Goal Seek Analysis Using Combined Normal and Abnormal Status Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning models built using both normal and abnormal status data records face reduced accuracy when attempting to return a system from an abnormal status to normal, as they are partially trained on incorrect data types, leading to ineffective goal seek analysis.
Innovation Solution
A combination model is created by building separate normal and abnormal status models and computing time-sequenced coefficient combinations, which are then used for goal seek analysis to adjust influential effect values and improve accuracy in returning the system to a normal status.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single machine learning model is built using both normal and abnormal status data records, then the model can handle both normal and abnormal situations, but the accuracy is reduced because the model is partially trained on incorrect data types
Solution Approach 1:
The patent divides the single machine learning model into two separate models: a normal status model trained only on normal status data records and an abnormal status model trained only on abnormal status data records. This segmentation allows each model to specialize in its specific data type, improving accuracy while maintaining adaptability through dynamic model selection based on current system status.
2Measurement precision
If separate normal and abnormal status models are built, then the accuracy of goal seek analysis improves, but the device complexity increases due to maintaining multiple models
Solution Approach 1:
The patent combines the normal status model and abnormal status model into a unified combination model that dynamically integrates both models based on current system status. This merging approach maintains the accuracy benefits of separate specialized models while reducing complexity by providing a single interface for goal seek analysis operations.
Solution Approach 2:
The patent implements dynamic model selection and coefficient adjustment based on real-time system status monitoring. The combination model adapts its composition and parameters dynamically, switching between normal and abnormal status models as needed, which automates model management and reduces operational complexity.
3Measurement precision
If time-sequenced coefficient combinations are computed and applied, then the accuracy in returning system to normal status improves, but the computation time and processing complexity increase
Solution Approach 1:
The patent pre-computes and stores optimal coefficient combinations for transitioning between different system statuses during the model training phase. These pre-computed coefficients are then directly applied during runtime goal seek analysis, eliminating the need for complex real-time optimization calculations and significantly reducing computation time while maintaining high accuracy.
Data Source
AI summary
An approach is provided in which the approach builds a combination model that includes a normal status model and an abnormal status model. The normal status model is built from a set of time-sequenced normal status records and the abnormal status model is built from a set of time-sequenced abnormal status records. The approach computes a set of time-sequenced coefficient combination values of the normal status model and the abnormal status model based on applying a set of fitting coefficient characteristics to the normal status model and the abnormal status model. The approach performs goal seek analysis on a system using the combination model and the set of time-sequenced coefficient combination values.