3-Phase DAB Fault Diagnosis Using Centroid Vector Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Three-phase dual active bridge (3p-DAB) converters face challenges in detecting and identifying open-circuit faults within their complex semiconductor device configurations, leading to increased losses and potential catastrophic failures due to magnetic saturation, which existing technologies fail to address effectively in real-time.
Innovation Solution
A real-time vector-based fault diagnosis scheme that senses phase currents on one side of the transformer, calculates a centroid vector in the alpha-beta stationary reference frame, and adjusts operations based on the vector's location to identify faults on either side of the transformer, allowing for early detection and prevention of catastrophic failures without requiring additional circuit modifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fault detection methods are used in 3p-DAB converters, then the system can operate with standard monitoring, but fault detection precision is insufficient leading to delayed identification of open-circuit faults
Solution Approach 1:
The patent transforms the fault detection problem from time-domain analysis to geometric space analysis by mapping centroid vectors onto a complex plane divided into 12 sectors. This dimensional transformation enables precise fault identification within 3-4 switching cycles by determining which sector contains the centroid vector, rather than relying on gradual threshold-based detection in the time domain.
Solution Approach 2:
The patent replaces traditional complex signal processing and mechanical threshold comparison systems with a geometric vector analysis method. By calculating the centroid vector of phase currents and determining its angular position in the complex plane, the system achieves rapid fault identification without requiring complex computational algorithms or multiple sensing points.
2Reliability
If current sensors are installed on both sides of the transformer with high gain, then measurement coverage is complete, but system cost and complexity increase significantly
Solution Approach 1:
The patent extracts only the essential measurement requirement by installing current sensors on a single side of the transformer (either primary or secondary). The centroid vector calculation method processes the three-phase currents from one side to identify faults on both sides, eliminating the need for redundant sensors on both sides while maintaining reliable fault detection capability.
Solution Approach 2:
The patent creates a universal fault detection system where sensors installed on one side of the transformer serve dual purposes: detecting faults on the installation side and detecting faults on the opposite side through transformer coupling relationships. This multi-functional approach reduces sensor quantity and system complexity while maintaining comprehensive monitoring coverage.
3Measurement precision
If high-bandwidth current sensors are used on both sides of the transformer, then measurement accuracy is sufficient, but system cost increases due to requiring high-gain sensors
Solution Approach 1:
The patent replaces expensive high-gain current sensors with lower-cost standard current sensors by implementing the centroid vector calculation method. The mathematical processing of signals from one side provides sufficient measurement precision for detecting faults on both sides, eliminating the need for costly high-gain sensors while maintaining adequate measurement accuracy for safety-critical fault detection.
Data Source
AI summary
Various examples are provided related to fault detection of power converters. In one example, a method includes sensing phase currents of a three-phase transformer; determining a centroid vector in an alpha-beta stationary reference frame based upon averaged values of the sensed phase currents; identifying a fault on a side of the transformer based upon a location of the centroid vector within the alpha-beta stationary reference frame, the location defined by a magnitude and an angle of the centroid vector; and adjusting operation of the power converter in response to the identified fault. In another example, a power converter includes a three-phase transformer connected between primary and secondary side three-phase bridges and control circuitry that can determine a centroid vector based upon sensed phase currents; identify a fault on a side of the three-phase transformer; and adjust operation of the power converter in response to the identified fault.