Loading [Contrib]/a11y/accessibility-menu.js
Skip to content

Water Tank Fault Injection with NuRV Runtime Monitor

Overview

This example shows a fault injection (FI) enabled digital twin (DT). A live DT is subjected to simulated faults received from the environment. The simulated faults is specified as part of DT configuration and can be changed for new instances of DTs. This version of the example is expanded with a monitor generated by NuRV. More information about NuRV is available here.

In this co-simulation based DT, a watertank case-study is used; co-simulation consists of a tank and controller. The goal of which is to keep the level of water in the tank between Level-1 and Level-2. The faults are injected into output of the water tank controller (Watertankcontroller-c.fmu) from 12 to 20 time units, such that the tank output is closed for a period of time, leading to the water level increasing in the tank beyond the desired level (Level-2). Additionally, a monitor is inserted in the simulation to check at runtime whether the level of the water stays below a maximum threshold.

Example Diagram

Water Tank System

Example Structure

Water Tank Structure

Digital Twin Configuration

This example uses two models and one tool. The specific assets used are:

Asset Type Names of Assets Visibility Reuse in Other Examples
Models watertankcontroller-c.fmu Private Yes
singlewatertank-20sim.fmu Private Yes
m1.fmu Private No
RtI.fmu Private Yes
Specification m1.smv Private No
Tool maestro-2.3.0-jar-with-dependencies.jar Common Yes

The multimodelFI.json and simulation-config.json are two DT configuration files used for executing the digital twin. You can change these two files to customize the DT to your needs.

The faults are defined in wt_fault.xml.

Lifecycle Phases

Lifecycle Phase Completed Tasks
Create Installs Java Development Kit for Maestro tool
Generates and compiles the monitor FMU
Execute Produces and stores output in data/water_tank_FI_monitor/output directory
Clean Clears run logs and outputs

Run the example

To run the example, change your present directory.

cd /workspace/examples/digital_twins/water_tank_FI_monitor

If required, change the execute permission of lifecycle scripts you need to execute, for example:

chmod +x lifecycle/create

Now, run the following scripts:

Create

Installs Open Java Development Kit 17 and pip dependencies. The pandas and matplotlib are the pip dependencies installated. The monitor FMU from the NuRV specification is generated and compiled.

lifecycle/create

Execute

Run the co-simulation. Generates the co-simulation output.csv file at /workspace/examples/data/water_tank_FI_monitor/output.

lifecycle/execute

Analyze phase

Process the output of co-simulation to produce a plot at: /workspace/examples/data/water_tank_FI_monitor/output/plots/.

lifecycle/analyze

A sample plot is given here.

sample plot

In the plot, four color-coded indicators are used to represent different values. The blue line shows the water tank level, while orange represents the control output and green indicates the monitor's verdict. A red dashed line serves as a reference point, marking the level checked by the monitor. As the water level exceeds this threshold, the monitor's verdict changes from unknown (0) to false (2).

Examine the results

The results can be found in the /workspace/examples/data/water_tank_FI_monitor/output directory.

You can also view run logs in the /workspace/examples/digital_twins/water_tank_FI_monitor.

Terminate phase

Clean up the temporary files and delete output plot

lifecycle/terminate

References

More details on this case-study can be found in the paper:

1
2
3
4
M. Frasheri, C. Thule, H. D. Macedo, K. Lausdahl, P. G. Larsen and
L. Esterle, "Fault Injecting Co-simulations for Safety,"
2021 5th International Conference on System Reliability and Safety (ICSRS),
Palermo, Italy, 2021.

The fault-injection plugin is an extension to the Maestro co-orchestration engine that enables injecting inputs and outputs of FMUs in an FMI-based co-simulation with tampered values. More details on the plugin can be found in fault injection git repository. The source code for this example is also in the same github repository in a example directory.