Predictive Process Variable Determination in Switchgear
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining process variables in technical systems, such as gas pressure in SF6 gas-insulated switchgear, face challenges due to measurement errors caused by stochastic disturbances and temperature fluctuations, requiring advanced computational power for accurate long-term and short-term change recognition.
Innovation Solution
A method that records and combines measured values to form weighted values, using linear regression to estimate the future course of the process variable and calculate the point at which it reaches a limit value, reducing computational effort and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all measured values are processed using complex mathematical approximation methods, then prediction accuracy is improved, but computational power requirements increase significantly
Solution Approach 1:
The patent segments the continuous measurement data into discrete measured values at specific time points. This segmentation allows the use of simpler discrete mathematical operations (weighted averaging and linear regression) instead of complex continuous mathematical approximation methods, thereby reducing computational power requirements while maintaining prediction accuracy for process variable trends.
Solution Approach 2:
The patent employs computationally inexpensive discrete mathematical operations (weighted averaging and linear regression) that can be performed with minimal computational resources. These simple calculations serve as effective substitutes for complex mathematical approximation methods, achieving sufficient prediction accuracy with significantly reduced computational power consumption.
2Measurement precision
If measured values are recorded continuously at short intervals, then short-term changes are detected accurately, but the number of measured values to be evaluated increases
Solution Approach 1:
The patent combines multiple continuously recorded measured values into a single weighted measured value using weighted averaging. This merging process reduces the large number of individual measured values into fewer composite values that retain the essential information about short-term changes, making the data more manageable for evaluation while preserving detection accuracy.
Solution Approach 2:
The patent performs preliminary processing of measured values by calculating weighted averages before the actual prediction analysis. This preliminary action reduces the volume of data that needs to be evaluated in subsequent steps, making the overall process more efficient while maintaining the ability to detect short-term changes accurately.
3Power
If discrete measured values are used for prediction, then computational effort is reduced, but disturbance variables may not be fully compensated
Solution Approach 1:
The patent applies parameter changes by using weighted averaging with specifically chosen weighting factors that give appropriate emphasis to different measured values. This parameter adjustment in the averaging process helps compensate for disturbance variables while maintaining computational simplicity, achieving a balance between reduced computational effort and maintained reliability of disturbance compensation.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The invention relates to a method for predictive determination of a process variable (P) in a technical installation, wherein measured values for the process variable (P) are recorded at predeterminable points in time, wherein temporally successive recorded measured values are combined to form a weighted measured value, wherein a discrete measured value is assigned to each weighted measured value, wherein the future temporal progression of the process variable (P) is estimated by means of a linear regression of the discrete measured values, and wherein a point in time at which the process variable (P) reaches a predeterminable limit value is calculated.