AI Converter Control Profiles for Loss and Stress Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing electronic converter control systems face challenges in optimizing operating properties such as power losses, aging, and stress on components due to fixed control signal profiles, which do not account for long-term behavior and system variables.

Innovation Solution

A method and device that utilize a trainable, data-based control signal model to predict and modify control signal profiles based on provided and predicted profiles, optimizing the temporal profile to improve the operating behavior of electronic converters, incorporating variables like thermal resistance and capacitance, and using models like artificial neural networks for training.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If fixed control signal profiles are used to operate electronic converters, then the control system is simple and easy to implement, but the operating properties such as power losses, aging, and stress on components cannot be optimized

Engineering Contradiction:
Improvepower lossesVSAvoidcontrol system complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The control signal profile is predicted in advance for a future time period, allowing the system to prepare optimal control signals before actual operation occurs. This predictive approach enables optimization of power losses and component stress without requiring complex real-time control adjustments

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses a trainable data-based control signal model that learns from historical operation data and provides feedback to continuously improve control signal profiles. This feedback mechanism enables the system to adapt to changing conditions and optimize operating properties while maintaining manageable complexity

Inventive Principle:
Principle #23Feedback

2Duration of action of stationary object

If fixed control signal profiles are used, then the control implementation is straightforward, but the longevity and aging characteristics of components are not optimized

Engineering Contradiction:
Improvecomponent longevityVSAvoidcontrol implementation complexity
Core Design Contradiction:
Duration of action of stationary objectVSDevice complexity

Solution Approach 1:

The system predicts control signal profiles in advance for future time periods, allowing optimization of component longevity before actual operation occurs. This proactive approach enables the system to adjust control signals to minimize stress and aging effects on components

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The trainable control signal model automatically learns and adapts to optimize component longevity through continuous operation. The system self-adjusts control parameters based on accumulated data, reducing the need for manual intervention and complex control implementation

Inventive Principle:
Principle #25Self-service

3Object-affected harmful factors

If fixed control signal profiles are used, then the system operation is simple, but stress on converter components cannot be reduced

Engineering Contradiction:
Improvestress on converter componentsVSAvoidcontrol signal generation complexity
Core Design Contradiction:
Object-affected harmful factorsVSDevice complexity

Solution Approach 1:

The system employs a data-based control signal model that continuously learns from operational data and provides feedback to optimize control signals. This feedback loop enables the system to identify and reduce stress patterns on converter components while maintaining simple operational control

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

Control signal profiles are predicted in advance, allowing the system to prepare stress-minimizing control signals before operation. This predictive capability enables the system to avoid high-stress conditions without requiring complex real-time monitoring and adjustment

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11855522B2Method and device for controlling an electronic converter in a technical system using methods of artificial intelligence
Publication Date: 2023.12.26 ROBERT BOSCH GMBH
  • US11855522B2 patent drawing
  • US11855522B2 patent drawing
  • US11855522B2 patent drawing

AI summary

A method is disclosed for operating a technical apparatus with an electronic converter controlled a control signal. A control signal profile is provided with which the electronic converter is to be operated. A predicted control signal profile is predicted based on the provided control signal profile. The predicted control signal profile is a predicted future profile of the control signal. A modified control signal profile of is obtained by modifying the provided control signal profile using a trainable, data-based control signal model. The control signal model is trained to determine the modified control signal profile based on the provided control signal profile and the predicted control signal profile. The electronic converter is operated using the modified control signal profile.