Adaptive Unit Space Generation for Plant Load Change Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing plant monitoring systems face challenges in accurately determining the operation state due to increased Mahalanobis distance values during consumable component replacement and load changes, leading to mistaken abnormality detections and the need for frequent unit space updates, which complicates and costs monitoring work.
Innovation Solution
A unit space generating device that acquires sampling data at fixed cycles, determines adoption probability based on change rates and amounts of output values, and generates unit spaces to reduce bias and mistaken detections, allowing for robust Mahalanobis distance calculations and reduced unit space variations across load states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If data is sampled at a fixed timing when generating a unit space, then the unit space generation process is simple, but a bias is likely to occur in data configuring the unit space, leading to hyper-sensitive response to state quantity changes and mistaken abnormality detection
Solution Approach 1:
The patent applies dynamics by making the sampling timing adaptive rather than fixed. The sampling timing changes based on the degree of change in output values, allowing the system to adjust to different operational conditions. This resolves the contradiction by maintaining operational simplicity while improving detection accuracy through dynamic adaptation to actual plant conditions.
Solution Approach 2:
The patent changes the parameter of sampling timing from a fixed value to a variable that depends on output value changes. By making the sampling interval dynamic based on the degree of change in output values, the system achieves both ease of operation and high measurement precision, avoiding mistaken abnormality detection while maintaining simple implementation.
2Measurement precision
If a different unit space is generated for each load band to reduce mistaken detection, then the abnormality detection accuracy improves, but the work of generating and updating multiple unit spaces becomes complicated, increasing monitoring costs
Solution Approach 1:
The patent creates a single unit space that serves multiple load bands through adaptive sampling. Instead of generating separate unit spaces for each load condition, the system uses one unified unit space that adapts to different operational states by adjusting sampling timing based on output value changes. This reduces management complexity while maintaining high detection accuracy across all load conditions.
Solution Approach 2:
The system performs self-adjustment by automatically modifying sampling timing based on observed output value changes. The unit space generation process adapts to different load conditions without requiring manual intervention to create separate unit spaces for each band. This self-service mechanism reduces the complexity of managing multiple unit spaces while maintaining high detection accuracy.
3Measurement precision
If sampling frequency is increased to capture load changes, then the detection of load changes improves, but the data processing load and unit space generation complexity increase
Solution Approach 1:
The patent applies dynamics by adjusting the sampling frequency based on the degree of change in output values. During periods of significant load change, the sampling frequency increases automatically to capture important transitions. During stable operation, the sampling frequency decreases to reduce data processing load. This dynamic adjustment maintains high detection capability while improving overall processing efficiency.
Solution Approach 2:
The patent applies partial action by increasing sampling frequency only when necessary - specifically when the degree of change in output values indicates significant load changes. Rather than continuously sampling at high frequency, the system applies enhanced sampling selectively during critical transitions, maintaining detection accuracy while reducing overall data processing requirements.
Data Source
AI summary
A unit space generating device (141) which generates a unit space for use when diagnosing an operating state of a plant on the basis of a Mahalanobis distance is provided with: a sampling data acquiring unit (141A) which acquires a sampling data group comprising a plurality of state quantities of the plant, measured with a fixed period; an adoption determining unit (141C) which, on the basis of an adoption probability calculated each time the sampling data group is acquired, determines whether the sampling data group is to be adopted as a unit space generation data group serving as the basis for a unit space; a unit space generating unit (141D) which generates unit spaces on the basis of a plurality of the adopted unit space generation data groups; and an output value acquiring unit (141B) which acquires an output value of the plant corresponding to the sampling data group.