Training of machine learning models with hardware-in-the-loop simulations

A system comprising:

• a software simulation system including a processor and memory for executing a software simulation of a physical system, the software simulation system being configured to generate emulated sensor data based on input to the software simulation system, the emulated sensor data emulating output from one or more sensors of the physical system, wherein the input to the software simulation system comprises a simulated fault of the physical system introduced to the software simulation by:
• utilizing a Monte Carlo technique to generate random data representing the simulated fault;
• and
• utilizing the Monte Carlo technique to inject the random data into the software simulation at a random interval;
• and
• a test control system emulating a control system of the physical system, the test control system configured to generate control system output based on the emulated sensor data, wherein the control system output comprises a simulated control signal generated by the test control system in response to the emulated sensor data.

A method comprising:

• receiving, at a hardware-in-loop (HIL) simulation of a physical system, programmatic data input including at least one injected fault, the at least one injected fault being representative of a first physical condition in a physical system, the at least one injected fault being injected into the programmatic data input using a Monte Carlo technique, wherein injecting the at least one injected fault into the programmatic data input comprises:
• utilizing the Monte Carlo technique to generate random data representing the at least one injected fault;
• and
• utilizing the Monte Carlo technique to inject the random data into the programmatic data input at a random interval;
• generating, by the HIL simulation, control system outputs, the control system outputs corresponding to outputs of a control system controlling the physical system, wherein the control system outputs comprise simulated control signals generated by the HIL simulation in response to the at least one injected fault;
• generating a dataset by tagging the control system outputs using the programmatic data input including the at least one injected fault;
• and
• training a control model using the dataset, the control model being trained to identify the first physical condition in the physical system based on the outputs of the control system controlling the physical system.