CN-122018318-A - Multi-parameter self-adaptive coordination system and method for high-alkali coal boiler

CN122018318ACN 122018318 ACN122018318 ACN 122018318ACN-122018318-A

Abstract

The invention belongs to the technical field of thermal power generation process control, and provides a multi-parameter self-adaptive coordination system and method for a high-alkali coal boiler, wherein the system comprises the steps of extracting operation state characteristics representing the current working condition based on boiler real-time operation data; the method comprises the steps of generating control action suggestions of at least two different sources in parallel based on the characteristics, synchronously carrying out security risk assessment, acquiring dynamic boundary constraint based on the running state characteristics, screening the control suggestions, carrying out weighted fusion and boundary correction on the screened suggestions and preset target weights, outputting final control instructions, sending the instructions to an execution mechanism for coordinated control, collecting actual system states, carrying out parameter on-line adjustment on an encoder based on actual and predicted errors, and integrating multi-source control experience and real-time optimization.

Inventors

Jiang Jiading
Liu yasuo
XUE JIANXIONG
YU QIANG
WANG JINGJIE
YE HUAN
ZHANG ZHENGSHUN
ZHU LINGKUN
FAN XIAOCHAO
ZHANG YONGTAO
WANG JIANGLEI
CUI ZUTAO
WANG YINGCHAO
CAO XIGUO
HUO ZHENGYU
WU KE

Assignees

新疆工程学院

Dates

Publication Date: 20260512
Application Date: 20260210

Claims (7)

1. The multi-parameter self-adaptive coordination method for the high-alkali coal boiler is characterized by comprising the following steps of: extracting the operation state characteristics representing the current operation working condition based on the real-time operation state data of the high-alkali coal boiler; Generating at least two control action suggestions with different sources in parallel based on the running state characteristics, and synchronously carrying out security risk assessment on various control action suggestions to obtain an assessment result; acquiring boundary constraint based on the running state characteristics, screening various control action suggestions based on the evaluation result, and carrying out weighted fusion and boundary correction by combining the screened control action suggestions, the running state characteristics and preset target weights to output a final control instruction; Issuing the final control instruction to an executing mechanism corresponding to the high-alkali coal boiler, performing multi-parameter coordination control, and collecting the actual system state after executing the final control instruction; and carrying out parameter on-line adjustment on the encoder extracting the running state characteristics based on the error between the actual system state and the prediction state corresponding to the final control instruction.
2. The multi-parameter adaptive coordination method for a high-alkali coal-oriented boiler according to claim 1, wherein the extracting of the operation state characteristics representing the current operation condition based on the real-time operation state data of the high-alkali coal-oriented boiler comprises the following steps: Acquiring original running state data of a high alkali coal boiler in real time, acquiring coal quality data representing the characteristics of the current coal alkali metal, and carrying out fusion processing on the original running state data and the coal quality data to generate enhanced state data; Calculating the contribution weight of each characteristic dimension in the enhanced state data to the current combustion working condition, weighting the enhanced state data according to the contribution weight, and outputting the weighted state characteristics; The weighted state features are mapped to the operating state features by a pre-trained encoder.
3. The multi-parameter self-adaptive coordination method for high-alkali coal-oriented boiler according to claim 2, wherein the generating control action suggestions of at least two different sources based on the running state features in parallel and synchronously performing security risk assessment on each type of control action suggestions to obtain an assessment result comprises: Taking the running state characteristics as input, inquiring and calling a control strategy obtained based on historical optimal control data training, outputting a first control action suggestion, and calculating to obtain a first safety risk value corresponding to the first control action suggestion; taking the running state characteristics as an initial state of rolling optimization, performing forward rolling calculation and optimization in a preset limited domain by taking the optimization meeting the multi-target performance index as a criterion, solving to obtain a second control action suggestion, and calculating to obtain a second safety risk value corresponding to the second control action suggestion; and the first security risk value and the second security risk value together form the evaluation result of various control action suggestions, and the evaluation result is bound with the control action suggestions corresponding to the evaluation result and output.
4. The multi-parameter adaptive coordination method for high-alkali coal-fired boiler according to claim 3, wherein the obtaining boundary constraint based on the running state characteristics, screening various control action suggestions based on the evaluation result, and performing weighted fusion and boundary correction by combining the screened control action suggestions, the running state characteristics and preset target weights, and outputting a final control instruction comprises: Based on the running state characteristics, identifying the category of the current running working condition, and matching and acquiring a corresponding basic boundary constraint set from a pre-constructed multi-working-condition constraint knowledge base according to the identified working condition category; based on the running state characteristics and the basic boundary constraint set, carrying out constraint fusion by combining with real-time equipment state information, and outputting boundary constraint; Performing compliance verification on the primarily screened control action suggestions according to the boundary constraint to form a to-be-fused control suggestion set; For each control action suggestion in the to-be-fused control suggestion set, taking the corresponding running state characteristic and the preset target weight as inputs, calculating to obtain a dynamic fusion weight of each suggestion under the current working condition, and carrying out weighted summation to generate an initial fusion control instruction; and carrying out boundary normalization on the parameter components of the initial fusion control instruction, and outputting the final control instruction meeting all physical and safety constraints.
5. The multi-parameter adaptive coordination method for the high-alkali coal boiler according to claim 4, wherein the issuing the final control instruction to the corresponding executing mechanism of the high-alkali coal boiler performs multi-parameter coordination control, and collects the actual system state after executing the final control instruction, and the method comprises the following steps: Based on the running state characteristics, performing multi-time scale decomposition on the final control instruction, and decomposing the final control instruction into an instruction set comprising a fast-change loop control instruction and a slow-change loop control instruction; before the instruction set is issued to each executing mechanism, carrying out self-adaptive dead zone compensation correction on the control instruction according to the real-time state and the preset dead zone characteristic of each executing mechanism, and generating a corrected control instruction; Synchronously issuing the corrected control instructions to corresponding execution mechanisms, and simultaneously sending synchronous execution signals to each execution mechanism to enable a fuel supply mechanism, an air inducing mechanism and a water supply mechanism to perform cooperative action according to the sequential logic of the corrected control instructions so as to realize multi-parameter coordination control; In the action process of each executing mechanism, acquiring process variables of the high-alkali coal boiler in real time, wherein the process variables comprise hearth temperature distribution, flue gas oxygen content, main steam parameters and actual feedback position signals of each executing mechanism to form executing process state data; and when the execution process state data reach a quasi-steady state, extracting data in a preset effective observation window from the execution process state data as the actual system state.
6. The multi-parameter adaptive coordination method for high-alkali coal-oriented boiler according to claim 5, wherein the performing parameter on-line adjustment on the encoder extracting the running state feature based on the error between the actual system state and the predicted state corresponding to the final control instruction comprises: Calculating model prediction errors based on the actual system state and a prediction state made according to the running state characteristics and the final control instruction, and constructing comprehensive loss for optimizing encoder parameters by combining historical distribution characteristics of the running state characteristics based on the model prediction errors; performing online incremental adjustment on parameters of the encoder with the aim of minimizing the comprehensive loss, and performing effect verification on the operation state characteristics re-extracted after adjustment; and storing the verified actual system state, the final control instruction and the corresponding optimized running state characteristics as a newly added sample.
7. A multi-parameter adaptive coordination system for high alkali coal-oriented boilers, for performing a multi-parameter adaptive coordination method for high alkali coal-oriented boilers as claimed in any one of claims 1 to 6, wherein said coordination system comprises: the characteristic acquisition module is used for extracting the operation state characteristics representing the current operation working condition based on the real-time operation state data of the high-alkali coal boiler; The risk assessment module is used for generating control action suggestions of at least two different sources in parallel based on the running state characteristics, and synchronously carrying out security risk assessment on various control action suggestions to obtain an assessment result; The control instruction output module is used for acquiring boundary constraint based on the running state characteristics, screening various control action suggestions based on the evaluation result, and carrying out weighted fusion and boundary correction by combining the screened control action suggestions, the running state characteristics and preset target weights to output a final control instruction; The instruction execution module is used for issuing the final control instruction to an execution mechanism corresponding to the high-alkali coal boiler, carrying out multi-parameter coordination control, and collecting the actual system state after the final control instruction is executed; And the parameter feedback module is used for carrying out parameter online adjustment on the encoder extracting the running state characteristics based on the error between the actual system state and the prediction state corresponding to the final control instruction.

Description

Multi-parameter self-adaptive coordination system and method for high-alkali coal boiler Technical Field The invention relates to the technical field of thermal power generation process control, in particular to a multi-parameter self-adaptive coordination system and method for a high-alkali coal boiler. Background High alkali coal is used as a fuel with special combustion characteristics, and the combustion process in a power station boiler shows complex dynamic behaviors. The high content of alkali metals (such as sodium, potassium, etc.) in coal is liable to cause serious problems of slagging, contamination and corrosion during combustion, and significantly affects combustion stability and pollutant formation characteristics. This makes the control of the operation of the high alkali coal boiler extremely complex and sensitive. In the actual operation of utility boilers, especially in the context of participating in grid depth peaking, the boiler needs to be subjected to frequent, large-scale load adjustment. The process involves the cooperative variation of several key parameters, such as fuel quantity, secondary air quantity, proportioning, burn-out air quantity, water supply flow quantity, etc. The traditional control scheme mostly depends on static parameter proportioning tables obtained by test under fixed working conditions, or adopts a single-loop control strategy based on classical PID. The method has obvious limitations that firstly, static proportion cannot adapt to inherent coal quality fluctuation (such as alkali metal content and volatile matter change) of high-alkali coal, so that control parameters deviate from an optimal value, combustion efficiency is reduced, emission of pollutants such as NOx exceeds standard, secondly, in the process of dynamic load change, a fixed parameter coupling relation is difficult to realize rapid and stable load response, and safety risks of unstable combustion and even fire extinguishment are easily caused by asynchronous adjustment of each parameter, and finally, the traditional method lacks dynamic weighing capability on multiple control targets (economy, environmental protection and flexibility), and cannot explore and track optimal operation working condition points in real time on the premise of meeting strict safety constraint. Although some advanced process control methods (such as fuzzy control, model predictive control and the like) are applied to boiler optimization in the prior art, the methods still have insufficient comprehensive challenges of strong nonlinearity, large delay, multivariable strong coupling, working condition time variation and the like of high-alkali coal combustion in the aspects of fusion utilization of multi-source control experience, dynamic perception of operation safety boundaries and online adaptability of a controller to slow time variation of combustion system characteristics. Therefore, the development of the multi-parameter self-adaptive coordination method of the high-alkali coal boiler, which can intelligently integrate multi-source decision, sense the safety boundary in real time and continuously self-evolve, has important significance for guaranteeing the safety of the boiler, improving the running economy and realizing clean and flexible peak regulation. Disclosure of Invention In order to solve the technical problems, the invention provides the multi-parameter self-adaptive coordination system and the method for the high-alkali coal boiler, which effectively solve the problems that the multi-parameter coordination of the high-alkali coal boiler is difficult, the safety guarantee and the multi-objective optimization are difficult to be compatible in the variable working condition operation, and provide an intelligent control solution of the system for realizing the safe, efficient, clean and flexible operation of the high-alkali coal boiler. The invention provides a multi-parameter self-adaptive coordination method for a high-alkali coal boiler, which comprises the following steps of: extracting the operation state characteristics representing the current operation working condition based on the real-time operation state data of the high-alkali coal boiler; Generating at least two control action suggestions with different sources in parallel based on the running state characteristics, and synchronously carrying out security risk assessment on various control action suggestions to obtain an assessment result; acquiring boundary constraint based on the running state characteristics, screening various control action suggestions based on the evaluation result, and carrying out weighted fusion and boundary correction by combining the screened control action suggestions, the running state characteristics and preset target weights to output a final control instruction; Issuing the final control instruction to an executing mechanism corresponding to the high-alkali coal boiler, performing multi-parameter coordination con