Control Device Software Image for Multi-Instance Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for developing and testing control devices require multiple real devices for simulation, which is costly and inefficient, and struggle with training software images using variable sampling rates and complex signal processing.
Innovation Solution
A method using artificial neural networks or support vector machines to create a software image of a real control device by mapping relevant input and output variables, allowing for numerical simulation and training through supervised or reinforcement learning, enabling the creation of a behavior map that can be used in various simulations without relying on multiple real devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple real control devices are used for simulation and testing, then the reliability and accuracy of simulation results are improved, but the cost and device complexity increase significantly
Solution Approach 1:
The patent creates a software image that copies the behavior and functionality of the real control device. This software image can be instantiated multiple times in simulations without requiring multiple physical devices, thus maintaining simulation accuracy while reducing hardware requirements
Solution Approach 2:
The patent transforms the control device into a software-based representation with adjustable parameters. By changing from physical hardware to software parameters, the system can be replicated and modified without additional hardware costs, resolving the contradiction between reliability and device complexity
2Adaptability or versatility
If multiple real control devices are deployed for comprehensive testing, then the coverage of test scenarios is improved, but the cost and resource requirements worsen
Solution Approach 1:
The software image serves multiple functions and can be used across different test scenarios and simulation environments. A single software image can be instantiated multiple times with different configurations, providing universal test coverage without requiring proportional increases in physical devices
3Reliability
If real control devices are used directly in simulations, then the authenticity of simulation data is improved, but the flexibility and speed of training and simulation deteriorate
Solution Approach 1:
The patent replaces the mechanical/physical control device system with a software-based system. This substitution maintains data authenticity by preserving the control logic while enabling faster operation, parallel processing, and more flexible training scenarios that are not constrained by physical hardware limitations
Data Source
AI summary
A method for creating a software image of at least a part of a control device for a numerical simulation, the control device mapping an input vector of control device input variables to an output vector of control device output variables during operation. The creation of the software image is formed by an artificial neural network or a support vector machine, using an input vector of map input variables having control device input variables of interest, and using an output vector of map output variables having control device output variables of interest. The software image is trained with the aid of supervised learning or with the aid of reinforcement learning, using a plurality of training input vectors of the control device input variables of interest and, in the case of the supervised learning, also using a plurality of corresponding training output vectors of the control device output variables of interest.