PMSM Current Increment Predictive Control for High-Speed Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing predictive control methods for permanent magnet synchronous motors are inadequate for high-speed operations due to inaccuracies caused by rotor position angle changes and parameter variations, leading to poor control performance and increased prediction errors.
Innovation Solution
A predictive control method that establishes a mathematical expression for stator voltage based on rotor position changes, uses a continuous time domain current model, and calculates optimal voltage increments to improve current prediction accuracy, reducing sensitivity to motor parameter variations and inverter dead-time effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the one-order forward Euler approximation method is used for model predictive control, then the control implementation is simple, but the prediction accuracy deteriorates under high-speed operation due to large rotor position angle changes during one control period
Solution Approach 1:
The patent transitions from a static approximation method (Euler) to a dynamic model that explicitly accounts for rotor position changes during the control period. The new model uses time-varying voltage expressions that incorporate the rotating coordinate transformation, making the prediction model adaptive to high-speed operation conditions where rotor position changes significantly.
Solution Approach 2:
The patent changes the fundamental parameters of the prediction model by deriving new voltage expressions that include time-varying components based on rotor position. Instead of using constant coefficient differential equations, the model now incorporates time-dependent voltage vectors that reflect the actual physical behavior during rotation, thereby improving prediction accuracy without excessive complexity.
2Ease of operation
If conventional predictive control methods are used, then the control algorithm is easy to implement, but control performance deteriorates due to parameter variations and inverter dead-time effects
Solution Approach 1:
The patent introduces a feedback mechanism by using the predicted current to calculate current increment, which is then used to evaluate voltage vector effectiveness. The cost function incorporates feedback from actual system behavior (current changes) to select optimal voltage vectors, thereby compensating for parameter variations and dead-time effects through iterative refinement.
Solution Approach 2:
The patent performs preliminary prediction of current behavior using the derived model before actual voltage application. By predicting the current increment and evaluating it against a cost function, the system pre-assesses the effectiveness of different voltage vectors, allowing selection of the optimal vector before implementation, thus improving reliability through advance planning.
Data Source
AI summary
The present disclosure provides a predictive control method of current increment for a permanent magnet synchronous motor includes: substituting a mathematical expression of a stator voltage during one control period into a continuous time domain current model to obtain a discrete current prediction model and a predicted current at the next time point; obtaining a predicted current increment from a current increment prediction model by subtracting a predictive current at a present time point from a predictive current at a next time point; establishing a cost function according to a preset reference current increment and the predicted current increment; obtaining an optimal voltage increment by minimizing the cost function; superposing the optimal voltage increment on a stator voltage of a present control period to obtain an optimal stator voltage of a next control period for controlling control the permanent magnet synchronous motor.