Electric Motor Current-Spectrum Diagnosis Across Load Variation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electric motor diagnosis methods require multiple sensors, increasing equipment size and cost, making them cumbersome and expensive.
Innovation Solution
An electric motor diagnosis device that utilizes only current information to diagnose abnormalities, employing normalization, FFT analysis, and correction techniques to enhance accuracy without additional sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors are used to collect current information, voltage information, and zero-phase current information, then diagnosis accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The invention extracts and utilizes only the current information component from the multiple information types (current, voltage, zero-phase current) that were previously required. By focusing solely on current information and applying sophisticated analysis methods (FFT, spectral analysis, feature extraction), the system achieves accurate diagnosis without needing voltage sensors or zero-phase current sensors, thereby reducing device complexity while maintaining diagnostic capability
Solution Approach 2:
The invention transforms the current information through various parameter changes including time-domain features (RMS, peak values, waveforms) and frequency-domain features (spectral components, harmonic analysis). This parameter transformation allows a single current sensor to provide rich diagnostic information that previously required multiple sensors, resolving the contradiction between measurement precision and device complexity
2Measurement precision
If multiple sensors are used to collect current information, voltage information, and zero-phase current information, then diagnosis accuracy is improved, but cost increases
Solution Approach 1:
The invention extracts and utilizes only the current information component from the multiple information types (current, voltage, zero-phase current) that were previously required. By focusing solely on current information and applying sophisticated analysis methods (FFT, spectral analysis, feature extraction), the system achieves accurate diagnosis without needing voltage sensors or zero-phase current sensors, thereby reducing device complexity while maintaining diagnostic capability
Solution Approach 2:
The invention replaces expensive multiple-sensor hardware with a more cost-effective approach using a single current sensor combined with advanced signal processing algorithms. The computational resources required for FFT and spectral analysis are significantly cheaper than the hardware cost of additional sensors, resolving the contradiction between diagnosis accuracy and manufacturing cost
Data Source
Figure 1
Figure 2~3
Figure 4A~4B
AI summary
Obtained is an electric motor diagnosis device (100) that can diagnose an abnormality in an electric motor by obtaining only current information of the electric motor. An electric motor diagnosis device includes: a current input unit (7) for inputting current value data of an electric motor; an FFT analysis unit (116) for performing FFT analysis on the current value data, to obtain power spectrum data; a normalized current calculation unit (110) for calculating normalized current from the current value data; a load rate calculation unit (111) for obtaining a load rate corresponding to the normalized current calculated by the normalized current calculation unit (110), by using a normalized current-load rate curve obtained from set information of the electric motor; a correction value data recording unit (113) in which a correction value for adjusting influence due to a difference in the load rate of the power spectrum data at a time of diagnosis relative to a reference value of the power spectrum data is recorded in a database so as to correspond to the load rate; an FFT analysis result correction value selection unit (112) for selecting the correction value on the basis of the database and the load rate, at the time of diagnosis, obtained by the load rate calculation unit (111); an FFT analysis result correction unit (121) for correcting the power spectrum data at the time of diagnosis by the correction value; and an abnormality diagnosis unit (122) for, with respect to each of the reference value and the corrected power spectrum data at the time of diagnosis, calculating a difference value between current signal intensities at a power supply frequency and at a rotational frequency, and diagnosing the electric motor through comparison of the difference values with each other.