Chunk-Based AI Simulation for Secure In-Place Anomaly Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in condition-based maintenance of mechanical systems is the difficulty in sharing user-specific data with AI vendors due to the risk of revealing secret information, improper data analysis, and the need for domain knowledge transfer, leading to inappropriate AI algorithm selection and inefficient data utilization.
Innovation Solution
A simulation apparatus and method that allows users to verify AI effectiveness using unsupervised training with their own data, employing a machine learning model with preprocessing units and a three-layer neural network for anomaly detection, enabling in-place analysis and reducing the risk of data exposure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If user-specific data is shared with AI vendors for training, then AI algorithm accuracy is improved, but the risk of revealing secret information increases
Solution Approach 1:
The patent creates a simulated environment that copies the essential characteristics of real industrial data without using actual proprietary data. The simulation apparatus generates synthetic data that mimics the statistical properties and patterns of real machinery data, allowing AI training while eliminating the risk of exposing sensitive information.
Solution Approach 2:
The simulation apparatus acts as an intermediary between the user's proprietary data and the AI training process. Instead of directly sharing real data, the system uses simulated data as a mediator that preserves the training value while removing sensitive information content.
2Measurement precision
If domain knowledge transfer is required for proper data analysis, then data analysis accuracy is improved, but the complexity of collaboration increases
Solution Approach 1:
The simulation apparatus enables users to independently perform AI algorithm verification without requiring external domain expertise. The system provides built-in simulation capabilities that allow users to test and validate AI algorithms using their own simulated data, eliminating the need for complex knowledge transfer to external vendors.
3Productivity
If AI algorithms are selected without in-place data verification, then development speed is improved, but algorithm appropriateness deteriorates
Solution Approach 1:
The system performs preliminary verification of AI algorithm appropriateness through simulation before actual deployment. Users can test and validate algorithm selection using simulated data in advance, ensuring algorithm suitability is confirmed beforehand without delaying the development timeline.
Data Source
AI summary
In the simulation apparatus, a model setting unit executes setting related to a chunk as a batch block of data in a case where data are sequentially entered into a machine learning model on a basis of loaded data. A model computing unit executes computations of unsupervised training and computations of prediction by sequentially entering the chunk into the machine learning model. A model storage unit is configured to non-temporarily store not only the machine learning model before execution of the computations of training but also the machine learning model after execution of at least part of the computations of training.