Rotational Frequency Measurement Using Variable Sliding Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for measuring rotational frequency in rotating machines face challenges in achieving fast and accurate measurements with low latency, particularly in capturing changes in rotational frequency and maintaining accuracy across varying speed conditions.
Innovation Solution
The implementation of a rotational frequency measurement system using variable sliding windows, where the size of the measurement window can be dynamically adjusted, and multiple windows can be used in parallel to quickly capture changes in rotational frequency, allowing for low latency and high accuracy measurements by utilizing a counter, memory, and frequency processor to determine rotational frequency based on count differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a traditional counter captures elapsed time between pulse signals to determine rotational frequency, then the measurement system is simple to implement, but the response time is slow and latency is high
Solution Approach 1:
The system pre-captures and stores multiple count values from the counter at different time points before they are needed for frequency calculation. By having these count values ready in advance in memory, the system eliminates the need to wait for new pulse signals to arrive before computing frequency changes, thus reducing measurement latency and improving response speed.
Solution Approach 2:
The measurement process is divided into discrete segments by storing individual count values from the counter at specific time intervals. Each count value represents a segmented snapshot of the rotational position, allowing the system to independently calculate frequency changes between any two segments without processing the entire measurement sequence, thereby accelerating response time.
2Measurement precision
If a fixed measurement window is used to determine rotational frequency, then the calculation is simple, but the system cannot quickly capture changes in rotational frequency
Solution Approach 1:
The measurement window is made dynamic by allowing it to slide across the stored count values in memory. Instead of being fixed, the window can move to compare different pairs of count values, enabling the system to adaptively track changes in rotational frequency at different time points and detect frequency changes more quickly while maintaining measurement accuracy.
Solution Approach 2:
The system uses the stored count values as feedback to continuously monitor and detect changes in rotational frequency. By comparing current count values with previously stored values, the system provides continuous feedback on frequency changes, allowing for rapid detection and response to varying rotational speeds while maintaining precise measurements.
3Loss of time
If multiple measurement windows are used in parallel to capture frequency changes quickly, then response time improves, but system complexity increases
Solution Approach 1:
Multiple measurement windows are merged into a single unified memory structure that stores all count values sequentially. Instead of implementing separate parallel measurement systems, the invention combines multiple measurement functions into one integrated memory and processing architecture, reducing system complexity while still enabling parallel frequency measurements through the sliding window mechanism.
Solution Approach 2:
The single memory structure serves multiple functions: it stores count values, provides data for multiple sliding windows, and supports both historical analysis and real-time frequency measurement. This multi-functional design eliminates the need for separate dedicated memory for each measurement window, simplifying the overall system while maintaining the ability to perform parallel measurements.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Systems and methods for measuring rotational frequency in rotating machines use variable sliding windows of measurement. The systems and methods count and store the number of internal clock cycles between the start of a measurement interval and each pulse signal from a pulse generator. Rotational frequency is determined by taking a difference between the count for a most recent pulse signal and the count for some previous pulse signal within the measurement interval. The number of pulse signals that have occurred between the most-recent pulse signal and the previous pulse signal represents a window of measurement. This window of measurement, or the size thereof, may then be used along with the count difference to determine the rotational frequency. The window of measurement may then be slid to the counts for next most recent pulse signal and the next previous pulse signal to obtain a new count difference, and so on. The pulses and the associated timestamp values may be written into a circular memory. The size of the measuring window may be adapted according to the current speed. Optionally, a frequency divider may process the sensor signal and pass only every N-th pulse, so that less data have to be stored. N may be optimized to take into account various system parameters.