Electric Rotating Machine Simulation from Minimal Input Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for creating a simulation model of electric rotating machines, especially asynchronous motors and generators, face challenges in determining operating behavior due to inadequate information, often requiring expensive measurements, which is impractical for older machines with limited data availability.
Innovation Solution
A computer-implemented method using a trained function to determine simulation model parameter values based on input data such as shaft height and number of pole pairs, allowing for the creation of a simulation model without expensive measurements, utilizing a neural network or multivariate regression for accurate parameter determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If expensive measurements (cold resistance measurement, electromagnetic series, load series) are conducted to create an accurate simulation model, then the manufacturing precision and reliability of the simulation model improve, but the cost and complexity of the process increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-training a neural network function using comprehensive measurement data from multiple machines before actual simulation model creation. The neural network is trained offline with data from cold resistance measurements, electromagnetic series, and load series, storing the learned relationships in a trained function. When creating a simulation model, instead of performing expensive measurements, the system simply inputs basic machine data (shaft height, number of pole pairs) into the pre-trained neural network, which instantly provides accurate parameter values. This transfers the measurement complexity to a one-time training phase rather than requiring it for each individual machine.
Solution Approach 2:
The patent uses copying by creating a virtual model (neural network) that replicates the behavior and relationships learned from expensive physical measurements. The trained function serves as a digital copy of the measurement process, capturing the complex relationships between machine geometry, electrical parameters, and performance characteristics. Once trained, this digital copy can be reused indefinitely without repeating the actual measurements, effectively copying the benefits of comprehensive measurements at minimal cost.
2Reliability
If comprehensive measurement data is collected to determine operating behavior, then the reliability of the simulation model improves, but the time and resources required increase
Solution Approach 1:
The patent performs preliminary action by conducting all complex measurements and data collection during the offline training phase of the neural network. The system gathers comprehensive measurement data from multiple machines, processes it through the neural network training algorithm, and stores the learned relationships in the trained function. During actual simulation model creation, only simple input data (shaft height, number of pole pairs) needs to be provided, and the neural network instantly outputs reliable parameter values. This shifts the time investment from individual machine processing to a one-time training process.
3Ease of operation
If simple methods (linear interpolation) are used to determine operating behavior, then the ease of operation improves, but the manufacturing precision and detail of specifications deteriorate
Solution Approach 1:
The patent replaces simple mechanical interpolation methods with an intelligent system based on neural networks. Instead of using basic linear interpolation algorithms that provide only torque information, the system uses a trained neural network that has learned complex relationships from comprehensive measurement data. This substitution maintains ease of operation (simple data input) while dramatically improving manufacturing precision (detailed specifications including efficiency and power factor). The neural network acts as an intelligent intermediary that combines the simplicity of basic methods with the accuracy of comprehensive measurements.
Data Source
AI summary
A computer-implemented method for providing a simulation model of an electric rotating machine is disclosed. The simulation model is defined by parameter values. Input data is obtained. The input data is collectable using the electric rotating machine when the electric rotating machine is not connected to an operating voltage and being characteristic of the electric rotating machine. The parameter values are determined from the input data using a trained function and the parameter values determined are provided.