Code Generation Verification Workflow with Simulink Test

Open the Fuel-Rate Controller Model

mdl = 'sltestFuelRateControlExample';
open_system(mdl);

Description of the Model

This example uses a model of a fuel-rate controller for a gasoline engine. The controller uses four sensors from the system to determine the proper fuel rate. The four sensors used from the system are throttle angle, speed, EGO and manifold absolute pressure [MAP].

The model uses three subsystems to calculate the fuel rate using the sensor inputs: control logic, airflow calc, and fuel_calc. The core control logic is implemented in the Stateflow® chart named control_logic. The control logic handles single sensor failures and engine overspeed protection. If a single sensor fails, operation continues but the air/fuel mixture is richer to allow smoother running at the cost of higher emissions. If more than one sensor has failed, the engine shuts down as a safety measure, since the air/fuel ratio cannot be controlled reliably.

The model estimates the airflow rate and multiplies the estimate by the reciprocal of the desired ratio to give the fuel rate.

Open the Test Harness

A Test Harness named fuel_rate_control_cgv has been created for the entire model. The harness can be opened by clicking on the perspectives pullout icon in the bottom-right corner of the model canvas and choosing the fuel_rate_control_cgv thumbnail. Ensure that the top level of the model is in view before you click on the icon. Note that the Scope that opens with the test harness is empty until you run the test. Alternately, you can open the harness using the following API:

sltest.harness.open(mdl,'fuel_rate_control_cgv');

Model the Plant

The test harness has been modeled as a closed-loop test with a Test Sequence block to drive fuel-rate controller. The computed fuel_rate from the output of the controller is used to drive a model of the gasoline engine. The fuel rate combines with the actual air flow in the Engine Gas Dynamics subsystem to determine the resulting mixture ratio as sensed at the exhaust. Feedback from the oxygen sensor to the Test Sequence block provides a closed-loop adjustment of the rate estimation in order to maintain the ideal mixture ratio.

Notice that the plant has been modeled in the test harness instead of the main model. The main model is free of extraneous clutter so that code can be easily built for an ECU with minimal changes to the model.

Model Sensor Failures

The Test Sequence block named Sequence Sensor Failures models various sensor failure and engine overspeed scenarios. It accepts feedback from the plant and drives the controller with sensor data. This modeling pattern allows the Test Sequence block to control the feedback signals received by the Controller block and function as a canvas for authoring test cases. Open the Test Sequence block to see the modeled test scenarios.

open_system('fuel_rate_control_cgv/Sequence Sensor Failures');

Test Scenarios

For the first 10 seconds of simulation, the test is in stabilization mode, where the closed loop inputs from the plant is passed through to the controller. The throttle and speed inputs are set to nominal values that are within the normal operating envelope of the controller. The Stabilize_Engine step models this state.

The Test then steps through the following modes:

Test Assessments

The Test Sequence block Assess Controller verifies the controller output for the various test cases modeled by the Sequence Sensor Failures block. The following assessments are modeled:

open_system('fuel_rate_control_cgv/Assess Controller');

Run a Simulation

Simulate the test harness by clicking Play in the toolstrip and observe the fuel_rate and air-fuel ratio signals in the scope. Alternately, run the following command: sim('fuel_rate_control_cgv')

Notice that no assertions trigger during the simulation, which indicate that all assessments modeled in Assess Controller pass.

Configure an Equivalence Test in the Test Manager

As part of code generation verification (CGV) for the controller system, it is important to assert that the functional behavior of the controller is the same during normal and software-in-the-loop (SIL) simulation modes. The Test Manager is used to perform this verification.

Use the function sltestmgr to open the Test Manager and load the example test file using the function: sltest.testmanager.TestFile('sltestFuelRateControlComparisonTestSuite.mldatx')

Model the Test Case

The equivalence test has been configured in the Test Manager so that the controller is simulated in normal and SIL modes and the numerical results are compared between these two runs. Explore the structure of the test case by clicking on different nodes of the test hierarchy in the Test Browser.

Run the Test Case

Run the test in the Test Manager using the function sltest.testmanager.run.

Alternately, in the Test Manager, select the CGV Test1 node in the Test Browser pane and click Run in the toolstrip. The pass/fail result is available in the Results and Artifacts pane.

Create the Report

A report can then be generated to view the result of the equivalence test. Use the following commands to generate the report. You can also launch the report after creation using the API with the LaunchReport option set to true.

sltest.testmanager.report(cgvresult,'cgvresult.zip','IncludeTestResults',int32(0));

close_system(mdl, 0);
clear mdl;