9 Using Equation Networks

Transcription

1 9 Using Equation Networks In this chapter Introduction to Equation Networks 244 Equation format 247 Using register address lists 254 Setting up an enable contact 255 Equations displayed within the Network Editor 256

2 ProWORX NxT User s Guide Introduction to Equation Networks An Equation Network provides an easy way to program complex math functions, with values stored in register locations. Equations in an Equation Network are presented in a regular, left-to-right format, technically known as infix notation. You program Equation Networks and set its enable contact and output coil(s) in the Equation Network Editor. Equation Networks were introduced in Quantum Rev. 2 controllers; not all controllers support Equation Networks. The easiest way to see if your controller supports Equation Networks is by trying to create a new one if your controller doesn t support it, the Equation Network option on the Insert menu won t be available. To create a new Equation Network: 1. In the main Network Editor, move to the logic network where you want the new Equation Network. Each Equation Network takes up an entire network s worth of logic. 2. Click Edit, then from the Insert menu, click Equation Network. The Equation Network Editor display appears and a blank Equation Network is inserted into your logic. To edit an existing Equation Network: 1. In the main Network Editor, move the cursor to the top row of the Equation Network you want to edit. 2. From the Edit menu, click Edit Equation Network. (When the cursor is on an Equation Network, this option replaces Edit Instruction in the Edit menu.) - Or - Double-click anywhere in the Equation Network. 244 Introduction to Equation Networks

3 The Equation Network Editor display The Equation Network Editor display looks like this: Chapter 9 Using Equation Networks Each element of an equation takes up a certain amount of space, measured in nodes. This text tells you how much space you have left. These buttons switch between four different panels in the Equation Network Editor: Coils (shown), Operators, Functions, and Symbols. You select an enable contact and its register address in this panel. The equation appears in this editor. You can type the equation in directly, if you wish. Click for a list of previously-used addresses. Click to view the Descriptor Summary. The register addresses for the Equation Network s output coils are set in the Coils panel. While you can type in the equation from the keyboard, three list panels. For example, the Operators, Functions and Symbols panels let you insert items directly into your equation. To add an item from a list panel to your Equation Network: 1. Select the list panel from which you want to add an item. 2. Move the equation cursor to where you want to add the item. 3. In the list box, double-click the item you want to add. If any text within the equation is selected, the new item will replace it. - Or - Drag the item from the panel onto the equation or address box. The Equation Network Editor display 245

4 ProWORX NxT User s Guide Coils panel You set addresses for the output coils in the Coils panel. There are five possible output coils for an Equation Network: Solved OK is set when the equation is being solved without errors. Result<0 is set when the equation result is less than zero. Result=0 is set when the equation result is equal to zero. Result>0 is set when the equation result is greater than zero. Error is set when errors have occurred while solving the equation. While online, if the Error coil receives power, an error message will appear under the coil describing the error. To change an address for one of the output coils: From the Equation Network Editor, click the appropriate text box in the output coils panel and type the register address for that coil. You can enter either the direct address (in X:Y numeric format) or a symbolic address. You can also insert addresses from the Symbols list panel, Used Register Address table and the Descriptor Summary. If you don't want to use a particular output coil, leave the address for that coil blank (or erase one already typed in). That coil will not be included in the Equation Network. Operators list panel The Operators list panel contains a list of mathematical operators recognized by the Equation Network Editor. This includes operators for addition, multiplication, exponentiation, boolean, and comparison operations. You can insert an item from the Operators panel directly into your equation. For a complete list of operators and their meanings, see Mathematical operations in equation networks on page The Equation Network Editor display

5 Equation format Functions list panel Chapter 9 Using Equation Networks The Functions list panel contains a list of standard math functions you can include in your equation. Each of these functions takes one argument enclosed in brackets following the function name. The argument can be any valid value or expression. For example, COS(# ) returns the cosine of 35 plus the number stored at address You can insert a function from the Functions panel directly into your equation. For a complete list of functions, see Mathematical functions in equation networks on page 253. Symbols list panel The Symbols list panel contains a list of all previously-defined symbols for use in an Equation Network. You can insert an item from the Symbols panel directly into your equation, an output coil address box or the enable contact address box. For more information on symbols, see Symbolic addressing on page 179. ProWORX NxT expects equation elements to appear in a specific format. Operations and functions each have their own format. Also, for each value, you must specify what kind of value it is (register address, constant or symbol) and its data type (signed integer, unsigned integer, etc.). Equation values and data types Each value can refer to a constant, register address or symbol. The Equation Network Editor determines which one it is based on the following format: Format Meaning Example Default (no # sign or single quotes) Register address Prefixed by # Constant #123 Enclosed in single quotes Symbol HEIGHT Equation values and data types 247

6 ProWORX NxT User s Guide The actual data type of a value is determined by its suffix, as shown in the following table: Suffix Meaning Example None 16-bit signed integer #38 U 16-bit unsigned integer 40001U L Long (32-bit) signed integer #-123L UL Long (32-bit) unsigned integer HEIGHT UL F 32-bit floating point (real) #+1.45E-4F Typically, you d first indicate the register address where the calculated result is to be stored, followed by an equal sign (the assignment operator ), followed by the calculation itself. For example: = 40002U + COS(40003UL) * #+1.35E-4F / 'HEIGHT'L 40002U is an address of a 16-bit unsigned integer. COS(40003UL) calculates the cosine of a long (32-bit) unsigned integer value stored at address #+1.35E-4F is the floating point value of , given in exponential notation. HEIGHT L is a symbol of the name HEIGHT, representing the address of a long (32-bit) signed integer = indicates that the result of the calculation is to be stored in register address as a 16-bit signed integer. Everything to the right of the assignment operator also constitutes an expression. An expression is any part of an equation that can be evaluated to a single value. This can be a single constant or register address, or a complete mathematical operation. For example, #35 is an expression, as are LOG(#10) and 40002U + COS(40003UL). Complex expressions can contain other expressions within them, as in #3 * (40002U + COS(40003UL)). For the most part, any operator or function can be performed on any expression, no matter how complex. It is good programming practice to enclose all expressions in parentheses, even when they re not actually needed. This makes the equation easier to read and ensures that operations in an equation are solved in the correct order. 248 Equation values and data types

7 Exponential notation Chapter 9 Using Equation Networks Floating point numbers are normally specified in exponential notation, as in: +1.35E-4 This represents 1.35 times 10 to the -4th power, or 1.35 times Thus, we would shift the decimal place four places to the left to get The -4 part is called the exponent (note the preceding E ) and can be a positive or negative number. In the Equation Network Editor, you must also indicate: That these numbers are constants; and Their data types. For example, integers or floating point numbers. The default data type is unsigned 16-bit integer. So, since the above value is a fraction (and therefore must be a floating point number), it would have to appear as #+1.35E-4F. With no data type suffix, numbers in exponential notation are assumed to be integers. For example, #+1.35E+2 represents the unsigned 16-bit integer value 135. Exponential notation is particularly useful for very large integers. Mathematical operations in equation networks The following table lists the mathematical operations you can include in your equation. Assignment operator The assignment operator = is used to assign a storage place for the results of the equation. All equations will use the assignment operator. The format is: ADDRESS = EXPRESSION Where ADDRESS is a valid register address and EXPRESSION is a valid value or expression assigned to the address. = Assignment. Mathematical operations in equation networks 249

8 ProWORX NxT User s Guide Unary operators Unary means single, so unary operators are used on only one value. The unary operator is placed just before the value or expression to which it is applied. For example, - (30002) returns -1 times the number stored at address Exponentiation operator Takes values to a specified power **3 returns the (integer) value stored at 40001, taken to the third power. Arithmetic operators These require two values, one before and one after the operator. These values can be any valid expression. For example, #4 * results in four multiplied by the value stored at address Negation. The result is -1 times the value. ~ Ones complement. This works on the binary representation of a value: all 1s are changed to 0s and vice versa. ** Exponentiation. * Multiplication / Division + Addition - Subtraction 250 Mathematical operations in equation networks

9 Chapter 9 Using Equation Networks Bitwise operators Bitwise operators work on binary (base 2) representations of values. In the case of AND, OR and XOR, the computer applies the operator to each digit in the two values: 010 XOR 011 (2 XOR 3 in decimal numbers) results in 001 (1 in decimal). In the case of shifting operators, the computer shifts all digits in the binary representation of the number the given number of places to the left or right. Digits on one side of the number are lost, and zeros fill in the blanks on the other side. For example, for 8-bit numbers, 77 << 2 means shifted left two digits. The binary result is , or 52 decimal Relational operators These operators describe a comparison between two values or expressions. The result is always true (1) or false (0). For example, #35 <= #42 evaluates to 1 (true). Relational operators are used in Conditional expressions. & AND. The single bit result of an AND operation is only true (1) if both bits are set to 1. OR. The single bit result of an OR operation is true (1) if either bit is set to 1. The result is false (0) only if both bits are set to 0. ^ XOR, short for exclusive or. The single bit result of an XOR operation is false (0) if both bits are the same, true (1) otherwise. << Left shift. The result of 40001<<#2 is the binary representation of the number stored at shifted left two (#2) places. Zeros are added on the right to fill in the gap. >> Right shift. The result of 40001>>#2 is the binary representation of the number stored at shifted right two (#2) places. Zeros are added on the left to fill in the gap. < Less than. <= Less than or equal to. = Equal to. <> Not equal to. => Greater than or equal to. > Greater than. Mathematical operations in equation networks 251

10 ProWORX NxT User s Guide Conditional operators See page252 for details. Parentheses Used to set precedence in solving equations. To make sure certain operations are solved before others, enclose those operations in parentheses.? : Used in a conditional expression. ( ) Conditional expressions In Equation Networks, conditional expressions take the following form: EXPR1 RELOP (EXPR2)? (EXPR3) : EXPR4 EXPR1 through EXPR4 can be any address, value or expression. If EXPR2 or EXPR3 are not single values or addresses, they must be enclosed in brackets. RELOP can be any relational operator, such as > (greater than) or <> (not equal to). When solving a conditional expression, the computer first compares EXPR1 and EXPR2, based on the relational operator. If the comparison is true, the value of the conditional expression is the result of EXPR3. If the comparison is false, the value of the conditional expression is EXPR4. It is especially good programming practice to enclose all expressions within a conditional expression in parentheses, even when they re not actually needed. This makes the equation easier to read and ensures that operations are solved in the correct order. 252 Mathematical operations in equation networks

11 Mathematical functions in equation networks Chapter 9 Using Equation Networks The following table lists the pre-defined math functions you can include in your equation. Each of these functions takes one argument enclosed in brackets following the function name. The argument can be any valid value or expression. For example, COS(# ) returns the cosine of 35 plus the number stored at address In this table, X refers to a function s argument (as in COS(X) ). ABS(X) ARCCOS(X) ARCSIN(X) ARCTAN(X) COS(X) COSD(X) EXP(X) FIX(X) FLOAT(X) Absolute value of X (i.e. negative numbers become positive). Arc cosine of X radians. Arc sine of X radians. Arc tangent of X radians Cosine of X radians. Cosine of X degrees. Calculates e (approximately ) to the Xth power. Converts floating point number X to an integer. Converts integer X to a floating point number. LN(X) Natural (base e) logarithm of X. LOG(X) Common (base 10) logarithm of X. SIN(X) SIND(X) Sine of X radians. Sine of X degrees. SQRT(X) Square root of X. TAN(X) TAND(X) Tangent of X radians. Tangent of X degrees. Mathematical functions in equation networks 253

12 ProWORX NxT User s Guide Using register address lists Click the Used to bring up a list of previously-used addresses register addresses that have been used elsewhere in your logic database. Click the Summary to bring up the Descriptor Summary display for the address currently selected. This includes addresses for output coils and the enable contact. If the cursor does not highlight a register address, the Descriptor Summary will jump to address Addresses from both tables can be inserted directly into the equation. To insert an address from the Previously-Used Address table or Descriptor Summary: 1. Move the cursor to the spot in the equation (or to a coil address box) where you want to insert the address. 2. Click Used or Summary. The appropriate table appears. 3. Navigate through the table until you find the address you want to add to the equation. 4. Double-click that address. The address is inserted at the current cursor position. If any text was selected, it will be overwritten. 254 Mathematical functions in equation networks

13 Chapter 9 Using Equation Networks Setting up an enable contact An Equation Network s enable contact, when set, activates the Equation Network. If an enable contact passes current, the Equation Network will be solved. You change settings for the enable contact in the Enable Editor display. To select a type for the enable contact: Select the symbol of the enable contact that corresponds with your chosen type. An enable contact can be a normally-open contact( ), normally-closed contact ( ), horizontal short ( ), or a horizontal open ( ). To select a register address for the enable contact: In the Enable Contact address field, type the direct address (in X:Y numeric format) or symbolic address for the enable contact coil. This field is only available if the enable contact type is a normally-open or normally closed contact. - Or - Insert an address into the Enable Contact address field from the Symbols list panel, Used-Register Address table, or the Descriptor Summary. Setting up an enable contact 255

14 ProWORX NxT User s Guide Equations displayed within the Network Editor When you scroll to an Equation Network from the main Network Editor, the display will show the text of the equation you created, along with any input or output coils. There are several things to keep in mind when viewing Equation Networks from the main Network Editor: Blocking: Blocking operations on Equation Networks are limited to single cells or full network blocks. The only blocking operation not allowed is a block save saved blocks cannot have Equation Networks. Cursor movement: You can only move the cursor onto the input contact or output coils of an Equation Network. Cut and Paste: You can Copy part of the Equation Network into the clipboard, but you cannot Paste into it. Similarly, the Cut operation will copy material into the clipboard, but will not delete anything. The Equation Network Editor must be used for modifying Equation Networks. Display menu: Tables and lists created using commands from the Display menu will include items used in Equation Networks. Find, Locate Coil and Global searches: All search operations will find (and replace) searched-for items within Equation Networks. The search result tables will show Eqn for any addresses that are part of the actual equation of an Equation Network. Instruction Speed buttons: These have no effect when viewing Equation Networks. Online mode: Power flow is shown using settings from the Network Display setup dialog box. Any contact or coil receiving power will show power flow. If the enable contact passes power, the equation box will also change to show power flow. Forcing an Equation Network output coil on or off affects how power is passed through that coil; however, the power flow of the output coil itself is still displayed based on the status of the Equation Network. For example, Result>0 will still show power if an Equation Network s Result is greater than 0. This may make for some strange visual effects, such as an output coil that seems to be passing flow but doesn't affect anything that follows. 256 Equations displayed within the Network Editor

15 Chapter 9 Using Equation Networks The following Network Editor features are disabled when you move the cursor onto an Equation Network: Assembly Register (AR) operations that modify parts of an Equation Network. Functions that force discrete variables will work as usual. Insert Row/Column Delete Element/Row/Column Drag and drop operations from side panels. Equations displayed within the Network Editor 257

16