Power Plant Control Using Relevant Parameter Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for controlling power generation systems are inefficient in determining optimal control methods with high computational effort and imprecise parameter selection, limiting effective system control.

Innovation Solution

The method employs a gradient-free optimization technique using an operating data set and system model to select relevant parameters through adaptive mutual information feature selection and model-based or model-free reinforcement learning, enabling improved control with reduced computational effort.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional control methods are used for power generation systems, then comprehensive control can be achieved, but computational effort is high and parameter selection is imprecise

Engineering Contradiction:
Improveparameter selection precisionVSAvoidcomputational effort
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and selects only the most relevant parameters from the complete system model using adaptive mutual information feature selection. This extraction process identifies key parameters that have the highest influence on system performance, eliminating redundant parameters and reducing computational complexity while maintaining control effectiveness.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality by treating different parameters with different levels of importance. Through adaptive mutual information analysis, the system identifies which parameters require precise control and which can be simplified, allocating computational resources efficiently to the most critical parameters rather than treating all parameters equally.

Inventive Principle:
Principle #3Local quality

2Reliability

If comprehensive parameters are selected for control, then system control accuracy is improved, but computational time increases

Engineering Contradiction:
Improvecontrol accuracyVSAvoidcomputational time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary action by conducting adaptive mutual information feature selection before the actual control process. This preliminary parameter selection step identifies and prepares the most relevant parameters in advance, so that during real-time control, only these pre-selected parameters need to be processed, significantly reducing computational time while maintaining control accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameter set dynamically by using adaptive mutual information to identify which parameters are most relevant under current operating conditions. This allows the system to adjust the number and type of parameters being controlled based on the specific situation, optimizing the balance between control accuracy and computational efficiency.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3489773B1Method for the computer-based control of a technical system, in particular an energy generation plant
Publication Date: 2024.12.25 SIEMENS AG
  • EP3489773B1 patent drawingFigure 1
  • EP3489773B1 patent drawingFigure 2
  • EP3489773B1 patent drawingFigure 3

AI summary

The invention relates to a method for computer-aided control of a technical system, in particular a power generation plant, to achieve a predetermined technical behavior of the technical system, wherein an operating data set is provided for controlling the system, wherein a system model is provided for describing the functioning of the technical system, wherein an optimization data set is determined using an optimization method based on the operating data set and the system model, wherein relevant parameters of the technical system are selected on the basis of the optimization data set using a selection procedure, which enable more advantageous control of the technical system than other parameters of the technical system, wherein a control method for the technical system is determined using the selected relevant parameters, and wherein the technical system is controlled using the control method.