Switching Device Service Life Prediction Using Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the service life of switching devices in vehicles are not precise, as they primarily rely on current variables and lack a comprehensive approach to predict potential failures, leading to potential disadvantageous failures.
Innovation Solution
A neural network is trained using current and switching device state variables, with monitored learning and cloud-based implementation, to accurately determine the remaining service life of switching devices, enabling more precise health assessment and timely replacement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods using only current variables are used to determine service life, then the method is simple to implement, but the determination precision is insufficient
Solution Approach 1:
The patent combines multiple input variables (current variables, switching device state variables, environmental variables) into a unified neural network model to determine service life. This merging of multiple data sources enables more precise determination while managing complexity through integrated processing.
Solution Approach 2:
The patent replaces traditional mechanical/methodological approaches to service life determination with a neural network-based system. This substitution enables higher precision by leveraging machine learning capabilities to process multiple variables and identify complex patterns that traditional methods cannot detect.
2Measurement precision
If a neural network with multiple variables is implemented locally, then the determination precision improves, but the device complexity and computational requirements increase
Solution Approach 1:
The patent extracts the complex neural network computational functionality from the local switching device and places it in a cloud-based environment. This extraction allows the switching device to remain simple while still benefiting from high-precision neural network-based service life determination through cloud connectivity.
Solution Approach 2:
The patent introduces a cloud-based processing layer as an intermediary between the switching device and the service life determination function. This intermediary handles the complex computational requirements, allowing the edge device to maintain simplicity while achieving high precision through the cloud-based neural network.
3Reliability
If comprehensive monitoring of current and state variables is performed, then the reliability of service life prediction improves, but the loss of information and data processing requirements increase
Solution Approach 1:
The patent performs preliminary actions by collecting and storing time-series data of current variables and switching device state variables before service life determination is needed. This preliminary data collection and organization enables reliable predictions when needed while managing information load through structured storage and preprocessing.
Data Source
AI summary
A method for determining the service life of a switching device, comprising the steps of:a) providing a neural network having at least two input variables and an output variable;b) determining at least a current variable, which represents a current flowing through the switching device, and a switching device state variable, which represents a sticking or jammed or fused switching device;c) inputting at least the current variable and the switching device state variable as input variables into the neural network;d) determining a remaining service life of the switching device by means of the neural network.A method for training a neural network for determining the service life of a switching device, a corresponding device for determining the service life, a corresponding computer program, and a machine-readable storage medium with the computer program.