EPS Motor Phase Failure Diagnosis Using Negative Sequence Current
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing phase failure diagnosis methods in electric power steering systems fail to detect circuit breaks caused by phase separation MOSFETs, posing a safety risk due to undetected failures during motor operation.
Innovation Solution
A phase failure diagnosis method based on current instruction reconstruction and negative sequence current extraction, utilizing formulas to calculate reference current values and negative sequence components, and comparing these to feedback currents to accurately identify phase failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If power-on self-inspection is used for phase failure diagnosis, then the diagnosis can be performed during initialization, but the failure cannot be detected during motor operation
Solution Approach 1:
The patent performs preliminary reconstruction of current instructions and calculation of negative sequence components during the power-on self-inspection phase. This preliminary action establishes a baseline reference state that enables continuous monitoring capability during subsequent motor operation, extending detection coverage beyond just initialization.
Solution Approach 2:
The patent implements continuous feedback monitoring by comparing real-time three-phase current measurements against the reconstructed current instructions. The negative sequence current components are continuously calculated and compared during operation, creating a closed-loop feedback system that detects failures dynamically throughout the motor's operational lifecycle.
2Measurement precision
If conventional current monitoring is used, then the system can monitor phase currents, but it cannot identify circuit breaks in phase separation MOSFETs
Solution Approach 1:
The patent introduces negative sequence current components as an intermediary diagnostic parameter. By transforming the three-phase current measurements into negative sequence components and comparing them against reconstructed reference values, the system creates an intermediate representation that makes MOSFET circuit break failures detectable, bridging the gap between conventional current monitoring and failure identification.
Solution Approach 2:
The patent changes the measurement parameter from raw three-phase currents to negative sequence current components. This parameter transformation enables the detection system to identify MOSFET failures that are invisible in the original current domain, effectively changing the diagnostic parameter space to reveal previously undetectable failure modes.
3Device complexity
If no phase failure diagnosis method is implemented, then the system complexity remains low, but safety risk increases due to undetected failures
Solution Approach 1:
The patent implements a self-service diagnostic approach where the motor control system uses its existing current sensors and control structure to perform phase failure detection. By reconstructing current instructions from available control signals and using the existing current measurements, the system achieves failure diagnosis without requiring additional external diagnostic hardware, keeping complexity low while improving safety.
Data Source
Figure 1
Figure 2
AI summary
The present invention provides a phase failure diagnosis method based on current instruction reconstruction and negative sequence current extraction. The method comprises the following steps: calculating a d-axis current instruction value and a q-axis current instruction value of a motor according to the current torque, the rotating speed and the voltage of the motor; performing inverse Clark transformation and inverse Park transformation on the d-axis current instruction value and the q-axis current instruction value of the motor, to obtain a U-phase current reference value, a V-phase current reference value and a W-phase current reference value; calculating a reference negative sequence component of the U-phase current according to the U-phase current reference value, the V-phase current reference value and the W-phase current reference value; calculating a negative sequence component of the U-phase current according to the three-phase current; and calculating the difference between the reference negative sequence component of the U-phase current and the negative sequence component of the U-phase current, and performing phase failure diagnosis according to the difference between the reference negative sequence component and the negative sequence component. Therefore, phase failure diagnosis based on current instruction reconstruction and negative sequence current extraction is realized.