Open-End Winding Inverter Control for Switching Failure Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing control devices struggle to accurately identify and distinguish between different types of failures, such as open-circuit and short-circuit failures, in switching elements of inverters connected to open-end windings of rotating electrical machines, particularly when failures occur in one of the two inverters, leading to asymmetrical and distorted current waveforms that complicate failure identification.
Innovation Solution
The solution involves independent control of two inverters connected to the open-end windings, using mixed pulse width modulation to determine failure patterns based on current integrated values' polarities, and active short-circuit control to identify and isolate failure-side arms, allowing continued operation of the rotating electrical machine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensors are disposed in each switching element for failure detection, then failure detection capability is improved, but device cost and complexity increase
Solution Approach 1:
The system uses existing control parameters (phase currents and voltages) that are already being measured for control purposes to detect switching element failures. The control device self-diagnoses failures by analyzing these parameters without requiring additional sensors in each switching element, thus achieving reliable failure detection while avoiding increased device complexity
Solution Approach 2:
The existing current and voltage measurement systems, originally designed for control purposes, are made to serve dual functions: both control operation and failure detection. By analyzing phase currents and voltages, the system simultaneously achieves control and diagnostic functions, eliminating the need for separate sensing infrastructure
2Device complexity
If phase currents and voltages are used to detect failures, then device complexity is reduced, but failure identification precision deteriorates due to influence from both inverters
Solution Approach 1:
The system segments the failure analysis by temporarily isolating one inverter at a time. By controlling one inverter while keeping the other inactive, the system can independently analyze phase currents and voltages to identify failures in the active inverter, eliminating the masking effect caused by simultaneous operation of both inverters
Solution Approach 2:
The system employs periodic switching between the two inverters, activating one inverter at a time in alternating periods. This periodic operation allows systematic testing and failure identification for each inverter independently, while maintaining overall system functionality through time-division operation
3Measurement precision
If mixed pulse width modulation control is used, then failure pattern identification capability is improved, but control complexity increases
Solution Approach 1:
The system changes control parameters by switching between different pulse width modulation modes (mixed PWM with different pulse patterns, and inactive states). By varying the modulation depth, pulse timing, and inverter activation patterns, the system creates distinct electrical signatures for different failure patterns, enabling precise failure identification through parameter comparison
Data Source
AI summary
In a first control state, it is determined which one of a first failure pattern (FP1) and a second failure pattern (FP2) is a failure pattern (FT), and in a second control state, it is determined which one of a first lower-stage-side failure pattern (LF1) and a second lower-stage-side failure pattern (LF2) is a lower-stage-side failure pattern (LF), and it is determined which one of a set of upper-stage-side arms of a first inverter (11), a set of lower-stage-side arms of the first inverter (12), a set of upper-stage-side arms of a second inverter (12), and a set of lower-stage-side arms of the second inverter (12) is failure-side arms, based on a result of the determination in the first control state and a result of the determination in the second control state.