Power Asset Monitoring with Parameter Evolution Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional model-based approaches for monitoring power system assets are inadequate as they fail to systematically capture long-term trends and require significant human expertise, leading to errors and inefficiencies in reflecting the actual state of the assets over time.
Innovation Solution
A hybrid data-driven and model-based system that uses a parameter evolution model to periodically update the parameter values of a power system asset model, allowing for systematic capture and prediction of trends, reducing human involvement through automated diagnostics and control signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional model-based approaches are used to monitor power system assets, then the system can operate with existing models, but the models fail to systematically capture long-term trends and require significant human expertise leading to errors
Solution Approach 1:
The patent introduces a parameter evolution model as an intermediary between the power system asset and the monitoring system. This model systematically captures long-term trends in asset parameters without requiring human expertise to interpret complex degradation patterns. The parameter evolution model acts as a mediator that translates raw asset data into meaningful trend information, resolving the contradiction between measurement precision and device complexity.
Solution Approach 2:
The monitoring system performs automated diagnostics and control signal generation without requiring human intervention. The parameter evolution model automatically updates and predicts asset state, enabling the system to serve itself by eliminating the need for human experts to continuously analyze asset conditions, thereby reducing errors while maintaining high measurement precision.
2Adaptability or versatility
If parameter values are re-determined in a recurrent manner to reflect current asset state, then the model can adapt to degradation, but long-term trends cannot be systematically captured
Solution Approach 1:
The patent employs a nested structure where the parameter evolution model is embedded within the power system asset model. The parameter evolution model captures long-term trends while the inner model handles short-term operational states. This nested arrangement allows both recurrent adaptation to current degradation and systematic capture of long-term trends to coexist, preventing loss of trend information while maintaining model adaptability.
3Ease of manufacture
If conventional model-based approaches are used, then configuration can be performed manually, but the process is error-prone and requires significant human involvement
Solution Approach 1:
The system automatically configures and updates the parameter evolution model without human intervention. The model self-adjusts to asset degradation patterns and generates control signals autonomously, eliminating manual configuration processes that are prone to human error. This automation maintains ease of operation while significantly improving reliability by removing the error-prone human element from the configuration process.
Data Source
AI summary
Techniques for monitoring and/or operating a power system asset are provided. A series of sets of first model parameter values (71-73) of a power system asset model is determined from data obtained by measurements. The series of sets of first model parameter values (71-73) is used to determine a set of second model parameter values (79) of a parameter evolution model different from the power system asset model. The parameter evolution model describes an evolution of one, several or all first model parameter values (71-73) of the power system asset model. An output is generated in dependence on at least one of the sets of first model parameter values (71-73) of the power system asset model and the set of second model parameter values (79) of the parameter evolution model.