Power Control Unit Failure Prediction Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for accurate prediction of power control unit failures in hybrid or electric vehicles, especially under multi-load conditions, to prevent significant vehicle failures, particularly in complex systems like autonomous driving.
Innovation Solution
A method and system utilizing a machine learning algorithm, such as the K-Nearest Neighbors algorithm, to predict failures by obtaining data from sensors under simulated and real multi-load conditions, comparing machine learning data to test data, and determining the likelihood of failure based on data clusters and standard deviations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional monitoring methods are used for power control units, then the system complexity remains low, but the prediction accuracy of failures under multi-load conditions is insufficient
Solution Approach 1:
The patent replaces traditional mechanical monitoring methods with a machine learning-based prediction system. The K-Nearest Neighbors algorithm processes sensor data to predict failures, substituting complex computational methods for simple threshold-based monitoring, thereby improving prediction accuracy while managing system complexity through software-based solutions.
Solution Approach 2:
The patent introduces multiple sensors as intermediaries between the power control unit and the monitoring system. These sensors (temperature, current, voltage, frequency) act as mediators that collect detailed operational data, enabling more accurate failure prediction without directly increasing the complexity of the power control unit itself.
2Reliability
If simulated multi-load conditions are used for training, then the machine learning model can be developed offline, but the model may not accurately reflect real-world operating conditions
Solution Approach 1:
The patent applies preliminary action by training the machine learning model offline using simulated multi-load conditions before deployment. This allows the K-Nearest Neighbors algorithm to learn from diverse operational scenarios in advance, improving model reliability without requiring real-time data collection during actual vehicle operation.
Solution Approach 2:
The patent uses parameter changes by varying multiple operational parameters simultaneously in simulated multi-load conditions (temperature, current, voltage, frequency combinations). This creates comprehensive training data that covers diverse operating scenarios, enabling the model to generalize better to real-world conditions while maintaining reliability.
3Measurement precision
If real-time data collection from multiple sensors is implemented, then the prediction capability is enhanced, but the data processing complexity and computational requirements increase
Solution Approach 1:
The patent replaces complex real-time data processing requirements with the K-Nearest Neighbors algorithm, which efficiently handles multi-dimensional sensor data. The algorithm substitutes complex computational processing with a simpler distance-based classification approach, maintaining high detection precision while reducing processing complexity.
Solution Approach 2:
The patent implements a universal machine learning model that handles multiple sensor types (temperature, current, voltage, frequency) and various load conditions through a single K-Nearest Neighbors algorithm. This multi-functional approach simplifies the data processing system by using one algorithm to handle diverse data inputs, reducing overall system complexity.
Data Source
AI summary
A method for predicting a failure of a power control unit of a vehicle is provided. The method includes obtaining data from a plurality of sensors of the power control unit of a vehicle subject to simulated multi-load conditions, implementing a machine learning algorithm on the data to obtain machine learning data, obtaining new data from the plurality of sensors of power control unit of the vehicle subject to real multi-load conditions, implementing the machine learning algorithm on the new data to obtain test data, predicting a failure of the power control unit based on a comparison between the test data and the machine learning data.