Buildings.Utilities.IO.Python27

This package contains blocks and functions that embed Python 2.7 in Modelica. Data can be sent to Python functions and received from Python functions. This allows for example data analysis in Python as part of a Modelica model, or data exchange as part of a hardware-in-the-loop simulation in which Python is used to communicate with hardware.

Package Content

Name	Description
UsersGuide	User's Guide
Real_Real	Block that exchanges a vector of real values with a Python function
Functions	Package with functions that call Python
Examples	Collection of models that illustrate model use and test models

Buildings.Utilities.IO.Python27.Real_Real

Information

For each element in the input vector uR[nDblWri], the value of the flag flaDblWri[nDblWri] determines whether the current value, the average over the sampling interval or the integral over the sampling interval is sent to Python. The following three options are allowed:

Parameters

Connectors

Modelica definition

Type	Name	Default	Description
Time	samplePeriod		Sample period of component [s]
Time	startTime	0	First sample time instant [s]
String	moduleName		Name of the python module that contains the function
String	functionName	moduleName	Name of the python function
Integer	nDblWri		Number of double values to write to Python
Integer	nDblRea		Number of double values to be read from the Python
Integer	flaDblWri[nDblWri]	zeros(nDblWri)	Flag for double values (0: use current value, 1: use average over interval, 2: use integral over interval)

Type	Name	Description
input RealInput	uR[nDblWri]	Real inputs to be sent to Python
output RealOutput	yR[nDblRea]	Real outputs received from Python

model Real_Real "Block that exchanges a vector of real values with a Python function" extends Modelica.Blocks.Interfaces.DiscreteBlock( startTime=0, firstTrigger(fixed=true, start=false)); parameter String moduleName "Name of the python module that contains the function"; parameter String functionName=moduleName "Name of the python function"; parameter Integer nDblWri(min=1) "Number of double values to write to Python"; parameter Integer nDblRea(min=1) "Number of double values to be read from the Python"; parameter Integer flaDblWri[nDblWri] = zeros(nDblWri) "Flag for double values (0: use current value, 1: use average over interval, 2: use integral over interval)"; Modelica.Blocks.Interfaces.RealInput uR[nDblWri] "Real inputs to be sent to Python"; Modelica.Blocks.Interfaces.RealOutput yR[nDblRea] "Real outputs received from Python"; Real uRInt[nDblWri] "Value of integral"; Real uRIntPre[nDblWri] "Value of integral at previous sampling instance"; Real uRWri[nDblWri] "Value to be sent to Python"; initial equation uRWri = pre(uR); uRInt = zeros(nDblWri); uRIntPre = zeros(nDblWri); for i in 1:nDblWri loop assert(flaDblWri[i]>=0 and flaDblWri[i]<=2, "Parameter flaDblWri out of range for " + String(i) + "-th component."); end for; // The assignment of yR avoids the warning // "initial conditions for variables of type Real are not fully specified". // At startTime, the sampleTrigger is true and hence this value will // be overwritten. yR = zeros(nDblRea); equation for i in 1:nDblWri loop der(uRInt[i]) = if (flaDblWri[i] > 0) then uR[i] else 0; end for; when {sampleTrigger} then // Compute value that will be sent to Python for i in 1:nDblWri loop if (flaDblWri[i] == 0) then uRWri[i] = pre(uR[i]); // Send the current value. else uRWri[i] = uRInt[i] - pre(uRIntPre[i]); // Integral over the sampling interval if (flaDblWri[i] == 1) then uRWri[i] = uRWri[i]/samplePeriod; // Average value over the sampling interval end if; end if; end for; // Exchange data yR = Buildings.Utilities.IO.Python27.Functions.exchange( moduleName=moduleName, functionName=functionName, dblWri=uRWri, intWri={0}, nDblWri=nDblWri, nDblRea=nDblRea, nIntWri=0, nIntRea=0, nStrWri=0, strWri={""}); // Store current value of integral uRIntPre= uRInt; end when; end Real_Real;

flaDblWri[i]	Value sent to Python
0	Current value of uR[i]
1	Average value of uR[i] over the sampling interval
2	Integral of uR[i] over the sampling interval