Optimization method for improving flexible operation of thermal power generating unit

Through the combination of real-time monitoring and machine learning models, a target adjustment strategy is generated, which solves the problems of degradation in the performance of the control system and insufficient economics during the operation of the thermal power unit, and realizes the intelligence and maximum benefits of the key control circuits of the unit.

CN119065238AActive Publication Date: 2024-12-03SHANGAN POWER PLANT OF HUANENG INT POWER CO LTD +1

Patent Information

Application Number
CN202411032371.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-12-03
Estimated Expiration
2044-07-30

AI Technical Summary

Technical Problem

During long-term operation, the thermal power unit has deteriorated the performance of the control system due to the aging of the environment and equipment, making it difficult to flexibly adjust and ignore economics, which increases the operating volume and monitoring pressure.

Method used

By monitoring the unit's operating status in real time and combining the parameter prediction model built based on machine learning algorithms, target adjustment strategies are generated to achieve intelligent control of the unit's critical control loop and parameter flexibility adjustment to maximize benefits.

Benefits of technology

It realizes intelligent control of key control circuits of thermal power units and flexible parameter adjustments, reduces operating costs, and improves the economy and operation efficiency of the unit.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119065238A_ABST
    Figure CN119065238A_ABST
Patent Text Reader

Abstract

The invention provides an optimization method for improving flexible operation of a thermal power generating unit, and relates to the technical field of big data, and the method comprises the steps: analyzing first operation state data obtained by monitoring the operation state of a target unit in real time, and obtaining a unit operation state result; predicting key control parameters of the target unit based on the first operation state data by using a parameter prediction model constructed by adopting a machine learning algorithm to obtain a control parameter prediction result; and analyzing the control parameter prediction result and the unit operation state result to generate a target adjustment strategy to act on the target unit. The running state of the unit at the current moment is obtained by monitoring and analyzing the running state of the unit in real time, and a target adjustment strategy is generated to act on the unit by combining the running state of the unit with a control parameter prediction result of the unit obtained by prediction of a parameter prediction model constructed based on a machine learning algorithm for analysis. The intelligent control of the key control loop of the unit is realized, and the benefit of the unit is maximized while the parameters are flexibly adjusted.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to an optimization method for improving the flexible operation of a thermal power unit. Background Art

[0002] The production process of thermal power units is complex, and the equipment system is in a high temperature, high pressure, high dust and high corrosion environment for a long time. As the unit runs for a long time, the equipment gradually ages and its characteristics change significantly, resulting in a decline in the performance of the unit control system, which in turn forces the operators to frequently intervene manually, greatly increasing the amount of operations and the pressure of monitoring. However, the operation of most units currently still relies on the experience of the operators, and from the perspective of safe operation of the units, it is difficult to make flexible adjustments to the changes in the units, and the economic efficiency of the units is also ignored. Therefore, how to achieve flexible adjustments to the key control loops of the units while improving economic efficiency has become one of the current research centers.

[0003] Therefore, the present invention provides an optimization method for improving the flexible operation of a thermal power unit. Summary of the invention

[0004] The present invention provides an optimization method for improving the flexible operation of a thermal power unit, which is used to obtain the current operating status of the unit by real-time monitoring and analyzing the operating status of the unit, and combine the analysis with the control parameter prediction results of the unit obtained by using a parameter prediction model built based on a machine learning algorithm to generate a target adjustment strategy to act on the unit, thereby realizing intelligent control of the key control loops of the unit, flexible parameter adjustment, and maximizing the unit benefit.

[0005] The present invention provides an optimization method for improving the flexible operation of a thermal power unit, comprising:

[0006] Step 1: monitor the operating status of the target unit in real time to obtain first operating status data;

[0007] Step 2: Based on the first operating status data, analyzing the operating status of the current target unit to obtain a unit operating status result;

[0008] Step 3: using a parameter prediction model constructed using a machine learning algorithm, predicting key control parameters of the target unit based on the first operating state data to obtain control parameter prediction results;

[0009] Step 4: Combine the control parameter prediction results with the unit operation status results for analysis, generate a target adjustment strategy to act on the target unit, and maximize the unit benefit.

[0010] Preferably, the first operating status data includes power generation power, efficiency, pressure and temperature, operating load, fuel consumption and emissions, unit flow and operating time.

[0011] Preferably, based on the first operating status data, the operating status of the current target unit is analyzed to obtain the unit operating status result, including:

[0012] Obtaining an operation data threshold table of each operation system in the target unit;

[0013] Based on the operation data threshold table, comparing the first operation status data of each operation system with the corresponding operation data threshold to obtain a first operation difference;

[0014] Compare and analyze the first operation difference with a preset difference threshold to determine a grade labeling result of the first operation status data;

[0015] Calculating a system risk coefficient of the current operating system based on the level labeling result of the first operating status data of each operating system;

[0016] The calculation formula of the system risk coefficient is as follows:

[0017] Where X represents the system risk coefficient of the current operating system; n i It represents the number of running status data of the i-th state level in the current running system, where i = 1, 2, 3; n0 represents the total number of running status data of the current running system; Δs ij It is represented by the first operation difference between the jth operation status data of the ith state level in the current operation system and the corresponding operation data threshold, where j = 1, 2, 3, ..., n i ;μ ij It is represented as the risk impact weight of the current operating system corresponding to the j-th operating status data of the i-th status level in the current operating system;

[0018] The system risk coefficient of each operating system of the target unit is aggregated to generate a unit operating status result.

[0019] Preferably, a parameter prediction model constructed using a machine learning algorithm is used to predict key control parameters of the target unit based on the first operating state data to obtain control parameter prediction results, including:

[0020] Extracting historical operating status data, equipment parameters, and environmental condition data of the target unit, and performing data preprocessing to obtain a first data set;

[0021] Dividing the first data set according to a set ratio to obtain a training data set and a test data set;

[0022] Extracting the correlation influence characteristics of the key control parameters of the target unit, and combining the training data set as input, training a neural network to obtain an initial prediction model;

[0023] Using the test data set to perform performance evaluation on the initial prediction model, and based on the performance evaluation results, optimizing the model to obtain a parameter prediction model;

[0024] The first operating state data is input into the parameter prediction model to obtain a control parameter prediction result.

[0025] Preferably, the control parameter prediction result is combined with the unit operation status result for analysis, and a target adjustment strategy is generated to act on the target unit to maximize the unit benefit, including:

[0026] Determine the safety risk level of each operating system at the current moment according to the system risk coefficient in the unit operation status result;

[0027] Extracting the first state data-adjustment strategy reference table of each running system at the current moment based on the security risk level from the risk level-adjustment strategy database;

[0028] Determine the data adjustment strategies for the first operating state data marked at different levels in the operating system from the first state data-adjustment strategy reference table, and output them as reference adjustment strategies;

[0029] Determining control input parameters for controlling key control parameters of the target unit;

[0030] The parameters directly related to the unit economy among the control input parameters are regarded as associated input parameters, and the parameters not directly related to the unit economy are regarded as non-associated input parameters;

[0031] According to the reference adjustment strategy, respectively obtain the control parameter adjustment strategy of the associated input parameter and the non-associated input parameter of each key control parameter;

[0032] Based on the control parameter prediction results, the target prediction results of the associated input parameters and the non-associated input parameters of the key control parameters are obtained by analyzing the target optimization algorithm;

[0033] The parameter adjustment strategy and target prediction results of the associated input parameters and non-associated input parameters of each key control parameter are combined and analyzed to generate a target adjustment strategy.

[0034] Preferably, based on the control parameter prediction results, the target prediction results of the associated input parameters and the non-associated input parameters of the key control parameters are obtained by analyzing the target optimization algorithm, including:

[0035] Obtain technical performance indicators and economic performance indicators of key control parameters;

[0036] The associated input parameters and non-associated input parameters of the key control parameters are used as input variables, and the output variable is the comprehensive cost of the control loop. The control parameter-economic model is established by combining the corresponding technical performance indicators and economic performance indicators.

[0037] Based on the control parameter-economic model, the lowest hourly comprehensive cost of the unit is taken as the objective function, the input variables are used as linear constraints, and a target optimization algorithm is used to build a parameter optimization prediction model;

[0038] Solving the parameter optimization prediction model to obtain first prediction result sequences of associated input parameters and non-associated input parameters of the current key control parameters within a set prediction time period;

[0039] The prediction result at the current sampling moment is extracted from the first prediction result sequence and output as the target prediction result of the associated input parameters and the non-associated input parameters.

[0040] Preferably, the control parameter adjustment strategy of the associated input parameters and the non-associated input parameters of each key control parameter and the target prediction results are combined and analyzed to generate a target adjustment strategy, including:

[0041] According to the combined analysis of the control parameter adjustment strategy and the grade annotation results of the associated input parameters, a first parameter adjustment range is obtained;

[0042] Obtaining a second parameter adjustment range based on a combined analysis of a control parameter adjustment strategy and a grade annotation result for non-correlated input parameters;

[0043] When the association-target prediction results of all the associated input parameters of the key control parameter belong to the corresponding first parameter adjustment range, the association-target prediction results are output as the associated input parameter optimization values ​​of the corresponding associated input parameters;

[0044] When the key control parameter has a single or multiple associated input parameters whose associated-target prediction results do not fall within the first parameter adjustment range, the corresponding associated input parameters are marked as first parameters;

[0045] The linear constraint condition of the first parameter is adjusted by using the constraint adjustment coefficient, and based on the adjusted linear constraint condition, the parameter optimization prediction model is re-solved to obtain the target prediction results of the associated input parameter and the non-associated input parameter;

[0046] Analyze the target prediction result of the associated input parameter and the corresponding first parameter adjustment range, determine the optimized value of the associated input parameter and output it;

[0047] If the non-correlated-target prediction result of the non-correlated input parameter falls within the second parameter adjustment range, the current non-correlated-target prediction result is used as the non-correlated input parameter optimization value;

[0048] If the non-correlated-target prediction result of the current non-correlated input parameter does not fall within the second parameter adjustment range, then the absolute differences between the non-correlated-target prediction result and the second parameter adjustment upper limit and the second parameter adjustment lower limit are calculated respectively, and the first absolute difference and the second absolute difference are obtained correspondingly;

[0049] If the first absolute difference is greater than the second absolute difference, the second parameter adjustment upper limit is used as the non-correlated input parameter optimization value;

[0050] Otherwise, the lower limit of the second parameter adjustment is used as the non-correlated input parameter optimization value;

[0051] A target adjustment strategy is generated based on the associated input parameter optimization values ​​of the associated input parameters of each key control parameter and the non-associated input parameter optimization values ​​of the non-associated input parameters.

[0052] Preferably, the calculation formula of the constraint adjustment coefficient is as follows:

[0053] In the formula, P is the constraint optimization coefficient; Δf is the linear constraint adjustment factor; c1 is the corresponding first parameter adjustment upper limit of the current first parameter; c0 is the corresponding first parameter adjustment lower limit of the current first parameter; Δc1 is the absolute difference between the association-target prediction result of the current first parameter and the corresponding first parameter adjustment upper limit; Δc0 is the absolute difference between the association-target prediction result of the current first parameter and the corresponding first parameter adjustment lower limit; α is the influence weight of the current first parameter on the safety of the unit; β is the contribution compensation coefficient of the current first parameter to the economy of the unit.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] By real-time monitoring and analysis of the unit's operating status, the unit's current operating status is obtained, and combined with the unit's control parameter prediction results obtained by using a parameter prediction model built based on a machine learning algorithm, a target adjustment strategy is generated and applied to the unit, thereby achieving intelligent control of the unit's key control loops, flexible parameter adjustment, and maximizing the unit's benefits.

[0056] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0057] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0059] Figure 1 The present invention is a flowchart of an optimization method for improving the flexible operation of a thermal power unit in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0061] The embodiment of the present invention provides an optimization method for improving the flexible operation of a thermal power unit, such as Figure 1 As shown, including:

[0062] Step 1: monitor the operating status of the target unit in real time to obtain first operating status data;

[0063] Step 2: Based on the first operating status data, analyzing the operating status of the current target unit to obtain a unit operating status result;

[0064] Step 3: using a parameter prediction model constructed using a machine learning algorithm, predicting key control parameters of the target unit based on the first operating state data to obtain control parameter prediction results;

[0065] Step 4: Combine the control parameter prediction results with the unit operation status results for analysis, generate a target adjustment strategy to act on the target unit, and maximize the unit benefit.

[0066] In this embodiment, the target unit refers to a complete set of equipment and systems that use thermal power to generate electrical energy; the first operating status data includes power generation power, efficiency, pressure and temperature, operating load, fuel consumption and emissions, unit flow and operating time; the unit operating status result refers to the system risk coefficient of each operating system of the target unit, wherein the operating system includes a boiler system, a turbine system, a generator system, an auxiliary system and a control system; the system risk coefficient is used to characterize the degree of abnormality of the operating status of the operating system of the current target unit.

[0067] In this embodiment, the machine learning algorithm refers to an algorithm that automatically learns rules and patterns from data by analyzing and learning the data, and uses these rules and patterns to perform tasks such as prediction, classification, clustering, and dimensionality reduction; the parameter prediction model is a model obtained by training a neural network using pre-processed historical operating status data, equipment parameters, and environmental conditions of the unit, and is used to predict the future values ​​of the key control parameters of the unit, where the key control parameters include main steam temperature, pressure, mid-point temperature, denitrification efficiency, etc., and refer to parameters of key control loops such as main steam temperature, pressure, mid-point temperature, and denitrification.

[0068] In this embodiment, the control parameter prediction result refers to the predicted parameter value obtained by predicting the key control parameters of the target unit based on the first operating state data using the parameter prediction model; the target adjustment strategy is used to adjust the associated input parameters that are directly related to the economy of the key control parameters, as well as the non-associated input parameters that are not directly related to the economy, so as to achieve precise control of the key control parameters and ensure the flexibility adjustment of the target unit and maximize the unit efficiency.

[0069] The beneficial effects of the above technical solution are: by real-time monitoring and analysis of the operating status of the unit, the operating status of the unit at the current moment is obtained, and combined with the control parameter prediction results of the unit obtained by using the parameter prediction model built based on the machine learning algorithm for analysis, a target adjustment strategy is generated to act on the unit, thereby realizing intelligent control of the key control loops of the unit, flexible parameter adjustment, and maximizing the unit efficiency.

[0070] The embodiment of the present invention provides an optimization method for improving the flexible operation of a thermal power unit, and based on the first operating status data, the operating status of the current target unit is analyzed to obtain the unit operating status result, including:

[0071] Obtaining an operation data threshold table of each operation system in the target unit;

[0072] Based on the operation data threshold table, comparing the first operation status data of each operation system with the corresponding operation data threshold to obtain a first operation difference;

[0073] Compare and analyze the first operation difference with a preset difference threshold to determine a grade labeling result of the first operation status data;

[0074] Calculating a system risk coefficient of the current operating system based on the level labeling result of the first operating status data of each operating system;

[0075] The calculation formula of the system risk coefficient is as follows:

[0076] Where X represents the system risk coefficient of the current operating system; n i It represents the number of running status data of the i-th state level in the current running system, where i = 1, 2, 3; n0 represents the total number of running status data of the current running system; Δs ij It is represented by the first operation difference between the jth operation status data of the ith state level in the current operation system and the corresponding operation data threshold, where j = 1, 2, 3, ..., n i ;μ ij It is represented as the risk impact weight of the current operating system corresponding to the j-th operating status data of the i-th status level in the current operating system;

[0077] The system risk coefficient of each operating system of the target unit is aggregated to generate a unit operating status result.

[0078] In this embodiment, the target unit refers to a complete set of equipment and systems that use thermal power to generate electricity; the first operating status data includes power generation power, efficiency, pressure and temperature, operating load, fuel consumption and emissions, unit flow and operating time; the operating system includes a boiler system, a turbine system, a generator system, an auxiliary system and a control system; the operating data threshold table refers to the safety threshold of the operating status data of each operating system, which is a table pre-established based on the analysis of unit design parameters, safety specifications and historical fault data.

[0079] In this embodiment, the first operating difference refers to the difference between the first operating status data and the corresponding operating data threshold; the preset difference threshold is set in advance; the level marking result refers to the operating status level of the first operating status data, including level one, level two, level three and normal; the system risk coefficient is used to characterize the degree of abnormality of the operating status of the operating system of the current target unit.

[0080] In this embodiment, comparing and analyzing the first operation difference with a preset difference threshold to determine the level labeling result of the first operation status data includes:

[0081] If the first operation difference is less than the preset difference threshold, and the first operation status data is greater than the corresponding operation data threshold, the current first operation status data is marked as secondary status data;

[0082] If the first operation difference is less than the preset difference threshold, and the first operation status data is not greater than the corresponding operation data threshold, the current first operation status data is marked as level 3 status data;

[0083] If the first operation difference is not less than the preset difference threshold, and the first operation status data is greater than the corresponding operation data threshold, the current first operation status data is marked as primary status data;

[0084] If the first operation difference is not less than the preset difference threshold, and the first operation status data is not greater than the corresponding operation data threshold, the current first operation status data is marked as normal status data.

[0085] The beneficial effect of the above technical solution is that by real-time analysis of the unit operation data, it helps to determine the safe operation risks and performance of each operating system of the unit, and provide data support for subsequent performance optimization and improvement, and maintenance costs.

[0086] The embodiment of the present invention provides an optimization method for improving the flexible operation of a thermal power unit, using a parameter prediction model constructed by a machine learning algorithm to predict key control parameters of a target unit based on the first operating state data, and obtaining control parameter prediction results, including:

[0087] Extracting historical operating status data, equipment parameters, and environmental condition data of the target unit, and performing data preprocessing to obtain a first data set;

[0088] Dividing the first data set according to a set ratio to obtain a training data set and a test data set;

[0089] Extracting the correlation influence characteristics of the key control parameters of the target unit, and combining the training data set as input, training a neural network to obtain an initial prediction model;

[0090] Using the test data set to perform performance evaluation on the initial prediction model, and based on the performance evaluation results, optimizing the model to obtain a parameter prediction model;

[0091] The first operating state data is input into the parameter prediction model to obtain a control parameter prediction result.

[0092] In this embodiment, the equipment parameters include operating status parameters, such as switch status, operating mode, performance parameters, such as efficiency, speed, configuration parameters and maintenance parameters, fault and alarm parameters; environmental condition data refers to temperature, humidity, vibration and dust concentration, etc.; data preprocessing includes data cleaning, denoising and normalization, etc., the purpose of which is to ensure data quality and improve model performance.

[0093] In this embodiment, the first data set is a data set obtained by combining historical operating status data, equipment parameters and environmental condition data of the target unit after data preprocessing; the set ratio is set in advance.

[0094] In this embodiment, the key control parameters include main steam temperature, pressure, intermediate point temperature, denitrification efficiency, etc.; the associated influence characteristics refer to characteristic variables that have a significant impact on the flexible adjustment of key control parameters; the initial prediction model is a model obtained by training a neural network with the associated influence characteristics of the key control parameters of the target unit and a training data set as input; the parameter prediction model is a model obtained after tuning the initial prediction model, which is used to predict the future values ​​of the key control parameters of the unit; the control parameter prediction result refers to the key control parameter prediction value output after the first operating state data is input into the parameter prediction model.

[0095] The beneficial effect of the above technical solution is: by using machine learning algorithms to build a parameter prediction model, accurate prediction of the key control parameters of the target unit can be achieved, providing data support for the subsequent flexible adjustment of key control parameters and reducing operating costs.

[0096] The embodiment of the present invention provides an optimization method for improving the flexible operation of a thermal power unit, combining and analyzing the control parameter prediction result with the unit operation status result, generating a target adjustment strategy to act on a target unit, and realizing flexibility adjustment of the target unit and maximizing the unit benefit, including:

[0097] Determine the safety risk level of each operating system at the current moment according to the system risk coefficient in the unit operation status result;

[0098] Extracting the first state data-adjustment strategy reference table of each running system at the current moment based on the security risk level from the risk level-adjustment strategy database;

[0099] Determine the data adjustment strategies for the first operating state data marked at different levels in the operating system from the first state data-adjustment strategy reference table, and output them as reference adjustment strategies;

[0100] Determining control input parameters for controlling key control parameters of the target unit;

[0101] The parameters directly related to the unit economy among the control input parameters are regarded as associated input parameters, and the parameters not directly related to the unit economy are regarded as non-associated input parameters;

[0102] According to the reference adjustment strategy, respectively obtain the control parameter adjustment strategy of the associated input parameter and the non-associated input parameter of each key control parameter;

[0103] Based on the control parameter prediction results, the target prediction results of the associated input parameters and the non-associated input parameters of the key control parameters are obtained by analyzing the target optimization algorithm;

[0104] The parameter adjustment strategy and target prediction results of the associated input parameters and non-associated input parameters of each key control parameter are combined and analyzed to generate a target adjustment strategy.

[0105] In this embodiment, the system risk coefficient is used to characterize the degree of abnormality of the operating state of the operating system of the current target unit; the safety risk level refers to different levels divided according to the severity of the risk, which is used to describe and evaluate potential safety threats or hazards.

[0106] In this embodiment, the risk level-adjustment strategy database is composed of the safety risk level of the operating coefficient and the corresponding first state data-adjustment strategy reference table, wherein the first state data-adjustment strategy reference table is composed of the first operating state data and the corresponding data adjustment strategy; the reference adjustment strategy refers to the data adjustment strategy of the first operating state data labeled with different levels extracted from the first state data-adjustment strategy reference table.

[0107] In this embodiment, the control input parameters refer to parameters that can affect and control the key control parameter values, such as fuel quantity, water supply quantity, water injection quantity, etc.; the associated input parameters refer to control input parameters that are directly related to the unit economy, such as fuel quantity; and the unassociated input parameters refer to control input parameters that are not directly related to the unit economy.

[0108] In this embodiment, the control parameter adjustment strategy refers to the data adjustment strategy of the associated input parameters and the non-associated input parameters extracted from the reference adjustment strategy; the target optimization algorithm refers to a pre-set method for finding the optimal solution of the future predicted input quantities of the associated input parameters and the non-associated input parameters of the key control parameters, such as the genetic optimization algorithm and the particle swarm optimization algorithm; the target adjustment strategy is used to adjust the associated input parameters that are directly related to the economy of the key control parameters, and the non-associated input parameters that are not directly related to the economy, so as to achieve precise control of the key control parameters, ensure the intelligent control of the key control loops of the target unit and maximize the unit benefits, wherein the key control loops include main steam temperature, pressure, intermediate point temperature, denitrification, etc.

[0109] The beneficial effect of the above technical solution is: by combining the control parameter prediction results with the unit operation status results to analyze and generate a target adjustment strategy to act on the target unit, it is possible to achieve intelligent control of key control loops with the goal of optimal economy on the basis of ensuring the safe operation of the unit and equipment.

[0110] The embodiment of the present invention provides an optimization method for improving the flexible operation of a thermal power unit, which uses a target optimization algorithm to analyze and obtain target prediction results of associated input parameters and non-associated input parameters of key control parameters based on the control parameter prediction results, including:

[0111] Obtain technical performance indicators and economic performance indicators of key control parameters;

[0112] The associated input parameters and non-associated input parameters of the key control parameters are used as input variables, and the output variable is the comprehensive cost of the control loop. The control parameter-economic model is established by combining the corresponding technical performance indicators and economic performance indicators.

[0113] Based on the control parameter-economic model, the lowest hourly comprehensive cost of the unit is taken as the objective function, the input variables are used as linear constraints, and a target optimization algorithm is used to build a parameter optimization prediction model;

[0114] Solving the parameter optimization prediction model to obtain first prediction result sequences of associated input parameters and non-associated input parameters of the current key control parameters within a set prediction time period;

[0115] The prediction result at the current sampling moment is extracted from the first prediction result sequence and output as the target prediction result of the associated input parameters and the non-associated input parameters.

[0116] In this embodiment, the key control parameters include main steam temperature, pressure, intermediate point temperature, denitrification efficiency, etc., which refer to the parameters of key control loops such as main steam temperature, pressure, intermediate point temperature, and denitrification; technical performance indicators refer to key parameters used to measure equipment performance; economic performance indicators are key parameters used to measure the economy of equipment, for example, the technical performance indicators of the main steam temperature include rated temperature and temperature fluctuation range, and the economic performance indicators include thermal efficiency, fuel consumption, and maintenance cost.

[0117] In this embodiment, the control parameter-economy model is established by combining the performance indicators of key control parameters with associated input parameters and non-associated input parameters as input variables and the output variable as the comprehensive cost of the control loop; the hourly comprehensive cost of the unit refers to the comprehensive cost generated by the unit during each hour of operation, including fuel cost, maintenance cost, labor cost, etc.

[0118] In this embodiment, the target optimization algorithm refers to a pre-set method for finding the optimal solution for the future predicted input quantities of the associated input parameters and non-associated input parameters of the key control parameters, such as a genetic optimization algorithm and a particle swarm optimization algorithm; the parameter optimization prediction model is used to predict the parameter values ​​of the associated input parameters and non-associated input parameters of the key control parameters within a set prediction time period; the set prediction time period is set in advance; the first prediction result sequence refers to the predicted control input quantities of the associated input parameters and non-associated input parameters of the current key control parameters within the set prediction time period obtained by solving the parameter optimization prediction model, wherein the associated input parameters refer to the control input parameters that are directly related to the unit economy, such as the fuel quantity; the non-associated input parameters refer to the control input parameters that are not directly related to the unit economy.

[0119] The beneficial effect of the above technical solution is: by utilizing the target optimization algorithm, the target prediction results of the associated input parameters and non-associated input parameters of the key control parameters are obtained based on the control parameter prediction result analysis, thereby providing effective data support for achieving precise control of the key control parameters.

[0120] The embodiment of the present invention provides an optimization method for improving the flexible operation of a thermal power unit, combining and analyzing the control parameter adjustment strategy of the associated input parameters and the non-associated input parameters of each key control parameter and the target prediction result to generate a target adjustment strategy, including:

[0121] According to the combined analysis of the control parameter adjustment strategy and the grade annotation results of the associated input parameters, a first parameter adjustment range is obtained;

[0122] Obtaining a second parameter adjustment range based on a combined analysis of a control parameter adjustment strategy and a grade annotation result for non-correlated input parameters;

[0123] When the association-target prediction results of all the associated input parameters of the key control parameter belong to the corresponding first parameter adjustment range, the association-target prediction results are output as the associated input parameter optimization values ​​of the corresponding associated input parameters;

[0124] When the key control parameter has a single or multiple associated input parameters whose associated-target prediction results do not fall within the first parameter adjustment range, the corresponding associated input parameters are marked as first parameters;

[0125] The linear constraint condition of the first parameter is adjusted by using the constraint adjustment coefficient, and based on the adjusted linear constraint condition, the parameter optimization prediction model is re-solved to obtain the target prediction results of the associated input parameter and the non-associated input parameter;

[0126] Analyze the target prediction result of the associated input parameter and the corresponding first parameter adjustment range, determine the optimized value of the associated input parameter and output it;

[0127] If the non-correlated-target prediction result of the non-correlated input parameter falls within the second parameter adjustment range, the current non-correlated-target prediction result is used as the non-correlated input parameter optimization value;

[0128] If the non-correlated-target prediction result of the current non-correlated input parameter does not fall within the second parameter adjustment range, then the absolute differences between the non-correlated-target prediction result and the second parameter adjustment upper limit and the second parameter adjustment lower limit are calculated respectively, and the first absolute difference and the second absolute difference are obtained correspondingly;

[0129] If the first absolute difference is greater than the second absolute difference, the second parameter adjustment upper limit is used as the non-correlated input parameter optimization value;

[0130] Otherwise, the lower limit of the second parameter adjustment is used as the non-correlated input parameter optimization value;

[0131] A target adjustment strategy is generated based on the associated input parameter optimization values ​​of the associated input parameters of each key control parameter and the non-associated input parameter optimization values ​​of the non-associated input parameters.

[0132] In this embodiment, the grade labeling results include primary, secondary, tertiary and normal levels; the first parameter adjustment range refers to the allowed range of parameter values ​​of the associated input parameters; the second parameter adjustment range refers to the allowed range of parameter values ​​of the non-associated input parameters; the associated-prediction result refers to the predicted value of the associated input parameter at the current sampling moment obtained by analyzing the control parameter prediction result using the target optimization algorithm.

[0133] In this embodiment, the optimized value of the associated input parameter refers to the optimal value of the associated input parameter that meets the prediction results of the key control parameters with the safety of the unit and equipment as the boundary and the optimal economy as the goal; the first parameter refers to the associated input parameter whose associated-target prediction result does not fall within the adjustment range of the first parameter; the constraint adjustment coefficient is used to adjust and optimize the linear constraint conditions; the non-associated-prediction result refers to the predicted value of the non-associated input parameter at the current sampling time obtained by analyzing the control parameter prediction result using the target optimization algorithm; the optimized value of the non-associated input parameter refers to the optimal value of the non-associated input parameter that meets the prediction results of the key control parameters with the safety of the unit and equipment as the boundary and the optimal economy as the goal.

[0134] In this embodiment, the first absolute difference refers to the absolute difference between the non-correlated prediction result and the second parameter adjustment upper limit, wherein the second parameter adjustment upper limit refers to the maximum value allowed by the parameter; the second absolute difference refers to the absolute difference between the non-correlated prediction result and the second parameter adjustment lower limit, wherein the second parameter adjustment lower limit refers to the minimum value allowed by the parameter.

[0135] The beneficial effect of the above technical solution is: by combining the control parameter adjustment strategy of the associated input parameters and non-associated input parameters of the key control parameters with the target prediction results determined by the reference target optimization algorithm, a target adjustment strategy is generated to act on the unit, which can effectively ensure the safe operation of the unit and equipment, with the goal of optimal economy, and realize intelligent and flexible control of the key control loops of the unit.

[0136] The embodiment of the present invention provides an optimization method for improving the flexible operation of a thermal power unit. The calculation formula of the constraint adjustment coefficient is as follows:

[0137] In the formula, P is the constraint optimization coefficient; Δf is the linear constraint adjustment factor; c1 is the corresponding first parameter adjustment upper limit of the current first parameter; c0 is the corresponding first parameter adjustment lower limit of the current first parameter; Δc1 is the absolute difference between the association-target prediction result of the current first parameter and the corresponding first parameter adjustment upper limit; Δc0 is the absolute difference between the association-target prediction result of the current first parameter and the corresponding first parameter adjustment lower limit; α is the influence weight of the current first parameter on the safety of the unit; β is the contribution compensation coefficient of the current first parameter to the economy of the unit.

[0138] The beneficial effect of the above technical solution is: by calculating the constraint adjustment coefficient, the linear constraint conditions of the associated input parameters whose associated-target prediction results do not fall within the corresponding first parameter adjustment range are adjusted, which helps to improve the performance of the parameter optimization prediction model and achieve precise control of key control parameters.

[0139] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An optimization method for improving the flexible operation of a thermal power unit, characterized in that: include: Step 1: monitor the operating status of the target unit in real time to obtain first operating status data; Step 2: Based on the first operating status data, analyzing the operating status of the current target unit to obtain a unit operating status result; Step 3: using a parameter prediction model constructed using a machine learning algorithm, predicting key control parameters of the target unit based on the first operating state data to obtain control parameter prediction results; Step 4: Combine the control parameter prediction results with the unit operation status results for analysis, generate a target adjustment strategy to act on the target unit, and maximize the unit benefit.

2. The optimization method for improving the flexible operation of a thermal power unit according to claim 1, characterized in that: The first operating status data includes power generation power, efficiency, pressure and temperature, operating load, fuel consumption and emission, unit flow and operating time.

3. The optimization method for improving the flexible operation of a thermal power unit according to claim 1, characterized in that: Based on the first operating status data, the operating status of the current target unit is analyzed to obtain the unit operating status result, including: Obtaining an operation data threshold table of each operation system in the target unit; Based on the operation data threshold table, comparing the first operation status data of each operation system with the corresponding operation data threshold to obtain a first operation difference; Compare and analyze the first operation difference with a preset difference threshold to determine a grade labeling result of the first operation status data; Calculating a system risk coefficient of the current operating system based on the level labeling result of the first operating status data of each operating system; The calculation formula of the system risk coefficient is as follows: Where X represents the system risk coefficient of the current operating system; n i It represents the number of running status data of the i-th state level in the current running system, where i = 1, 2, 3; n0 represents the total number of running status data of the current running system; Δs ij It is represented by the first operation difference between the jth operation status data of the ith state level in the current operation system and the corresponding operation data threshold, where j = 1, 2, 3, ..., n i ;μ ij It is represented as the risk impact weight of the current operating system corresponding to the j-th operating status data of the i-th status level in the current operating system; The system risk coefficient of each operating system of the target unit is aggregated to generate a unit operating status result.

4. The optimization method for improving the flexible operation of a thermal power unit according to claim 1, characterized in that: Using a parameter prediction model constructed using a machine learning algorithm, the key control parameters of the target unit are predicted based on the first operating state data to obtain control parameter prediction results, including: Extracting historical operating status data, equipment parameters, and environmental condition data of the target unit, and performing data preprocessing to obtain a first data set; Dividing the first data set according to a set ratio to obtain a training data set and a test data set; Extracting the correlation influence characteristics of the key control parameters of the target unit, and combining the training data set as input, training a neural network to obtain an initial prediction model; Using the test data set to perform performance evaluation on the initial prediction model, and based on the performance evaluation results, optimizing the model to obtain a parameter prediction model; The first operating state data is input into the parameter prediction model to obtain a control parameter prediction result.

5. The optimization method for improving the flexible operation of a thermal power unit according to claim 1, characterized in that: The control parameter prediction results are combined with the unit operation status results for analysis, and a target adjustment strategy is generated to act on the target unit to maximize the unit benefit, including: Determine the safety risk level of each operating system at the current moment according to the system risk coefficient in the unit operation status result; Extracting the first state data-adjustment strategy reference table of each running system at the current moment based on the security risk level from the risk level-adjustment strategy database; Determine the data adjustment strategies for the first operating state data marked at different levels in the operating system from the first state data-adjustment strategy reference table, and output them as reference adjustment strategies; Determining control input parameters for controlling key control parameters of the target unit; The parameters directly related to the unit economy among the control input parameters are regarded as associated input parameters, and the parameters not directly related to the unit economy are regarded as non-associated input parameters; According to the reference adjustment strategy, respectively obtain the control parameter adjustment strategy of the associated input parameter and the non-associated input parameter of each key control parameter; Based on the control parameter prediction results, the target prediction results of the associated input parameters and the non-associated input parameters of the key control parameters are obtained by analyzing the target optimization algorithm; The parameter adjustment strategy and target prediction results of the associated input parameters and non-associated input parameters of each key control parameter are combined and analyzed to generate a target adjustment strategy.

6. The optimization method for improving the flexible operation of a thermal power unit according to claim 5, characterized in that: Based on the control parameter prediction results, the target prediction results of the associated input parameters and non-associated input parameters of the key control parameters are obtained by using the target optimization algorithm analysis, including: Obtain technical performance indicators and economic performance indicators of key control parameters; The associated input parameters and non-associated input parameters of the key control parameters are used as input variables, and the output variable is the comprehensive cost of the control loop. The control parameter-economic model is established by combining the corresponding technical performance indicators and economic performance indicators. Based on the control parameter-economic model, the lowest hourly comprehensive cost of the unit is taken as the objective function, the input variables are used as linear constraints, and a target optimization algorithm is used to build a parameter optimization prediction model; Solving the parameter optimization prediction model to obtain first prediction result sequences of associated input parameters and non-associated input parameters of the current key control parameters within a set prediction time period; The prediction result at the current sampling moment is extracted from the first prediction result sequence and output as the target prediction result of the associated input parameters and the non-associated input parameters.

7. The optimization method for improving the flexible operation of a thermal power unit according to claim 5, characterized in that: The parameter adjustment strategy and target prediction results of the associated input parameters and non-associated input parameters of each key control parameter are combined and analyzed to generate a target adjustment strategy, including: According to the combined analysis of the control parameter adjustment strategy and the grade annotation results of the associated input parameters, a first parameter adjustment range is obtained; Obtaining a second parameter adjustment range based on a combined analysis of a control parameter adjustment strategy and a grade annotation result for non-correlated input parameters; When the association-target prediction results of all the associated input parameters of the key control parameter belong to the corresponding first parameter adjustment range, the association-target prediction results are output as the associated input parameter optimization values ​​of the corresponding associated input parameters; When the key control parameter has a single or multiple associated input parameters whose associated-target prediction results do not fall within the first parameter adjustment range, the corresponding associated input parameters are marked as first parameters; The linear constraint condition of the first parameter is adjusted by using the constraint adjustment coefficient, and based on the adjusted linear constraint condition, the parameter optimization prediction model is re-solved to obtain the target prediction results of the associated input parameter and the non-associated input parameter; Analyze the target prediction result of the associated input parameter and the corresponding first parameter adjustment range, determine the optimized value of the associated input parameter and output it; If the non-correlated-target prediction result of the non-correlated input parameter falls within the second parameter adjustment range, the current non-correlated-target prediction result is used as the non-correlated input parameter optimization value; If the non-correlated-target prediction result of the current non-correlated input parameter does not fall within the second parameter adjustment range, then the absolute differences between the non-correlated-target prediction result and the second parameter adjustment upper limit and the second parameter adjustment lower limit are calculated respectively, and the first absolute difference and the second absolute difference are obtained correspondingly; If the first absolute difference is greater than the second absolute difference, the second parameter adjustment upper limit is used as the non-correlated input parameter optimization value; Otherwise, the lower limit of the second parameter adjustment is used as the non-correlated input parameter optimization value; A target adjustment strategy is generated based on the associated input parameter optimization values ​​of the associated input parameters of each key control parameter and the non-associated input parameter optimization values ​​of the non-associated input parameters.

8. The optimization method for improving the flexible operation of a thermal power unit according to claim 7, characterized in that: The constraint adjustment coefficient is calculated as follows: In the formula, P represents the constraint optimization coefficient; Δf represents the linear constraint adjustment factor; c1 represents the first parameter adjustment upper limit corresponding to the current first parameter; c0 represents the first parameter adjustment lower limit corresponding to the current first parameter; Δc1 represents the absolute difference between the association-target prediction result of the current first parameter and the upper limit of the corresponding first parameter adjustment; Δc0 is represented by the absolute difference between the association-target prediction result of the current first parameter and the corresponding first parameter adjustment lower limit; α is represented by the influence weight of the current first parameter on the safety of the unit; β is represented by the contribution compensation coefficient of the current first parameter to the economy of the unit.

Citation Information

Patent Citations

  • Multiple-target operation optimizing and coordinating control method and device of garbage power generator

    CN103064289A

  • Foreseeable predictive control method of thermal power generating unit coordination system

    CN111290282A

  • Thermal power plant intelligent combustion control method based on cloud data and cloud computing

    CN112859780A

  • Optimization control method and device of system and electronic equipment

    CN113759708A

  • Scheduling method and system of combined heat and power generation system

    CN114610215A

Cited By

  • Operation optimization method, device and equipment of thermal power generating unit, medium and product

    CN120233683A

  • A method, device, equipment, medium and product for optimizing operation of thermal power units

    CN120233683B