltes s the Fgure 9. Example : feasble convex hull for desgn 1. AIChE Journal June 001 Vol. 47, No

8 Fgure 10. Example : feasble convex hull for desgn. s for desgn 1 and for desgn. Consequently, by evaluatng the feasble convex hull, we can better compare the feasblty of dfferent desgns snce, as shown n Fgures 9 and 10, desgn 1 has actually larger feasble regon than desgn whch could not be captured by the flexblty ndex. The major advantage of the proposed approach compared to the flexblty ndex s that t s not lmted by one drecton, but can capture the overall desgn behavor under parameter varablty. However, t should be ponted out that snce the feasble regon s 1-DQC and not jontly convex on 1,, there s no guarantee that the convex hull defned by the ponts found through steps 1 and would be nscrbed wthn the feasble regon. To overcome ths dffculty the approach descrbed n the followng secton s modfed as suggested. Example 3 The example consdered here s a modfcaton of the example of Pstkopoulos and Ierapetrtou Ž wth three uncertan parameters. The constrants descrbed the feasble regon of the desgn Ž d, d. have the followng form 1 f sy zy q0.5 q.0 q d y3 d y8f0 1 1 f sy zy r3y y r3q d q8r3f0 f s zq y y d q y4f where z s the control varable, 1,, 3 are the uncertan parameters wth nomnal value of 1 N s N s 3 N s and expected devatons 1 s 1 s 1 s. All the constrants are jontly convex on Ž. and Ž z.. The desgn examned here corresponds to Ž d, d. sž 3,1. 1. The proposed approach s appled for ths desgn and the followng ponts are frst dentfed at the boundary of the feasble regon. Step 1. At the frst step, the ponts towards the vertces of Fgure 11. Example 3: feasble convex hull for 3-D feasble regon. the expected uncertanty range: p1: Ž,,. sž.18,.18,.18. p: Ž,,. sž 3.000, 3.000, p3: Ž,,. sž.5, 1.748,.5. p4: Ž,,. sž 1.637,.363,.363. p5: Ž,,. sž.486, 1.514, p6: Ž,,. sž 1.514, 1.514,.455. p7: Ž,,. sž 0.000, 4.000, p8: Ž,,. sž 0.000, 0.000, Step. Varyng only one uncertan parameter at the tme we got the followng ponts: p9: Ž,,. sž.6,.000,.000. p10: Ž,,. sž 0.000,.000,.000. p11: Ž,,. sž.000,.000,.31. p1: Ž,,. sž.000,.000, p13: Ž,,. sž.000, 3.646,.000. p14: Ž,,. sž.000, 0.667, Step 3. Then the quckhull algorthm s appled to determne the feasble convex hull of the ponts obtaned at steps 1 and. The resultng convex hull s llustrated n Fgure 11. Step 4. Usng Delaunay Trangulaton the volume of the convex hull s determned to be equal to 1.6 that corresponds to 19% of the total expected range of uncertanty. To compare wth the results of the flexblty analyss the flexblty of the desgn s also determned F s The volume of the correspondng rectangular s approxmately that corresponds to only 0.1% of the expected uncertanty range Ž see Fgure June 001 Vol. 47, No. 6 AIChE Journal

9 lem. Snce then constrants consdered here are jontly quasconvex n z and 1-D quas-convex n Ž., t s proven that the lnearzaton would always result n underestmaton of the feasble regon Ž see Appendx A.. In the followng secton an example nvolvng 1-D quas-convex functons s consdered to llustrate the results of the proposed approach for ths case. Example 4 The example presented n ths secton corresponds to an example that exhbts a 1-D convex feasble regon. To llustrate the proposed deas, the example s smple and descrbed by the followng constrants f sy q q F f s Ž 6q. y350f0 1 f s q y60f f s y10 y50f Fgure 1. Example 3: Flexblty range and overall expected range of uncertanty. 1DQC Convex Feasble Regon Frst, t should be ponted out that only 1-D convex regons are examned here and not general nonconvex regons whch s the subject of the second part of ths work. For the case of the 1-D convex regon, the proposed approach s smlar to the approach for the convex case wth the dfference that after evaluatng the ponts at the boundary towards the vertces as descrbed at step 1, the nonconvex constrants are lnearzed at these ponts. It should be noted that only the nonconvex constrants that are actve at these ponts are lnearzed and not all of the nonlnear constrants n the prob- Note that the constrants satsfy propertes Ž. and Ž. 3 proved by Swaney and Grossmann Ž 1985a. to be suffcent n resultng n the 1DQC feasble regon Ž as also shown n Fgure 13.. The proposed approach s appled for ths example and the feasble convex hull s determned Ž Fgure 13.. Note that, for comparson purposes, the approach s appled for both the orgnal nonconvex constrants, as well as the convexfed ones. As suggested n the prevous secton, the convexfcaton s obtaned after the determnaton of the ponts at the boundary. For ths example, the pont used to convexfy the nonconvex constrant f s the pont Ž P1. shown n Fgure 14 dentfed from the optmzaton problem of maxmzng the devaton from the nomnal pont towards the vertex that corresponds to the maxmum postve devatons of 1,. The convexfcaton results n the followng lnear constrant shown n Fgure 14 wth a dashed lne ˆf s y51.385q4.17 F0.0 1 Fgure 13. Example 4: feasblty convex hull for 1DQC feasble regon. Fgure 14. Example 4: convexfcaton of constrant f. AIChE Journal June 001 Vol. 47, No

10 plotng the nherent dstrbuted nature of the ntal steps of the proposed approach, and Ž. c the extenson of the proposed work to ncorporate the probablty of occurrence of uncertan parameter realzatons. Acknowledgments The author gratefully acknowledges fnancal support from the Natonal Scence Foundaton under the NSF CAREER program CTS Specal thanks to the anonymous revewer that brought to our attenton smlar work done n the area of electrcal engneerng. Fgure 15. Example 4: feasble convex hull for 1DQC feasble regon usng convexfed constrant. Usng ths constrant nstead of constrant f results n the determnaton of pont Ž Pn. to be used on feasble convex hull evaluaton nstead of pont P Ž Fgure 14.. Snce the dfference s very small n ths example, we llustrate the result by focusng Fgure 15 at the area of pont Ž P. whch s the only dfference between the feasble convex hull of the orgnal problem and the feasble convex hull of the convexfed problem. Dscusson and Future Drectons A novel approach for evaluatng the range of feasblty of a gven process s proposed n ths artcle. The basc dea s to determne the convex hull that can be nscrbed wthn the feasble regon. The convex hulls were computed usng the Quckhull Algorthm whch s an ncremental algorthm for evaluatng the convex hull gven a set of ponts. The ponts used to determne the feasble convex hull are ponts at the boundary of the feasble regon that are obtaned from the soluton of a seres of small optmzaton subproblems. The outcome of the applcaton of the Quckhull Algorthm s not only the computaton of the convex hull descrbed by a set of lnear constrants but also ts volume. Ths s acheved based on the deas of Delaunay Trangulaton. A new metrc was ntroduced that corresponds to the rato of the feasble convex hull volume dvded by the overall expected range of uncertanty. In all examples consdered we found that the proposed approach results n much better descrpton of the feasble space. Moreover, the new metrc ntroduced to represent the feasblty of a gven process was found to descrbe accurately the feasble space and thus results n correct comparson between dfferent desgns andror processes. Although the examples addressed n ths work are of small scale, the results obtaned are very promsng regardng the applcablty of the proposed work n large-scale problems. Dfferent extensons are currently under nvestgaton regardng Ž. a the applcablty of the work to nonconvex case, and Ž b. the development of effcent mplementaton scheme ex- Lterature Cted Androulaks, I. P., A. Srdeshapande, and M. G. Ierapetrtou, Determnaton of the Range of Valdty of Reduced Knetc Mechansms, n press AIChE J., Ž Barber, C. B., D. P. Dobkn, and H. Huhdanpaa, The Quckhull Algorthm for Convex Hulls, ACM Trans. Math. Soft.,, 469 Ž de Berg, M., M. van Kreveld, M. Overmars, and O. Schwarzkopf, Computatonal Geometry. Algorthms and Applcatons, Spreger Ž Floudas, C. A., Z. Gumus, and M. G. Ierapetrtou, Global Optmzaton for Feasblty Test and Flexblty Index Problems, Ind. Eng. Chem. Res., Ž Grossmann, I. E., and K. P. Halemane, Decomposton Strategy for Desgnng Flexble Chemcal Plants, AIChE J., 8, 686 Ž Grossmann, I. E., and C. A. Floudas, Actve Constrant Strategy for Flexblty Analyss n Chemcal Processes, Comput. Chem. Eng., 11, 675 Ž Halemane, K. P., and I. E. Grossmann, Optmal Process Desgn under Uncertanty, AIChE J., 9, 45 Ž Kubc, W. L., and F. P. Sten, A Theory of Desgn Relablty Usng Probablty and Fuzzy Sets, AIChE J., 34, 583 Ž Mgdalas, A., P. M. Pardalos, and P. Varbrand, Multle el Optmzaton: Algorthms and Applcatons, Kluwer Academc Publshers, Dordrecht, The Netherlands Ž Ostrovsky, G. M., Y. M. Voln, E. I. Bart, and M. M. Senyavn, Flexblty Analyss and Optmzaton of Chemcal Plants wth Uncertan Parameters, Comput. Chem. Eng., 18, 755 Ž Pstkopoulos, E. N., and M. G. Ierapetrtou, A Novel Approach for Optmal Process Desgn Under Uncertanty, Comput. Chem. Eng., 19, 1089 Ž Pstkopoulos, E. N., and T. A. Mazzuch, A Novel Flexblty Analyss Approach for Processes wth Stochastc Parameters, Comput. Chem. Eng., 14, 991 Ž Saboo, A. K., M. Morar, and D. C. Woodcock, Desgn of Reslent Processng Plants: VIII. A Reslence Index for Heat Exchanger Networks, Chem. Eng. Sc., 40, 1553 Ž Straub, D. A., and I. E. Grossmann, Desgn Optmzaton of Stochastc Flexblty, Comput. Chem. Eng., 17, 339 Ž Swaney, R. E., and I. E. Grossmann, An Index for Operatonal Flexblty n Chemcal Process Desgn Part I: Formulaton and Theory, AIChE J., 36, 139 Ž 1985a.. Swaney, R. E., and I. E. Grossmann, An Index for Operatonal Flexblty n Chemcal Process Desgn Part II: Computatonal Algorthms, AIChE J., 31, 631 Ž 1985b.. Appendx: Proof of Feasble Regon Underestmaton Proposton If f Ž d, z,. j are jontly quas-convex n z and 1-D quas- convex n, then the lnearzaton around and extreme pont underestmates the feasble regon. Proof As llustrated n Fgure A1, we need to bascally to prove that the pont 1 belongs n the feasble regon, that s, 1416 June 001 Vol. 47, No. 6 AIChE Journal

11 Also, we assume that the nomnal pont s a feasble pont Ž N. 1 and thus d, F0. Snce belongs to the lne wth slope Ž.Ž. 1 N fj d, z, there exst a scalar such that s q Ž. 1y. Based on theorem proved by Swaney and Grossmann Ž 1985a., f the constrant functons f Ž d, z,. j are jontly quas-convex n z and one-dmensonal quas-convex n, then the feasblty functon Ž d,. s 1-D quas-convex n, Ž d, 1. Fmax Ž d,., Ž d, N. 4. Fgure A1. Underestmaton of feasble regon. Ž. Ž N.4 Ž 1. Snce max d,, d, s 0, d, F 0 and the proposton has been proved. Ž 1. d, F0. Snce and are at the boundary Žd,. s Žd,. s0 as proved by Swaney and Grossmann Ž 1985a.. Manuscrpt rece ed June 7, 000, and re son rece ed No. 6,000. AIChE Journal June 001 Vol. 47, No

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan

More information

A Five-Point Subdivision Scheme with Two Parameters and a Four-Point Shape-Preserving Scheme

A Five-Point Subdivision Scheme with Two Parameters and a Four-Point Shape-Preserving Scheme Mathematcal and Computatonal Applcatons Artcle A Fve-Pont Subdvson Scheme wth Two Parameters and a Four-Pont Shape-Preservng Scheme Jeqng Tan,2, Bo Wang, * and Jun Sh School of Mathematcs, Hefe Unversty

More information

GSLM Operations Research II Fall 13/14

GSLM Operations Research II Fall 13/14 GSLM 58 Operatons Research II Fall /4 6. Separable Programmng Consder a general NLP mn f(x) s.t. g j (x) b j j =. m. Defnton 6.. The NLP s a separable program f ts objectve functon and all constrants are

More information

Review of approximation techniques

Review of approximation techniques CHAPTER 2 Revew of appromaton technques 2. Introducton Optmzaton problems n engneerng desgn are characterzed by the followng assocated features: the objectve functon and constrants are mplct functons evaluated

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

UNIT 2 : INEQUALITIES AND CONVEX SETS

UNIT 2 : INEQUALITIES AND CONVEX SETS UNT 2 : NEQUALTES AND CONVEX SETS ' Structure 2. ntroducton Objectves, nequaltes and ther Graphs Convex Sets and ther Geometry Noton of Convex Sets Extreme Ponts of Convex Set Hyper Planes and Half Spaces

More information

Intra-Parametric Analysis of a Fuzzy MOLP

Intra-Parametric Analysis of a Fuzzy MOLP Intra-Parametrc Analyss of a Fuzzy MOLP a MIAO-LING WANG a Department of Industral Engneerng and Management a Mnghsn Insttute of Technology and Hsnchu Tawan, ROC b HSIAO-FAN WANG b Insttute of Industral

More information

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

More information

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

More information

Mathematics 256 a course in differential equations for engineering students

Mathematics 256 a course in differential equations for engineering students Mathematcs 56 a course n dfferental equatons for engneerng students Chapter 5. More effcent methods of numercal soluton Euler s method s qute neffcent. Because the error s essentally proportonal to the

More information

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

More information

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data

Type-2 Fuzzy Non-uniform Rational B-spline Model with Type-2 Fuzzy Data Malaysan Journal of Mathematcal Scences 11(S) Aprl : 35 46 (2017) Specal Issue: The 2nd Internatonal Conference and Workshop on Mathematcal Analyss (ICWOMA 2016) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES

More information

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique //00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy

More information

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation Intellgent Informaton Management, 013, 5, 191-195 Publshed Onlne November 013 (http://www.scrp.org/journal/m) http://dx.do.org/10.36/m.013.5601 Qualty Improvement Algorthm for Tetrahedral Mesh Based on

More information

Chapter 6 Programmng the fnte element method Inow turn to the man subject of ths book: The mplementaton of the fnte element algorthm n computer programs. In order to make my dscusson as straghtforward

More information

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes SPH3UW Unt 7.3 Sphercal Concave Mrrors Page 1 of 1 Notes Physcs Tool box Concave Mrror If the reflectng surface takes place on the nner surface of the sphercal shape so that the centre of the mrror bulges

More information

Hermite Splines in Lie Groups as Products of Geodesics

Hermite Splines in Lie Groups as Products of Geodesics Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES

VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES UbCC 2011, Volume 6, 5002981-x manuscrpts OPEN ACCES UbCC Journal ISSN 1992-8424 www.ubcc.org VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints TPL-ware Dsplacement-drven Detaled Placement Refnement wth Colorng Constrants Tao Ln Iowa State Unversty tln@astate.edu Chrs Chu Iowa State Unversty cnchu@astate.edu BSTRCT To mnmze the effect of process

More information

A Facet Generation Procedure. for solving 0/1 integer programs

A Facet Generation Procedure. for solving 0/1 integer programs A Facet Generaton Procedure for solvng 0/ nteger programs by Gyana R. Parja IBM Corporaton, Poughkeepse, NY 260 Radu Gaddov Emery Worldwde Arlnes, Vandala, Oho 45377 and Wlbert E. Wlhelm Teas A&M Unversty,

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned

More information

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

Feedback Min-Max Model Predictive Control Based on a Quadratic Cost Function

Feedback Min-Max Model Predictive Control Based on a Quadratic Cost Function Proceedngs of the 26 Amercan Control Conference Mnneapols, Mnnesota, USA, June 14-16, 26 WeC5.5 Feedback Mn-Max Model Predctve Control Based on a Quadratc Cost Functon D. Muñoz de la Peña,T.Alamo, A. Bemporad

More information

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT 3. - 5. 5., Brno, Czech Republc, EU APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT Abstract Josef TOŠENOVSKÝ ) Lenka MONSPORTOVÁ ) Flp TOŠENOVSKÝ

More information

Design of Structure Optimization with APDL

Design of Structure Optimization with APDL Desgn of Structure Optmzaton wth APDL Yanyun School of Cvl Engneerng and Archtecture, East Chna Jaotong Unversty Nanchang 330013 Chna Abstract In ths paper, the desgn process of structure optmzaton wth

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

More information

Ecient Computation of the Most Probable Motion from Fuzzy. Moshe Ben-Ezra Shmuel Peleg Michael Werman. The Hebrew University of Jerusalem

Ecient Computation of the Most Probable Motion from Fuzzy. Moshe Ben-Ezra Shmuel Peleg Michael Werman. The Hebrew University of Jerusalem Ecent Computaton of the Most Probable Moton from Fuzzy Correspondences Moshe Ben-Ezra Shmuel Peleg Mchael Werman Insttute of Computer Scence The Hebrew Unversty of Jerusalem 91904 Jerusalem, Israel Emal:

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

Lecture 5: Multilayer Perceptrons

Lecture 5: Multilayer Perceptrons Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented

More information

Face Recognition University at Buffalo CSE666 Lecture Slides Resources:

Face Recognition University at Buffalo CSE666 Lecture Slides Resources: Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural

More information

An Entropy-Based Approach to Integrated Information Needs Assessment

An Entropy-Based Approach to Integrated Information Needs Assessment Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS Academc Research Internatonal ISS-L: 3-9553, ISS: 3-9944 Vol., o. 3, May 0 EVALUATIO OF THE PERFORMACES OF ARTIFICIAL BEE COLOY AD IVASIVE WEED OPTIMIZATIO ALGORITHMS O THE MODIFIED BECHMARK FUCTIOS Dlay

More information

Lecture 4: Principal components

Lecture 4: Principal components /3/6 Lecture 4: Prncpal components 3..6 Multvarate lnear regresson MLR s optmal for the estmaton data...but poor for handlng collnear data Covarance matrx s not nvertble (large condton number) Robustness

More information

Conditional Speculative Decimal Addition*

Conditional Speculative Decimal Addition* Condtonal Speculatve Decmal Addton Alvaro Vazquez and Elsardo Antelo Dep. of Electronc and Computer Engneerng Unv. of Santago de Compostela, Span Ths work was supported n part by Xunta de Galca under grant

More information

y and the total sum of

y and the total sum of Lnear regresson Testng for non-lnearty In analytcal chemstry, lnear regresson s commonly used n the constructon of calbraton functons requred for analytcal technques such as gas chromatography, atomc absorpton

More information

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following.

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following. Complex Numbers The last topc n ths secton s not really related to most of what we ve done n ths chapter, although t s somewhat related to the radcals secton as we wll see. We also won t need the materal

More information

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements Module 3: Element Propertes Lecture : Lagrange and Serendpty Elements 5 In last lecture note, the nterpolaton functons are derved on the bass of assumed polynomal from Pascal s trangle for the fled varable.

More information

An Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method

An Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method Internatonal Journal of Computatonal and Appled Mathematcs. ISSN 89-4966 Volume, Number (07), pp. 33-4 Research Inda Publcatons http://www.rpublcaton.com An Accurate Evaluaton of Integrals n Convex and

More information

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers 62626262621 Journal of Uncertan Systems Vol.5, No.1, pp.62-71, 211 Onlne at: www.us.org.u A Smple and Effcent Goal Programmng Model for Computng of Fuzzy Lnear Regresson Parameters wth Consderng Outlers

More information

5 The Primal-Dual Method

5 The Primal-Dual Method 5 The Prmal-Dual Method Orgnally desgned as a method for solvng lnear programs, where t reduces weghted optmzaton problems to smpler combnatoral ones, the prmal-dual method (PDM) has receved much attenton

More information

Repeater Insertion for Two-Terminal Nets in Three-Dimensional Integrated Circuits

Repeater Insertion for Two-Terminal Nets in Three-Dimensional Integrated Circuits Repeater Inserton for Two-Termnal Nets n Three-Dmensonal Integrated Crcuts Hu Xu, Vasls F. Pavlds, and Govann De Mchel LSI - EPFL, CH-5, Swtzerland, {hu.xu,vasleos.pavlds,govann.demchel}@epfl.ch Abstract.

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

Meta-heuristics for Multidimensional Knapsack Problems

Meta-heuristics for Multidimensional Knapsack Problems 2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information

Positive Semi-definite Programming Localization in Wireless Sensor Networks

Positive Semi-definite Programming Localization in Wireless Sensor Networks Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer

More information

USING GRAPHING SKILLS

USING GRAPHING SKILLS Name: BOLOGY: Date: _ Class: USNG GRAPHNG SKLLS NTRODUCTON: Recorded data can be plotted on a graph. A graph s a pctoral representaton of nformaton recorded n a data table. t s used to show a relatonshp

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

A Robust Method for Estimating the Fundamental Matrix

A Robust Method for Estimating the Fundamental Matrix Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.

More information

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Overvew 2 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Introducton Mult- Smulator MASIM Theoretcal Work and Smulaton Results Concluson Jay Wagenpfel, Adran Trachte Motvaton and Tasks Basc Setup

More information

Solitary and Traveling Wave Solutions to a Model. of Long Range Diffusion Involving Flux with. Stability Analysis

Solitary and Traveling Wave Solutions to a Model. of Long Range Diffusion Involving Flux with. Stability Analysis Internatonal Mathematcal Forum, Vol. 6,, no. 7, 8 Soltary and Travelng Wave Solutons to a Model of Long Range ffuson Involvng Flux wth Stablty Analyss Manar A. Al-Qudah Math epartment, Rabgh Faculty of

More information

User Authentication Based On Behavioral Mouse Dynamics Biometrics

User Authentication Based On Behavioral Mouse Dynamics Biometrics User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

LECTURE NOTES Duality Theory, Sensitivity Analysis, and Parametric Programming

LECTURE NOTES Duality Theory, Sensitivity Analysis, and Parametric Programming CEE 60 Davd Rosenberg p. LECTURE NOTES Dualty Theory, Senstvty Analyss, and Parametrc Programmng Learnng Objectves. Revew the prmal LP model formulaton 2. Formulate the Dual Problem of an LP problem (TUES)

More information

An efficient iterative source routing algorithm

An efficient iterative source routing algorithm An effcent teratve source routng algorthm Gang Cheng Ye Tan Nrwan Ansar Advanced Networng Lab Department of Electrcal Computer Engneerng New Jersey Insttute of Technology Newar NJ 7 {gc yt Ansar}@ntedu

More information

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

More information

Circuit Analysis I (ENGR 2405) Chapter 3 Method of Analysis Nodal(KCL) and Mesh(KVL)

Circuit Analysis I (ENGR 2405) Chapter 3 Method of Analysis Nodal(KCL) and Mesh(KVL) Crcut Analyss I (ENG 405) Chapter Method of Analyss Nodal(KCL) and Mesh(KVL) Nodal Analyss If nstead of focusng on the oltages of the crcut elements, one looks at the oltages at the nodes of the crcut,

More information

Optimization Methods: Integer Programming Integer Linear Programming 1. Module 7 Lecture Notes 1. Integer Linear Programming

Optimization Methods: Integer Programming Integer Linear Programming 1. Module 7 Lecture Notes 1. Integer Linear Programming Optzaton Methods: Integer Prograng Integer Lnear Prograng Module Lecture Notes Integer Lnear Prograng Introducton In all the prevous lectures n lnear prograng dscussed so far, the desgn varables consdered

More information

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc.

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc. [Type text] [Type text] [Type text] ISSN : 0974-74 Volume 0 Issue BoTechnology 04 An Indan Journal FULL PAPER BTAIJ 0() 04 [684-689] Revew on Chna s sports ndustry fnancng market based on market -orented

More information

Wavefront Reconstructor

Wavefront Reconstructor A Dstrbuted Smplex B-Splne Based Wavefront Reconstructor Coen de Vsser and Mchel Verhaegen 14-12-201212 2012 Delft Unversty of Technology Contents Introducton Wavefront reconstructon usng Smplex B-Splnes

More information

Load-Balanced Anycast Routing

Load-Balanced Anycast Routing Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance

More information

Parallel matrix-vector multiplication

Parallel matrix-vector multiplication Appendx A Parallel matrx-vector multplcaton The reduced transton matrx of the three-dmensonal cage model for gel electrophoress, descrbed n secton 3.2, becomes excessvely large for polymer lengths more

More information

LS-TaSC Version 2.1. Willem Roux Livermore Software Technology Corporation, Livermore, CA, USA. Abstract

LS-TaSC Version 2.1. Willem Roux Livermore Software Technology Corporation, Livermore, CA, USA. Abstract 12 th Internatonal LS-DYNA Users Conference Optmzaton(1) LS-TaSC Verson 2.1 Wllem Roux Lvermore Software Technology Corporaton, Lvermore, CA, USA Abstract Ths paper gves an overvew of LS-TaSC verson 2.1,

More information

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1) Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A

More information

Active Contours/Snakes

Active Contours/Snakes Actve Contours/Snakes Erkut Erdem Acknowledgement: The sldes are adapted from the sldes prepared by K. Grauman of Unversty of Texas at Austn Fttng: Edges vs. boundares Edges useful sgnal to ndcate occludng

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

An Application of Network Simplex Method for Minimum Cost Flow Problems

An Application of Network Simplex Method for Minimum Cost Flow Problems BALKANJM 0 (0) -0 Contents lsts avalable at BALKANJM BALKAN JOURNAL OF MATHEMATICS journal homepage: www.balkanjm.com An Applcaton of Network Smplex Method for Mnmum Cost Flow Problems Ergun EROGLU *a

More information

F Geometric Mean Graphs

F Geometric Mean Graphs Avalable at http://pvamu.edu/aam Appl. Appl. Math. ISSN: 1932-9466 Vol. 10, Issue 2 (December 2015), pp. 937-952 Applcatons and Appled Mathematcs: An Internatonal Journal (AAM) F Geometrc Mean Graphs A.

More information

SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING USING A CALCULATOR

SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING USING A CALCULATOR SENSITIVITY ANALYSIS IN LINEAR PROGRAMMING USING A CALCULATOR Judth Aronow Rchard Jarvnen Independent Consultant Dept of Math/Stat 559 Frost Wnona State Unversty Beaumont, TX 7776 Wnona, MN 55987 aronowju@hal.lamar.edu

More information

3D vector computer graphics

3D vector computer graphics 3D vector computer graphcs Paolo Varagnolo: freelance engneer Padova Aprl 2016 Prvate Practce ----------------------------------- 1. Introducton Vector 3D model representaton n computer graphcs requres

More information

Backpropagation: In Search of Performance Parameters

Backpropagation: In Search of Performance Parameters Bacpropagaton: In Search of Performance Parameters ANIL KUMAR ENUMULAPALLY, LINGGUO BU, and KHOSROW KAIKHAH, Ph.D. Computer Scence Department Texas State Unversty-San Marcos San Marcos, TX-78666 USA ae049@txstate.edu,

More information

Polyhedral Compilation Foundations

Polyhedral Compilation Foundations Polyhedral Complaton Foundatons Lous-Noël Pouchet pouchet@cse.oho-state.edu Dept. of Computer Scence and Engneerng, the Oho State Unversty Feb 8, 200 888., Class # Introducton: Polyhedral Complaton Foundatons

More information

Optimal Workload-based Weighted Wavelet Synopses

Optimal Workload-based Weighted Wavelet Synopses Optmal Workload-based Weghted Wavelet Synopses Yoss Matas School of Computer Scence Tel Avv Unversty Tel Avv 69978, Israel matas@tau.ac.l Danel Urel School of Computer Scence Tel Avv Unversty Tel Avv 69978,

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Handout 5 Luca Trevisan September 7, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Handout 5 Luca Trevisan September 7, 2017 U.C. Bereley CS294: Beyond Worst-Case Analyss Handout 5 Luca Trevsan September 7, 207 Scrbed by Haars Khan Last modfed 0/3/207 Lecture 5 In whch we study the SDP relaxaton of Max Cut n random graphs. Quc

More information

CHAPTER 2 DECOMPOSITION OF GRAPHS

CHAPTER 2 DECOMPOSITION OF GRAPHS CHAPTER DECOMPOSITION OF GRAPHS. INTRODUCTION A graph H s called a Supersubdvson of a graph G f H s obtaned from G by replacng every edge uv of G by a bpartte graph,m (m may vary for each edge by dentfyng

More information

Interpolation of the Irregular Curve Network of Ship Hull Form Using Subdivision Surfaces

Interpolation of the Irregular Curve Network of Ship Hull Form Using Subdivision Surfaces 7 Interpolaton of the Irregular Curve Network of Shp Hull Form Usng Subdvson Surfaces Kyu-Yeul Lee, Doo-Yeoun Cho and Tae-Wan Km Seoul Natonal Unversty, kylee@snu.ac.kr,whendus@snu.ac.kr,taewan}@snu.ac.kr

More information

Problem Set 3 Solutions

Problem Set 3 Solutions Introducton to Algorthms October 4, 2002 Massachusetts Insttute of Technology 6046J/18410J Professors Erk Demane and Shaf Goldwasser Handout 14 Problem Set 3 Solutons (Exercses were not to be turned n,

More information

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming CS 4/560 Desgn and Analyss of Algorthms Kent State Unversty Dept. of Math & Computer Scence LECT-6 Dynamc Programmng 2 Dynamc Programmng Dynamc Programmng, lke the dvde-and-conquer method, solves problems

More information

Learning to Project in Multi-Objective Binary Linear Programming

Learning to Project in Multi-Objective Binary Linear Programming Learnng to Project n Mult-Objectve Bnary Lnear Programmng Alvaro Serra-Altamranda Department of Industral and Management System Engneerng, Unversty of South Florda, Tampa, FL, 33620 USA, amserra@mal.usf.edu,

More information

CMPS 10 Introduction to Computer Science Lecture Notes

CMPS 10 Introduction to Computer Science Lecture Notes CPS 0 Introducton to Computer Scence Lecture Notes Chapter : Algorthm Desgn How should we present algorthms? Natural languages lke Englsh, Spansh, or French whch are rch n nterpretaton and meanng are not

More information