Electric Drive Thermal Fault Prediction Using Two-Layer State Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting temperature faults in electric drives in industrial networks are limited by the availability of measurement data and rely on limited input parameters, making it difficult to accurately monitor the condition and prevent thermal failures.
Innovation Solution
A method using a network device to estimate temperatures of electric drive components by obtaining input and state parameters, employing a two-layer state space model to predict temperature values, and determining the condition based on these estimates, allowing for proactive maintenance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If prior art techniques use limited input parameters (switching frequency and current) for temperature fault prediction, then the prediction process is simple, but the prediction accuracy is insufficient because temperature depends on various input and state parameters
Solution Approach 1:
The prediction model is segmented into two distinct layers: a first layer that estimates state parameters from input parameters and measured temperatures, and a second layer that predicts future temperatures using the estimated state parameters. This segmentation allows the system to handle multiple parameters systematically while maintaining manageable complexity in each layer.
Solution Approach 2:
The invention transitions from using only directly measurable parameters to incorporating estimated state parameters as an additional dimension. By introducing state parameters that represent internal drive conditions not directly measurable, the model gains a more comprehensive view of temperature dependencies without requiring direct measurement of all influencing factors.
2Reliability
If the system collects and processes multiple input and state parameters for temperature prediction, then the prediction accuracy improves, but the data processing complexity and computational requirements increase
Solution Approach 1:
The first layer of the model performs preliminary estimation of state parameters before the temperature prediction in the second layer. This preliminary action prepares the necessary intermediate values that simplify the final temperature prediction process, allowing the system to handle multiple parameters in a structured sequence rather than simultaneously.
Solution Approach 2:
The state parameters act as intermediaries between the directly measurable input parameters and the target temperature values. These intermediate state parameters bridge the gap between what can be easily measured and what needs to be predicted, enabling accurate temperature forecasting without requiring direct measurement or simultaneous processing of all influencing factors.
3Measurement precision
If the system uses a comprehensive model considering multiple parameters for temperature prediction, then the ability to predict temperature faults improves, but the difficulty of implementing and maintaining the monitoring system increases
Solution Approach 1:
The monitoring system is divided into modular functional layers that can be independently implemented and maintained. The first layer handles state parameter estimation while the second layer handles temperature prediction, allowing separate optimization and maintenance of each function without affecting the other.
Solution Approach 2:
The system uses readily available data from the electric drive's control unit (input parameters and measured temperatures) to automatically estimate state parameters and predict future temperatures. This self-service approach eliminates the need for additional sensors or manual measurements, reducing implementation and maintenance complexity while maintaining high prediction precision.
Data Source
AI summary
The present disclosure relates to monitoring a condition of an electric drive in an industrial network. A method comprises obtaining values of input parameters, state parameters, and one or more temperatures associated with one or more components, at a first time instant. The method further comprises estimating values of the state parameters at a second time instant with a first layer of a state space model of the electric drive and the values of input parameters at the first time instant. In addition, the method comprises estimating values of the one or more temperatures at a third time instant with a second layer of the model, the values estimated for the state variables and the values of the temperatures at the first time instant. A condition of the electric drive is determined from the values of the temperatures estimated for the third time instant and one or more predetermined thresholds.