Condition Monitoring Signal Windowing for Mobile Machine Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Condition monitoring apparatuses face challenges in maintaining accuracy of anomaly detection while reducing the processing amount of sensor signals, particularly in mechanical devices with varying operating conditions such as acceleration, deceleration, and door operations.
Innovation Solution
A condition monitoring apparatus that includes a processing circuitry to collect sensor signals, determine speed and acceleration conditions, and adjust analysis length and segment detection based on these conditions to cut out sensor signals at specific time widths, using unsupervised machine learning for anomaly detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor signals are continuously monitored with fixed analysis length, then anomaly detection accuracy is maintained, but processing amount increases
Solution Approach 1:
The analysis length is made dynamic by adjusting it according to the operating condition (speed and acceleration). During acceleration and deceleration phases, a shorter analysis length is used, while during constant speed operation, a longer analysis length can be applied. This dynamic adaptation reduces the total processing amount while maintaining detection accuracy for transient anomalies.
Solution Approach 2:
The patent changes the parameter of analysis length based on operating conditions (speed and acceleration). By monitoring these parameters and adjusting the analysis window accordingly, the system optimizes the balance between detection accuracy and processing load, using shorter windows during high-dynamic phases where anomalies are more likely to occur.
2Productivity
If analysis length is reduced to decrease processing amount, then computational load is reduced, but anomaly detection accuracy deteriorates
Solution Approach 1:
Different analysis lengths are applied to different operating conditions. During acceleration and deceleration, shorter analysis lengths are used locally, while during constant speed operation, longer analysis lengths are used. This local differentiation ensures adequate detection accuracy is maintained where needed while reducing processing load in stable operating phases.
Solution Approach 2:
The system performs preliminary detection of speed and acceleration conditions to determine the appropriate analysis length before processing the sensor signals. This preliminary action allows the system to pre-adjust the processing parameters based on the current operating state, ensuring optimal detection accuracy is maintained while minimizing unnecessary processing.
3Device complexity
If fixed segment detection is used, then processing is simplified, but adaptability to different operating conditions deteriorates
Solution Approach 1:
The segment detection process is made dynamic by adjusting the analysis length according to detected operating conditions (speed and acceleration). The system continuously monitors these conditions and adapts the signal processing parameters accordingly, enabling it to handle various operating phases effectively without requiring completely different processing algorithms for each condition.
Data Source
AI summary
According to one embodiment, a condition monitoring apparatus includes a processing circuitry. The processing circuitry is configured to collect a sensor signal output from a sensor that monitors a condition of a mechanical device that is at least partially mobile. The processing circuitry is configured to diagnose a presence or absence of an anomaly in the mechanical device based on the sensor signal. The processing circuitry is configured to cut out the sensor signal in a time width according to any one or more of a speed, an acceleration, and a jerk of the mechanical device. The processing circuitry is configured to determine the presence or absence of an anomaly based on the cut out sensor signal.