Model Based Design and System Simulation with SimulationX and Tool Integration with optislang

Transcription

1 Model Based Design and Sstem Simulation with SimulationX and Tool Integration with optislang Dr. Andreas Uhlig Uwe Grätz ITI GmbH

2 Introduction 2 Who we are Where we areout ITI Multi-faceted high-technolog Leader in virtual sstem engineering Headquarters located in Dresden downtown Frankfurt Berlin Dresden Munich

3 Introduction 3 Core Business Offering complete solutions for sstem modeling, simulation, analsis and testing development of simulation software engineering services product distribution product integration

4 Content 4 Content Model Based Design Equation Based Modeling Design of a Membrane Clinder an Eample for Sensitivit Analsis Interface to optislang Summar and Outlook

5 Model Based Design 5 Sstem Simulation / Model Based Design New challenges in sstems engineering Time, cost, Qualit, safet, Energ efficienc Dependenc of subsstems Earl assessment of designs Virtual prototping - of sstem! Use models and simulations Re-use models

6 Model Based Design 6 Requirements for Model Based Sstem Design Multidomain modeling Hierarchical modeling Replaceable models One model for multiple analses

7 Model Based Design Requirements Multidomain Modeling 7 Multidomain Modeling (SimulationX Libraries) Signal Blocks Mechanics Powertrain Electro- Mechanics Magnetics Pneumatics Hdraulics Thermo

8 Model Based Design Requirements Multidomain Modeling 8 Domain Specific Workspaces for the User Multi-bod Sstems Network Elements Statecharts Signal Blocks, Controls

9 Model Based Design Requirements Hierarchical Modeling 9 Level 1 Basic Model Simple rigid powertrain model including main components Analsis of feasibilit and general phsical relations Load Engine Flwheel Gearbo Gearbo Output Shaft Wheel Tire-Road Contact Car Driving Resistance 1D 2D 3D

10 Model Based Design Requirements Hierarchical Modeling 10 Level 2 Detailed Model Detailing of interesting elements Regard of elastic forces Simple engine model Merge elements in compounds Case studies Proof of concept simulations Engine Flwheel Gearbo Output Shaft Driving Resistance Gearbo Car 1D 2D 3D DoF: < 10 Parameter: Frequenc: < 100 Hz

11 Model Based Design Requirements Hierarchical Modeling 11 Level 3 Comple Model Multi-domain compounds Detailed engine & gearbo model StateCharts Flwheel Gearbo Output Shaft Driving Resistance Engine Car Turbocharger Gearbo 1D 2D 3D Clinders, Crank, PTOs DoF: > 10 Parameter: > 50 Frequenc: > 100 Hz

12 Model Based Design Requirements 12 Replaceable Models Easil switch to compatible tpe Efficient modeling

13 Model Based Design Requirements 13 One model for multiple analses Transient in time domain Static equilibrium (DC analses) Stationar simulation (non-linear, frequenc domain) Linear sstem analsis of the entire sstem: Eigenfrequencies, Eigenmodes Energ analses Frequenc response Poles/Zeros

14 Equation Based Modeling 14 Equation Based Modeling Phsical laws are described in tet books in terms of formulas like F = m a Equation In programs this is implemented dependent on the goal of the simulation F := m a or a := F / m or m := F / a Assigments

15 Equation Based Modeling 15 Modeling Concept Lumped (Network) Elements Definition of potential and flow quantities for each phsical domain Models consist of Elements and Connections Connections calculate potential quantities and define conservation equations (e.g. ΣF = 0 for mechanical nodes) F F F = 0 element F F,, &... F F element Elements define relations between flow- und potential variables within elements (e.g. F = k * in a mechanical spring model) Node element

16 Equation Based Modeling 16 Analogies in Phsical Domains SOURCES Potential Quantit Flow Quantit BASIC ELEMENT TYPES Capacitive Element Inductive Element Dissipative Element Electronics Voltage V Current I Capacitor I = C U & Inductor = 1 I U dt L Resistor U I = R Mechanics translational Velocit v Force F Mass F = m v & Spring F = c v dt Damper F = d v Mechanics rotational Angular Velocit ω Torque T Inertia T = J ω ω & Rotational Spring T = c ω dt Rotational Damper T = d ω ω Hdraulics Pressure p Volume Flow Q Volume Q = V p & Line (without loss) = 1 Q L pdt H Throttle p Q = R H Thermics Temperature T Heat Flow P Heat Capacit P = C T & - Heat Resistance R P = th T

17 Equation Based Modeling Modelica Language Allows Acausal and causal modeling Distinguishes Through and Across variables Multi-domain modeling Variables carr Attributes like units, documentation, min value, ma value and nominal value Modeling based on a modeling language Standard Modelica is a registered trademark of Modelica Association

18 Equation Based Modeling Who is Modelica? Standardized b Modelica Association Formed in September 1996 An independent, international, nonprofit modeling standards organization A clearing house for public domain Modelica libraries and documentation Organizer of Modelica design meetings and conferences SimulationX is based on Modelica Modelica is a registered trademark of Modelica Association

19 Equation Based Modeling 19 Mathematical Models represented b Modelica Differential Algebraic Equation (DAE) 0 = f (, &, p, d, r, t) DAE (semi-eplizit) & = f (,, p, d, r, t) 0 = f (,, p, d, r, t) continuous states algebraic variables p parameters d discrete variable R root functions t time

20 Equation Based Modeling 20 Start Equation sstem preparation Iterate discrete states Calc. consistent initial values Global smbolic analsis Event iteration Time step Time iteration Message No Success? Yes Prepare new Time step Message Yes Finished? No No Yes Event? STOP

21 Equation Based Modeling Smbolic Equation Handling No need for the user to decide which variables to solve for Reduces the number of different components in a librar Redundant states can be eliminated in models Parts of models can be solved smbolicall rather than numericall Smbolic initialization of model variables Increased performance of the simulation is gained at translation time Modelica is a registered trademark of Modelica Association

22 Application Membrane Clinder 22 Pneumatic actuation of testing machines Stead state tests and oscillating loads with higher frequenc Membrane clinder: Low friction losses Low sticking forces Low tolerance required -> low cost Small stroke Application with membrane clinders Fatigue tests up to 50Hz and ±0,5 mm Testing machines (e.g. dental implants ) Zahnimplantat Fiedler, M.: Modellbildung und numerische Optimierung am Beispiel eines servopneumatischen Membranzlinderantriebs, TU Dresden, Dissertation 2010 Quelle: QZM e.v.

23 Application Membrane Clinder 23 Development Objectives Enlargement of amplitude of the membrane clinder wrt. the whole drive concept Modification of the clinder design especiall of the inner geometr Decreasing package dimensions, too Comple relation between stroke, pressure p, chamber volume V, force F F= f(p, A) A=f(p, )

24 Application Membrane Clinder 24 Workflow of optimization prozess Elasticit of membrane FEM Simulation of the membrane Membrane behavior (effective area, volume) Simulation modell of membrane clinder Geometr data Optimization of membrane clinder

25 Application Membrane Clinder 25 Simulation Model Definition of limits of considered design parameters Control of the clinder b a pneumatic valve Calculation of the influence of design parameters to the stead state membrane behavior

26 26 Simulation Model Membrane Clinder - Membrane behavoir dp_bar no friction and damping Implementation of maps for effective area and chamber F_Last volumes _k v_ab dp F_zl _k Membrane Clinder Mechanics pa_minus_pb 2 1 in1 ctr2 source1 3 ctr1 ctr1 ctr2 endstop1 Signal Processing ctr2 lever1 P-Controler (onl option, for valve control) _w 1 sum1 2 K ctr1 mass1 ctr1 ctr2 pb 3 s Application Membrane Clinder sensor1 ctr1 v singlepass1 maimum minimum 1 ampl 3 2 a p Ideal Valve Behavior no dnamics flow like VP l/min Ventil_soll_V Ventil_Signal pa p portp portb V_A spring1 ctr1 ctr2 ctr1 preset1 portp portb V_B Auslass_A dirvalve53 porta pinth in1 porta pinth p_0 in1 port port portr portp ports va Anschluss_A porta porta porta portb portbvb porta portb pinth portb porta Anschluss_B portb Auslass_B port pinth

27 27 Simulation Model for Optimization Membrane Clinder - Membrane behavoir dp_bar no friction and damping Implementation of maps for effective area and chamber F_Last volumes _k v_ab dp F_zl _k Membrane Clinder Mechanics pa_minus_pb 2 1 in1 ctr2 source1 3 ctr1 ctr1 ctr2 endstop1 Signal Processing ctr2 lever1 P-Controler (onl option, for valve control) _w 1 sum1 2 K ctr1 mass1 ctr1 ctr2 pb 3 s Application Membrane Clinder sensor1 ctr1 v singlepass1 maimum minimum 1 ampl 3 2 a p Ideal Valve Behavior no dnamics flow like VP l/min Ventil_soll_V Ventil_Signal pa p portp portb V_A spring1 ctr1 ctr2 ctr1 preset1 portp portb V_B Auslass_A dirvalve53 porta pinth in1 porta pinth p_0 in1 port port portr portp ports va Anschluss_A porta porta porta portb portbvb porta portb pinth portb porta Anschluss_B portb Auslass_B port pinth

28 Optimization with optislang 28 Optimization with optislang ang

29 Optimization with optislang 29 Optimization with optislang ang

30 Optimization with optislang 30 Sensitivit Analsis with optislang First Attempt: 4 Parameters included Sensitivit Analses eecuted CoP 1% results not usable Check model again Include more parameters

31 Optimization with optislang 31 Sensitivit Analsis with optislang 6 Parameters included Sensitivit Analses eecuted CoP 83% Continue with optimization.

32 Interfaces to optislang 32 Benefit of Integration optislang 4 offers a range of direct integration nodes. In the case of SimulationX, the direct interface allows an eas and user-friendl parameter and response definition. In the optimization or calibration analsis, the specified properties are modified directl in the SimulationX model according to the defined ranges and the response values calculated for each design. Using the SimulationX API (COM based), the SimulationX model components, including their properties, can be directl accessed in the parametrization process of optislang.

33 API of SimulationX 33 (Component) Object Model Active Document Document SimObjects Parameters Parameter Objects & Collections SimObjects Results Results Application Simulation models Model components Parameters Result variables Element Tpes Result windows Selection Application Document Tpes Connections Connection Parameters Parameter Results Results Selection Entities Parameters Parameter Results Results SimObjects SimObjects Connections Connection Tpe Entit Result Window Result Window

34 Summar and Outlook 34 Summar Model based design is core of future sstems engineering Use noncausal models for different tasks (Modelica) SimulationX as powerful platform for sstems engineering A pneumatic application used as test case for sensitivit analsis Productiv optimization workflow with optislang 4 Net Steps Further increase efficienc Jump start guide (SimulationX + optislang)

35 Supporting our vision ITI GmbH Headquarters Webergasse 1 Haus C Dresden German T + 49 (0)