Boiler Combustion Control Using Neural Network Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current boiler control methods in thermal power plants prioritize stable combustion over optimal conditions, leading to suboptimal combustion efficiency and increased emissions, requiring manual expertise and inefficient emission management.

Innovation Solution

A system and method that utilize a boiler combustion model, optimized by an artificial neural network, to calculate and implement real-time control values for improving combustion efficiency while minimizing emissions, incorporating dynamic bias tracking to gradually adjust control objects without sudden state changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual expert control is used to maintain stable combustion, then boiler operational stability is improved, but combustion efficiency deteriorates and emissions increase

Engineering Contradiction:

Improveboiler operational stabilityVSAvoidcombustion efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent replaces manual expert control (mechanical/systematic approach) with an artificial neural network-based automated control system. The neural network learns optimal combustion parameters from historical data and automatically adjusts control variables, substituting human expertise with an intelligent algorithm that can simultaneously optimize both stability and efficiency.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system dynamically changes combustion control parameters (such as fuel-air ratio, combustion air flow, mill outlet temperature) based on real-time conditions and learned patterns. The neural network continuously optimizes these parameters to achieve the best combination of operational stability and combustion efficiency, rather than maintaining fixed manual settings.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If manual expert control is used to maintain stable combustion, then boiler operational stability is improved, but harmful emissions increase

Engineering Contradiction:

Improveboiler operational stabilityVSAvoidemissions (nitrogen oxides and carbon dioxide)

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system implements a feedback mechanism where the neural network continuously monitors combustion parameters and emissions, compares actual performance with target values, and automatically adjusts control variables to minimize emissions while maintaining operational stability. The feedback loop enables continuous optimization of the combustion process.

Inventive Principle:

Principle #23Feedback

3Productivity

If sudden optimization changes are applied to improve combustion efficiency, then combustion efficiency is improved, but operational stability deteriorates

Engineering Contradiction:

Improvecombustion efficiencyVSAvoidoperational stability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts combustion parameters based on real-time conditions and the rate of change. The neural network incorporates dynamic constraints that prevent abrupt changes in control variables, allowing the system to adapt to changing conditions while maintaining operational stability through controlled, gradual adjustments rather than sudden optimizations.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system applies cushioning by implementing rate-of-change constraints and gradual transition mechanisms. Before applying optimization changes, the system buffers sudden adjustments by limiting the maximum rate of change for control variables, preventing operational disruptions while still achieving combustion efficiency improvements over time.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS11287126B2System and method for controlling operation of boiler

Publication Date: 2022.03.29 DOOSAN HEAVY IND & CONSTR CO LTD

AI summary

A system for controlling a boiler in a power plant to ensure combust under optimized conditions is provided. The system for controlling an operation of the boiler may include an optimizer configured to perform a combustion optimization operation for the boiler using a boiler combustion model to calculate an optimum control value for at least one control object of the boiler, and an output controller configured to receive the calculated optimum control value from the optimizer and control the control object according to the optimum control value.