Multiphase Inverter Fault Identification Using Switched Current Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying faults in power converter inverter circuits and current sensors are inefficient, requiring excessive time and risking large currents due to through currents exceeding rated values, and often lead to erroneous determinations.
Innovation Solution
A power converter with a multiphase inverter circuit, current sensors, and a fault detection system that uses specific switching patterns and voltage/current monitoring to accurately identify faults in switching elements and current sensors in a short time, preventing large currents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a switching pattern for turning on switching elements one by one is used to identify fault location, then fault location can be identified, but a large through current exceeding rated current may occur causing heat generation and further faults
Solution Approach 1:
The patent applies preliminary action by first detecting abnormality during normal operation before initiating the fault identification switching pattern. The abnormality detection unit monitors current values and determines if a fault exists before the switching elements are turned on one by one, ensuring that the identification process only starts when necessary and with proper precautions in place.
Solution Approach 2:
The patent introduces an intermediary control mechanism where the control unit manages the switching pattern based on abnormality detection results. The control unit acts as an intermediary that coordinates between the abnormality detection unit and the switching elements, ensuring that switching patterns are applied only after confirming abnormality and with appropriate current monitoring to prevent excessive through current.
2Measurement precision
If waiting for motor rotation speed to become lower than specified value before fault identification, then erroneous determination due to back electromotive force is prevented, but identification time becomes excessively long
Solution Approach 1:
The patent applies preliminary action by performing abnormality detection during normal motor operation before the fault identification process begins. The system continuously monitors current values and determines abnormality in advance, so when fault identification is initiated, the decision can be made quickly without waiting for motor speed to change, thus reducing identification time while maintaining accuracy.
Solution Approach 2:
The patent applies dynamics by adapting the fault identification approach based on motor operating conditions. The control unit dynamically selects appropriate switching patterns and evaluation methods based on whether the motor is rotating or stationary, and based on the type of abnormality detected, allowing efficient fault identification across different operating states without being constrained by a single fixed procedure.
3Measurement precision
If a switching pattern for driving switching elements at higher frequency than motor rotation speed is used, then fault in current sensor can be identified based on high-frequency component, but fault in switching element cannot be identified and large through current may occur
Solution Approach 1:
The patent applies dynamics by dynamically selecting different switching patterns based on the type of abnormality detected. The control unit adapts the switching strategy - using high-frequency switching patterns when current sensor faults are suspected, and using patterns suitable for switching element fault identification when those are suspected, based on real-time analysis of current values and abnormality detection results.
Solution Approach 2:
The patent applies segmentation by dividing the fault identification process into separate evaluation stages. The control unit evaluates different aspects of the current values separately - analyzing high-frequency components for current sensor faults, analyzing through current characteristics for switching element faults, and making determination based on the specific abnormality type detected, rather than using a single unified approach.
Data Source
AI summary
A power converter includes: a multiphase inverter circuit including multiple switching elements; multiple current sensors each of which detects a current of a respective phase of the inverter circuit; a terminal voltage detector that detects a voltage of each terminal of a load connected to the inverter circuit; a switching element drive circuit that controls the switching elements according to a switching pattern given; a fault detector that detects a fault in the inverter circuit or the current sensors; and a fault identifier that, after the fault detector detects the fault, switches the switching pattern to be given to the switching element drive circuit according to types of faults detectable by the fault detector, and identifies a type and a location of the fault based on the value of the current detected or the value of the voltage detected during the control according to the switching pattern.