Sensor Measurement Filtering for Reliable Turbine Operation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Technical systems like gas or wind turbines face reliability issues due to unreliable sensor measurements under extreme conditions, leading to inaccurate assessment of their operating status and potential operational disruptions.
Innovation Solution
A method using a program-controlled device to categorize sensor measurements as normal or anomalous through threshold comparisons and statistical analysis, including location parameters and moving windows, to identify and filter out erroneous data, thereby improving measurement reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If sensor measurements are used under extreme conditions (high temperatures, pressures, flow rates), then the technical system can operate in demanding environments, but the error rates in sensor measurements increase leading to unreliable data
Solution Approach 1:
The patent introduces an intermediary data processing system that acts as a mediator between the sensors and the control system. This intermediary layer applies statistical methods (median calculation, deviation analysis) to filter and validate sensor measurements, thereby maintaining measurement reliability even when sensors operate under extreme conditions where error rates increase.
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors measurement values, compares them against statistical thresholds and median values, and adjusts or flags anomalous readings. This feedback loop ensures that unreliable measurements under extreme conditions are identified and handled appropriately, maintaining overall system reliability.
2Reliability
If statistical analysis with multiple location parameters and moving windows is applied to filter anomalous measurements, then measurement reliability improves, but computational complexity and processing time increase
Solution Approach 1:
The patent segments the data processing into distinct stages: acquiring measurement values, calculating median values, determining deviation values, comparing against thresholds, and handling anomalous readings. This segmentation of the statistical analysis process makes the complex computation more manageable and implementable in real-time control systems.
Solution Approach 2:
The patent applies partial statistical analysis by focusing on key metrics (median value, deviation from median) rather than comprehensive statistical treatment. The moving window approach processes only recent measurements rather than entire historical datasets, reducing computational burden while maintaining reliability improvements.
3Measurement precision
If anomalous measurements are filtered out using threshold comparisons and statistical methods, then operational decisions become more accurate, but the processing time for data analysis increases
Solution Approach 1:
The patent performs preliminary actions by pre-calculating median values and establishing threshold criteria before making operational decisions. The system prepares statistical benchmarks in advance, so when measurements need evaluation, the comparison process is accelerated because the reference criteria are already determined.
Solution Approach 2:
The patent uses dynamic moving windows that adapt to changing conditions, adjusting the time period and number of measurements used for statistical calculations. This dynamic approach allows the system to maintain assessment accuracy while optimizing processing time based on the current operational context and rate of change.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method for providing measured values for a technical installation (2) is proposed in which measured values (7) of at least one measurement series are captured, wherein a respective measured value is provided by a measurement sensor (MS1-MSm) for a respective physical measured variable in the technical installation (2) for a respective measurement time, and the measured values are categorised as normal measured values (10) or anomalous measured values (11) by means of a threshold value comparison and at least one further method step, the further method step comprising computation of one or more statistical position parameters for selected measured values from one and the same and/or different measurement series. The method can increase the reliability of the measured values provided. Moreover, a technical system (1) having the technical installation (2), having at least one measurement sensor (MS1-MSm) and having a program-controlled device (4) and a method for operating the technical system (1) are proposed.