Power Transformer RUL Estimation Using Real-Time Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current power transformer monitoring techniques, such as dissolved gas analysis (DGA), are time-consuming, expensive, and reactive, failing to provide timely and accurate assessments of transformer health, leading to potential blackouts and property damage.
Innovation Solution
A system that uses real-time sensor data and periodic dissolved gas analysis measurements to estimate the remaining useful life (RUL) of power transformers through an inferential model, sequential probability ratio test (SPRT), and logistic-regression model, generating notifications for replacement based on computed risk indices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dissolved gas analysis (DGA) is performed periodically to detect transformer health issues, then measurement precision of transformer condition is improved, but loss of time and productivity deteriorate due to infrequent monitoring
Solution Approach 1:
The patent replaces the mechanical/chemical DGA process (oil sampling and laboratory analysis) with an electronic sensor-based monitoring system that continuously measures transformer parameters such as temperature, voltage, and current. This substitution enables real-time health assessment without the time-consuming sampling and analysis procedures, resolving the contradiction between measurement precision and monitoring frequency.
Solution Approach 2:
The patent implements continuous monitoring of transformer parameters through sensors that operate without interruption, providing ongoing health assessment. This continuous action replaces the periodic, discontinuous nature of traditional DGA, allowing the system to maintain high measurement precision while eliminating time losses associated with periodic sampling and analysis.
2Reliability
If dissolved gas analysis (DGA) is performed to detect transformer problems, then reliability of fault detection is improved, but device complexity and cost worsen due to extensive chemical analysis procedures
Solution Approach 1:
The patent replaces the complex chemical analysis system (requiring oil sampling, transportation, laboratory equipment, and chemical procedures) with a simplified electronic sensor system. The sensors directly measure electrical and thermal parameters, eliminating the need for chemical DGA procedures while maintaining or improving fault detection reliability through continuous real-time data collection.
Solution Approach 2:
The patent extracts only the essential monitoring function from the complex DGA process, focusing on measuring key transformer parameters (temperature, voltage, current) that directly indicate health status. By taking out and focusing on these critical measurements, the system achieves reliable fault detection without the unnecessary complexity of full chemical analysis procedures.
3Measurement precision
If dissolved gas analysis (DGA) is used to detect transformer issues, then measurement precision of degradation is improved, but loss of time worsens because DGA is reactive rather than prognostic
Solution Approach 1:
The patent implements preliminary action by continuously monitoring transformer parameters and using predictive algorithms to forecast future degradation trends before they lead to failure. The system analyzes real-time data to predict remaining useful life and potential failure modes, enabling proactive maintenance scheduling. This preliminary assessment allows the system to detect and respond to degradation issues before they become critical, eliminating the reactive delay inherent in traditional DGA.
Solution Approach 2:
The patent incorporates feedback mechanisms where real-time sensor measurements are continuously fed into predictive models that update degradation assessments and adjust maintenance recommendations. This closed-loop feedback system enables the monitoring to be both precise in measuring current degradation and proactive in predicting future states, resolving the contradiction between measurement precision and response time by providing both accurate current status and forecasted future conditions.
Data Source
AI summary
During a surveillance mode, the system receives present time-series signals gathered from sensors in the power transformer. Next, the system uses an inferential model to generate estimated values for the present time-series signals, and performs a pairwise differencing operation between actual values and the estimated values for the present time-series signals to produce residuals. The system then performs a sequential probability ratio test on the residuals to produce alarms having associated tripping frequencies (TFs). Next, the system uses a logistic-regression model to compute a risk index for the power transformer based on the TFs. If the risk index exceeds a threshold, the system generates a notification that the power transformer needs to be replaced. The system also periodically updates the logistic-regression model based on the results of periodic dissolved gas analyses for the transformer to more accurately compute the index for the power transformer.