Power Module Fault Data Recording via Segmented Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-power power electronic systems face challenges in recording and analyzing fault data due to limited storage capacity, low sampling rates, and the inability to observe waveform changes during faults, leading to difficulties in diagnosing the root cause of issues in power modules.
Innovation Solution
A recording method and apparatus for power modules that acquire and store real-time parameters, including fault information, before and after fault occurrences, using a combination of SRAM and flash memory, allowing for the generation of waveforms for recent faults, enabling detailed analysis of fault causes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If the storage capacity is increased to store more fault data and parameters, then the recording time and data completeness improve, but the cost increases and data reading time increases
Solution Approach 1:
The storage system is segmented into two distinct parts: a first storage unit (circular buffer) for continuous real-time parameter recording, and a second storage unit (FIFO queue) for storing fault information and associated parameters. This segmentation allows each storage unit to be optimized for its specific function, avoiding the need for a single large-capacity storage device while maintaining data completeness.
Solution Approach 2:
The patent extracts and prioritizes critical fault-related data (fault type, occurrence time, position, and parameters within a preset time window) from the continuous parameter stream, storing only these essential elements in the second storage unit. This extraction approach ensures that the most important fault information is preserved without requiring proportional increases in overall storage capacity.
2Measurement precision
If the parameter sampling rate is increased to capture detailed fault waveforms, then the measurement precision improves, but the storage capacity is exhausted faster and recording time decreases
Solution Approach 1:
The system dynamically adjusts the recording strategy based on operational state: during normal operation, parameters are recorded at a standard sampling rate in the first storage unit; when a fault is detected, the system switches to high-resolution recording mode, capturing detailed parameter variations within a preset time window around the fault event. This dynamic approach ensures high measurement precision during critical fault moments while maintaining reasonable recording duration during normal operation.
Solution Approach 2:
The system preliminarily records continuous parameters at a lower sampling rate in the circular buffer before faults occur. When a fault is detected, these pre-recorded parameters are immediately preserved in the FIFO queue along with fault information, ensuring that detailed waveform data is captured without requiring continuously high sampling rates, thus balancing precision and recording duration.
3Device complexity
If only fault type information is stored to simplify the storage system, then the device complexity decreases, but the ability to analyze root causes deteriorates
Solution Approach 1:
The storage system applies local quality by storing different types of information with different levels of detail in different storage units. The first storage unit stores continuous parameters at standard resolution, while the second storage unit stores detailed fault-related parameters (including timing information and parameters within a preset window around fault occurrence) at high resolution. This localized enhancement ensures comprehensive fault analysis capability without uniformly increasing complexity across the entire storage system.
4Measurement precision
If the system waits for fault reproduction to analyze faults, then the measurement conditions are controlled, but the time required for diagnosis increases and productivity decreases
Solution Approach 1:
The system performs preliminary recording of all necessary parameters and fault information in real-time during normal operation and fault events. By pre-capturing and preserving complete fault data (including timing, position, and parameter waveforms) in the FIFO queue, the system eliminates the need for time-consuming fault reproduction during diagnosis, allowing maintainers to directly analyze the stored data and significantly improving diagnosis speed while maintaining measurement accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure discloses a recording method and a recording apparatus of a power module. The method includes: step 1: acquiring (S110) a plurality of operating parameters of the power module through an acquiring unit (410); step 2: reading and processing (S120) the operating parameters through a processor (420), to obtain a plurality of real-time parameters of the power module; step 3: storing (S130) the real-time parameters in a first storage unit (430) after the power module starts operating; and step 4: when the processor (420) determines that an abnormality occurs in the power module, storing (S150) corresponding fault information and the real-time parameters of the power module in a corresponding storage address of a second storage unit (440) obtained according to a preset strategy. Through a recording apparatus built in the power module, the present disclosure may realize the real-time recording function and may have good scalability.