Torque-to-Current Mapping Using Polynomial Lookup Tables

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional torque-to-current mapping modules in electric machine drive systems require significant memory resources due to the use of large lookup tables, exceeding available memory in many systems.

Innovation Solution

The proposed solution involves selecting a subset of one-dimensional lookup tables based on DC input voltage and angular rotation speed, applying coefficients within polynomial functions to generate current commands, and interpolating output values to reduce memory consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional lookup tables are used for torque-to-current mapping, then mapping accuracy is maintained, but memory resource consumption exceeds available memory

Engineering Contradiction:
Improvemapping accuracyVSAvoidmemory resource consumption
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent divides the conventional large two-dimensional lookup table into multiple smaller one-dimensional lookup tables, each storing coefficients for polynomial functions. This segmentation reduces the memory footprint while maintaining mapping accuracy through mathematical reconstruction of the original mapping relationship.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent replaces the direct lookup table approach with a mathematical model using polynomial functions. Instead of storing complete mapping data, the system uses polynomial expressions with coefficients from lookup tables to compute current commands, substituting direct memory access with computational processing.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Quantity of substance

If polynomial functions with coefficients are used, then memory consumption is reduced, but computation complexity increases

Engineering Contradiction:
Improvememory consumptionVSAvoidcomputation complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent changes the representation parameters from storing complete mapping values to storing polynomial coefficients. This parameter transformation reduces memory requirements while the polynomial computation provides a systematic way to reconstruct the mapping relationship with controlled computational complexity.

Inventive Principle:
Principle #35Parameter changes

3Quantity of substance

If fewer lookup tables are used, then memory resources are conserved, but interpolation requirements increase

Engineering Contradiction:
Improvememory resourcesVSAvoidinterpolation operations
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent uses a partial representation of the mapping data through polynomial coefficients rather than complete lookup tables. This partial action approach stores only essential parameters (coefficients) and performs computations to generate full mapping results, reducing memory usage while maintaining functional completeness.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS8729847B2Methods, systems and apparatus for generating current commands used to control operation of an electric machine
Publication Date: 2014.05.20 GM GLOBAL TECHNOLOGY OPERATIONS LLC
  • US8729847B2 patent drawing
  • US8729847B2 patent drawing
  • US8729847B2 patent drawing

AI summary

Embodiments of the present disclosure relate to methods, systems and apparatus for mapping torque to current to generate current commands used to control operation of an electric machine. Based on a DC input voltage, lookup tables (LUTs) are selected. Each of the selected LUTs includes a plurality (B) of first entries that correspond a particular input value of an angular rotation speed of the electric machine, and a set of coefficients are output from each of the selected LUTs that can be applied within a first polynomial function to generate a plurality of first polynomial functions each having a different sets of coefficients. A plurality of particular output values for a first current command are generated via the plurality of first polynomial functions. The particular output values for the first current command can be interpolated to generate a final output value for the first current command.