Dual T-Type SiC Drive Control for Torque and Capacitor Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for predictive torque control and capacitor balancing in silicon-carbide (SiC) based dual T-type drive systems face challenges due to high computational complexity and the need for extensive evaluations, particularly with the large number of switching states in multilevel converters, which increases computational time and complicates capacitor voltage balancing.
Innovation Solution
A method is introduced that splits the evaluation of the cost function into two cascaded steps, first optimizing torque and flux using a reduced set of voltage vectors and then optimizing capacitor voltage balancing using redundant switching states, thereby reducing the total number of evaluations and eliminating the need for weighting factor tuning, specifically for a dual three-level T-type multilevel converter connected to an open-ends induction motor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional predictive torque control evaluates all switching states (729 evaluations), then comprehensive torque and flux control is achieved, but computational time and complexity increase significantly
Solution Approach 1:
The patent segments the cost function evaluation into two separate steps: first evaluating voltage vectors (61 evaluations) to determine optimal torque and flux control, then evaluating only the redundant switching states (4 evaluations) for capacitor balancing. This segmentation reduces total evaluations from 729 to 65 while maintaining control precision.
Solution Approach 2:
The patent extracts the capacitor balancing function from the main torque control cost function evaluation. By separating the evaluation of voltage vectors from the evaluation of redundant switching states, it removes the computational burden of evaluating all 729 switching states while preserving both torque control and capacitor balancing functions.
2Stability of the object's composition
If comprehensive cost function evaluation including capacitor balancing is performed, then capacitor voltage balancing is achieved, but the number of evaluations and computational complexity increase
Solution Approach 1:
The patent segments the control algorithm into two distinct evaluation phases: voltage vector evaluation for torque control and redundant switching state evaluation for capacitor balancing. This reduces algorithmic complexity by avoiding the need to evaluate all 729 switching states while achieving both control objectives.
Solution Approach 2:
The patent extracts capacitor balancing from the comprehensive switching state evaluation and implements it separately using only the four redundant switching states identified in the first evaluation step. This extraction significantly reduces computational complexity while maintaining capacitor voltage balance.
3Measurement precision
If all switching states are evaluated for optimal control, then precise torque and flux control is achieved, but the execution time becomes impractical for real-time control
Solution Approach 1:
The patent segments the control execution into two fast evaluation steps totaling 65 evaluations, compared to the impractical 729 evaluations required for complete switching state assessment. This segmentation enables real-time control execution while maintaining flux control precision through the first evaluation step.
Solution Approach 2:
The patent performs preliminary evaluation of voltage vectors to identify the optimal voltage vector and its associated four redundant switching states before performing the second evaluation for capacitor balancing. This preliminary action eliminates the need for subsequent evaluation of the remaining 661 switching states, enabling real-time execution.
Data Source
AI summary
Capacitor balancing of a dual three-level (3L) T-type converter based on silicon carbide (SiC) discrete semiconductors was performed with the converter feeding an open-ends induction motor (OEIM). A model predictive control (MPC) using a two step cost function calculation was developed to balance the DC link capacitors and control the machine torque simultaneously. The number of redundant switching states used was reduced without affecting the operating voltage vectors, which substantially reduced the computational time. A simulation and experimental results are in good agreement.