PWM Control Fixed Calculation Variable Sampling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for controlling electrical machines using PWM signals at a variable frequency require all control algorithms to be executed at this varying frequency, leading to inefficiencies and issues with harmonics that affect electromagnetic compatibility and vibration noise.

Innovation Solution

Calculating PWM signals at a predetermined fixed calculation frequency and writing them into a register for rereading at the variable sampling frequency, using a device with a first clock for calculation and a second clock for sampling, thereby separating the calculation and sampling processes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If PWM signals are generated at a variable sampling frequency to reduce harmonic energy concentration, then electromagnetic compatibility and vibration noise are improved, but all control algorithms must be executed at this variable frequency which increases computational complexity and processing burden

Engineering Contradiction:
Improveharmonic energy concentrationVSAvoidcontrol algorithm execution complexity
Core Design Contradiction:
Object-affected harmful factorsVSDevice complexity

Solution Approach 1:

The patent divides the control system into two independent parts: a calculation unit that generates PWM signals at a fixed frequency, and a register that reads these signals at a variable frequency. This segmentation allows the computational algorithms to run at a stable fixed frequency while the PWM output frequency varies, resolving the contradiction between harmonic reduction and computational complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a register as an intermediary component between the calculation unit and the PWM output. This register acts as a buffer that decouples the fixed-frequency calculation from the variable-frequency PWM generation, enabling harmonic energy spreading without requiring algorithms to execute at variable frequencies.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If PWM signals are calculated at a fixed frequency to simplify control algorithm execution, then computational processing is simplified, but harmonic energy remains concentrated causing electromagnetic compatibility and vibration noise issues

Engineering Contradiction:
Improvecontrol algorithm execution simplicityVSAvoidharmonic energy concentration
Core Design Contradiction:
Device complexityVSObject-affected harmful factors

Solution Approach 1:

The system is segmented into a fixed-frequency calculation portion and a variable-frequency output portion. The calculation unit maintains simple fixed-frequency operations while the register introduces variable frequency sampling to spread harmonic energy, thus maintaining algorithm simplicity while reducing harmful harmonics.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The register serves as an intermediary that transforms the fixed-frequency calculated signals into variable-frequency PWM outputs. This allows the calculation to remain simple and fixed-frequency while the output frequency varies to distribute harmonic energy across a broader spectrum, reducing peak harmonic concentrations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3298687B1Method and device for controlling a rotating electrical machine by pwm signals, and corresponding electrical machine of a motor vehicle
Publication Date: 2019.03.13 VALEO EQUIP ELECTRIC MOTEUR
  • EP3298687B1 patent drawingFigure 1
  • EP3298687B1 patent drawingFigure 2a~2b

AI summary

The method according to the invention relates to the control of an electrical machine (1) by PWM signals (C) of the type where the PWM signals are sampled at a variable sampling frequency (Fpwm). According to the invention, a step of calculating the PWM signals at a predetermined fixed calculation frequency (Fcal), a step of writing the PWM signals in a register (11), and a step of rereading the PWM signals in the register at the sampling frequency (Fpwm) are performed successively. According to one particular embodiment, the calculation is a determination of the control vectors of a vector pulse-width modulation.