Example of programme control in modelling
Introduction
In this example, we will create a model using programmatic control methods. Our goal is to show how a model can be implemented programmatically using the workspace tools - command line 
and script editor 
.
To create a model using programmatic control methods, we will open the command line or the Engee script editor. The choice of tool is not fundamental and will not differ in syntax if an example is created. However, the script editor will allow refactoring already initialised lines of code, which can be useful for creating your own models. In the future we will use the script editor.
Creating/loading a model
By default, when you start Engee, the newmodel_1 model is created, which is what we will be working with later. However, you can create or load another model by selecting the method you need from the following:
Create/load/open model
Use the create method to create a model:
model = "newmodel_2" # создает переменную model со значением "model_name"
engee.create(model) # создает модель из переменнойor create the model directly without initialising the variable:
engee.create("newmodel_2")You can load an existing model using the load method:
engee.load("/user/newmodel_2.engee")
# или
engee.load("/user/newmodel_2.engee"; force = true) # force = true обязательно указывается в случае, если модель уже была загружена ранееAdding a block
First, let’s add blocks from the library to our newmodel_1 model using the add_block method:
engee.add_block("/Basic/Sources/Sine Wave", "newmodel_1/") # добавляет блок Sin Wave из библиотеки Basic/Sources и присваивает имя автоматическиand similarly for the Terminator block:
engee.add_block("/Basic/Sinks/Terminator", "newmodel_1/") # аналогично добавляет блок Terminator, но уже из другой библиотеки| In this case names are assigned automatically, for example Terminator will change its name to Terminator-1 if a block with this name already exists in the model. |
If necessary, you can name the blocks yourself by adding their name after the slash / of the model name:
engee.add_block("/Basic/Sources/Sine Wave", "newmodel_1/Sine Wave-x") #добавит блок Sine Wave с именем Sin Wave-xYou can delete an unnecessary block via the delete_block method:
engee.delete_block("newmodel_1/Sine Wave-x") # удалит блок Sine Wave-x и все связанные с ним линии и блоки из системыAdding lines
Let’s add a connecting signal line between the added blocks using the add_line method:
engee.add_line("Sine Wave/1", "Terminator/1") # устанавливает сигнал между выходным портом №1 у блока Sin Wave и входным портом №1 блока Terminator|
Enable signal recording for the created signal line between Sin Wave and Terminator blocks by left-clicking on the signal and selecting Record:
This is necessary to retrieve data from the block after the model simulation is complete. |
Setting the parameters of the model
Let’s get the parameters of the model using the get_param method:
engee.get_param("newmodel_1") - # получение параметров моделированияOutput get_param with constant step
ModelParameters( :EnableMultiTasking => false :GenerateComments => true #Параметры интегратора: :StartTime => 0.0 :StopTime => 10 :SolverType => fixed-step :SolverName => Euler :FixedStep => 0.01 )
The list of configurable parameters of the models differs depending on the selected step (constant or variable). For example, get_param for the variable step model mentioned above will show the following parameters:
Variable Step get_param output.
ModelParameters(
:EnableMultiTasking => false
:GenerateComments => true
#Параметры интегратора:
:StartTime => 0.0
:StopTime => 10
:SolverName => Tsit5
:SolverType => variable-step
:MaxStep => auto #максимальный размер шага
:MinStep => auto #минимальный размер шага
:InitialStep => auto #начальный размер шага
:RelTol => auto #относительная точность
:AbsTol => auto #абсолютная точность
:OutputOption => true #плотная выдача
:OutputTimes => 1e-2 #интервал
)Let’s change a few parameters of the model using the set_param! method:
engee.set_param!("newmodel_1", "FixedStep" => 0.05, "StopTime" => 40) # меняем фиксированный размер шага и время окончания симуляцииNext, reuse get_param to track parameter changes:
The output of get_param with changed parameters
ModelParameters( :EnableMultiTasking => false :GenerateComments => true #Параметры интегратора: :StartTime => 0.0 :StopTime => 40.0 :SolverType => fixed-step :SolverName => Euler :FixedStep => 0.05 )
Simulation
Let’s run a simulation of our model using the run method:
engee.run("newmodel_1")Thanks to the previously enabled signal recording, the simout variable storing the simulation results has been given the value SimulationResult("newmodel_1/Sine Wave.1" => WorkspaceArray("newmodel_1/Sine Wave.1")), represented as a table DataFrame:
Output
Dict{String, DataFrames.DataFrame} with 1 entry:
"Sine Wave.1" => 801×2 DataFrame…Let’s save the results of the simulation to the Engee RAM using the collect function as a result variable:
result = collect(simout["newmodel_1/Sine Wave.1"])|
Additionally, let’s get the simulation results using the method get_results without resorting to the simout variable: Both variables with simulation results ( |
For convenience, let’s continue with the result variable. The DataFrame of the result variable consists of two columns - time and value. Let’s visualise the DataFrame using the Plots library:
using Plots
plot(result.time, result.value)
|
Additionally, you can save the simulation results in CSV format: |
Codogeneration
Let’s generate the model code using the generate_code method:
engee.generate_code("/user/newmodel_1.engee", "/user/codegen_dir") # сгенерировали код из модели newmodel_1.engee, код будет храниться в папке codegen_dirThe generated code will be located in the codegen_dir folder in the files newmodel_1.h, newmodel_1.c and main.c.
At the end, let’s save our model using the save method:
engee.save("/user/newmodel_1.engee"; force=true)