Adaptive Window FIR Filter for Transient Signal Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Protective relays in electric power systems face challenges in detecting short circuits and switching events due to transients in voltage and current signals, which can lead to delayed tripping, false tripping, or failure to trip, as existing filtering methods introduce delays that compromise speed and accuracy.
Innovation Solution
Implementing a band-pass finite impulse response (FIR) filter with adaptive window resizing, where the window length changes in response to disturbances, allowing for faster operation by excluding pre-disturbance signals and gradually increasing accuracy as the window grows, without requiring frequency tracking or constant sampling rate adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a fixed window length filter is used, then the filter provides stable performance under normal conditions, but it introduces delays and filtering errors during transient conditions
Solution Approach 1:
The filter window length is made dynamic rather than fixed. The system automatically adjusts the window length based on system conditions: using a first window length during normal operation and a second, different window length during transient conditions. This dynamic adaptation allows the filter to optimize between stability and speed depending on the operational state, resolving the contradiction between reliable stable performance and fast transient response.
2Speed
If the filter window length is reduced to improve speed, then the response time decreases, but the filtering accuracy deteriorates
Solution Approach 1:
The window length is dynamically adjusted based on system conditions. During transient conditions requiring fast response, a shorter window length is used to improve speed. During normal conditions where accuracy is paramount, a longer window length is used to enhance filtering precision. This dynamic switching resolves the contradiction by allowing the system to optimize for speed or accuracy depending on the operational context.
Solution Approach 2:
The filter parameter (window length) is changed based on system state. The system transitions between different window length parameters: a first window length for normal operation and a second window length for transient conditions. This parameter adaptation allows the filter to achieve both fast response during transients and high accuracy during normal operation, resolving the speed-accuracy tradeoff.
3Stability of the object's composition
If existing filtering methods are applied to mitigate transient impact, then the signal stability improves, but the tripping speed and protection reliability deteriorate
Solution Approach 1:
The filter dynamically adapts its window length based on whether the system is in a normal or transient state. During transients, the system uses a window length optimized for fast response to maintain protection speed. During normal operation, it uses a window length optimized for signal stability. This dynamic behavior resolves the contradiction between signal stability and protection speed by applying the appropriate filter characteristics for each operational phase.
Data Source
AI summary
Variable window filtered power system signals for electric power system monitoring and protection operations of an electric power system are provided herein. Upon detection of a power system disturbance, the filter window is decreased after a predetermined resize delay such that pre-disturbance samples are not included in the new window. As additional samples are obtained, the filter window grows to include new samples until the window reaches an initial filter window length. Gain and group delay correction factors accounting for window size and signal frequency are approximated.