Double Virtual Voltage Vectors Predictive Torque Control for Five-Phase PMSM

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing model predictive torque control methods for five-phase permanent magnet synchronous motors face challenges with high computation burden, weighting factor complexity, and poor static performance due to the lack of effective harmonic suppression and flux ripple management.

Innovation Solution

A double virtual voltage vectors predictive torque control method is introduced, which eliminates the weighting factor by using virtual voltage vectors to suppress harmonics and employs the deadbeat principle for direct and quick vector selection, combined with voltage error tracking to enhance static performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If traditional model predictive torque control method is used, then the control structure is simple and dynamic response is fast, but the computation burden is heavy due to vector screening process

Engineering Contradiction:

Improvedynamic response speedVSAvoidcomputation burden

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The patent segments the voltage vectors into virtual voltage vectors that group multiple basic voltage vectors together. Instead of screening all individual basic vectors, the controller only needs to select from the reduced set of virtual vectors, significantly reducing computational burden while maintaining fast dynamic response.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension by synthesizing virtual voltage vectors from multiple basic vectors. This dimensional transformation allows the control system to operate in a simplified vector space where computation is reduced, while still achieving the desired torque and flux control through the synthesized virtual vectors.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Device complexity

If model predictive control method with reduced computation burden is used, then the computation burden is reduced, but the weighting factor is difficult to debug

Engineering Contradiction:

Improvecomputation burdenVSAvoiddebugging difficulty

Core Design Contradiction:

Device complexityVSEase of operation

Solution Approach 1:

The patent extracts and eliminates the weighting factor from the cost function by using a modified prediction approach. The virtual voltage vector selection is based on direct comparison of predicted torque and flux errors without requiring weighting factors, making the control parameters easier to debug and adjust.

Inventive Principle:

Principle #2Taking out (Extraction)

3Ease of operation

If model predictive control method with error ranking is used, then the weighting factor is eliminated, but the computation burden remains heavy due to multiple error calculations

Engineering Contradiction:

Improveweighting factor eliminationVSAvoidcomputation burden

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent segments the error calculation process by using virtual voltage vectors that pre-group multiple basic vectors. This segmentation reduces the number of error calculations needed, as selecting one virtual vector replaces the need to evaluate multiple individual basic vectors, eliminating both weighting factors and excessive computation.

Inventive Principle:

Principle #1Segmentation

4Device complexity

If model predictive control method with reduced candidate vectors is used, then the computational burden is reduced, but the static performance is poor due to lack of torque and flux ripple suppression

Engineering Contradiction:

Improvecomputation burdenVSAvoidstatic performance

Core Design Contradiction:

Device complexityVSManufacturing precision

Solution Approach 1:

The patent merges multiple basic voltage vectors into synthesized virtual voltage vectors. This merging allows the system to achieve better torque and flux ripple suppression through the combined effect of multiple vectors, improving static performance while maintaining reduced computation burden from the smaller virtual vector set.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent changes the parameter representation from individual basic voltage vectors to synthesized virtual voltage vectors with different amplitude and phase characteristics. This parameter transformation enables better control of torque and flux ripples, improving static performance without increasing computational complexity.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11356045B2Double virtual voltage vectors predictive torque control method without weighting factor for five-phase permanent magnet synchronous motor

Publication Date: 2022.06.07 JIANGSU UNIV

AI summary

A double virtual voltage vectors predictive torque control method without weighting factor for five-phase permanent magnet synchronous motor includes: obtaining the current component in the two-phase stationary coordinate system and the outputting voltage at k interval; one step delay compensation is performed to obtain the current component in the two-phase stationary coordinate system at k+1 interval; predicting the flux and torque of motor at k+1 interval; calculating the reference voltage vector needed by the motor at k+1 interval according to the deadbeat principle and selecting the first virtual voltage vector; selecting the second virtual voltage vector according to the voltage error tracking principle and calculating the duration of the first virtual voltage vector and the second virtual voltage vector respectively and then synthesizing the two vectors and outputting.