Environment Controller Validation of Neural Network Predictive Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current environment control systems face challenges in seamlessly integrating neural networks for predictive modeling while maintaining compatibility with legacy software modules, and there is a need for a method to validate neural network models within these systems.
Innovation Solution
An environment controller with a communication interface and processing unit that receives environmental data, determines input variables, and transmits them to a training server for neural network processing, allowing for comparison and validation of output variables with legacy software module outputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a neural network predictive model is integrated into the environment control system, then the system becomes more resilient to changes and unexpected conditions, but the device complexity increases due to the need for training servers, communication interfaces, and model validation mechanisms
Solution Approach 1:
The patent introduces a training server as an intermediary component that handles the complex neural network training and validation processes externally. The environment controller communicates with this intermediary to obtain predictive models without having to implement the full training infrastructure itself, thus reducing device complexity while maintaining reliability benefits
Solution Approach 2:
The system is segmented into distinct functional components: the environment controller for data collection and command execution, the training server for model development and validation, and the predictive model as a separate asset. This segmentation allows each component to be optimized independently and simplifies the overall system architecture
2Adaptability or versatility
If the legacy environment control software module is kept alongside the neural network technology, then the system maintains compatibility with existing infrastructure, but the device complexity increases due to running and validating multiple software modules
Solution Approach 1:
The validation mechanism uses feedback from the legacy software module's outputs to assess the predictive model's performance. By comparing predictions against the legacy module's decisions on the same input data, the system can iteratively improve the neural network model while maintaining legacy compatibility, managing complexity through structured comparison protocols
Solution Approach 2:
The system creates a copy of the validation process where the same input data is processed by both the legacy software module and the predictive model. This copying approach enables parallel validation without requiring modification of the original legacy module, maintaining compatibility while enabling model assessment
Data Source
AI summary
Method and environment controller for validating a predictive model of a neural network. The environment controller receives at least one environmental characteristic value and determines a plurality of input variables. At least one of the plurality of input variables is based on one among the environmental characteristic value(s). The environment controller executes an environment control software module for calculating at least one output variable based on the plurality of input variables. The environment controller transmits the plurality of input variables to a training server executing a neural network training engine using the predictive model; and receives at least one inferred output variable from the training server. Each inferred output variable corresponds to one of the at least one output variable calculated by the environment control software module. The environment controller compares each inferred output variable with the corresponding calculated output variable; and sends a feedback to the training server.