Intelligent optimization control method for boiler combustion

Abstract

The invention provides an intelligent optimization control method for boiler combustion, and relates to the technical field of optimization control, and the method comprises the steps: collecting boiler operation data of a target boiler based on a boiler DCS system, and carrying out the data processing, and obtaining first processing data; determining an optimization learning algorithm based on the boiler characteristics of the target boiler so as to obtain a combustion optimization model of the target boiler; performing parameter optimization on the combustion parameters of the target boiler based on the combustion optimization model, and performing optimization control on the target boiler based on the optimized combustion parameters; and boiler operation data of the target boiler after optimization control are monitored in real time, so that the optimization control effect of the target boiler is judged, and control optimization is carried out. The combustion optimization model of the target boiler is optimized by combining the boiler characteristics, so that the combustion optimization control of the target boiler is more timely and effective, and the optimization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of optimization control, and in particular to a boiler combustion intelligent optimization control method. Background Art

[0002] At present, big data and artificial intelligence are the major strategies for the development of national science and technology. China is actively carrying out the construction of intelligent power plants, with the goal of adopting advanced control strategies and technologies to achieve fine control of important process parameters and ensure safe, economical and environmentally friendly operation of the units. Among them, boiler combustion optimization is one of the key points of intelligent power plant construction, which is of great significance to energy conservation and emission reduction of thermal power units.

[0003] However, the existing intelligent optimization control methods for boiler combustion have problems such as low optimization efficiency and insufficient optimization.

[0004] Therefore, the present invention provides a boiler combustion intelligent optimization control method. Summary of the invention

[0005] The present invention provides a boiler combustion intelligent optimization control method, which is used to solve the problems of low optimization efficiency, insufficient optimization, etc. in the prior art.

[0006] The present invention provides a boiler combustion intelligent optimization control method, comprising:

[0007] Step 1: Collect boiler operation data of a target boiler based on a boiler DCS system, and process the data to obtain first processed data;

[0008] Step 2: Determine an optimization learning algorithm based on the boiler characteristics of the target boiler, thereby obtaining a combustion optimization model of the target boiler;

[0009] Step 3: Optimizing the combustion parameters of the target boiler based on the combustion optimization model, and optimizing the control of the target boiler based on the optimized combustion parameters;

[0010] Step 4: Monitor the boiler operation data of the target boiler after optimized control in real time to determine the optimized control effect of the target boiler and perform control optimization.

[0011] The boiler DCS system provided by the present invention collects boiler operation data of a target boiler and processes the data to obtain first processed data, including:

[0012] Step 11: extracting key operating data of the target boiler based on the boiler DCS system to obtain first operating data;

[0013] Step 12: Perform data cleaning and data standardization on the first operation data to obtain initial processed data;

[0014] Step 13: Determine the data structure of the initial processed data, and adjust the structure so that the data structure of each data in the initial processed data is consistent, thereby obtaining the first processed data.

[0015] According to the present invention, an optimization learning algorithm is determined based on the boiler characteristics of a target boiler, thereby obtaining a combustion optimization model of the target boiler, including:

[0016] Step 21: analyzing the boiler characteristics of the target boiler, and extracting the optimization learning algorithm with the highest degree of consistency with the target boiler from the algorithm database based on the boiler characteristics as the first learning algorithm;

[0017] Step 22: using the first processed data as a training sample to perform algorithm training on the first learning algorithm, thereby constructing an initial combustion optimization model of the target boiler based on the training result;

[0018] Step 23: Randomly extract historical operation data of the target boiler in the historical operation process, and perform model verification on the initial combustion optimization model, and obtain the combustion optimization model of the target boiler based on the verification optimization results.

[0019] According to the present invention, historical operation data of a target boiler in a historical operation process is randomly extracted, and the initial combustion optimization model is verified, and the combustion optimization model of the target boiler is obtained based on the verification optimization result, including:

[0020] Step 231: randomly extracting historical operation data of the target boiler during its historical operation, which data should include operating parameters and combustion performance indicators of the boiler under different operating conditions;

[0021] Step 232: Use historical operation data to verify the initial combustion optimization model and compare the difference between the model prediction results and the actual operation results;

[0022] Step 233: Based on the verification results, evaluate the accuracy, stability and applicability of the model;

[0023] If the model performs poorly, the first learning algorithm is retrained to obtain an initial combustion optimization model, and the model is revalidated;

[0024] If the model performs well, it will be used as the combustion optimization model of the target boiler.

[0025] According to the present invention, the combustion parameters of the target boiler are optimized based on the combustion optimization model, and the target boiler is optimized and controlled based on the optimized combustion parameters, including:

[0026] Step 31: Input the real-time boiler operation data into the combustion optimization model to obtain the combustion efficiency and emission performance predicted by the model;

[0027] Step 32: performing a first comparison between the combustion efficiency predicted by the model and the minimum combustion efficiency of the target boiler, thereby determining a first combustion performance of the combustion optimization model based on the first comparison result;

[0028] Step 33: performing a second comparison between the emission performance predicted by the model and the standard emission performance of the target boiler, thereby determining a second combustion performance of the combustion optimization model based on the second comparison result;

[0029] Step 34: combining the first combustion performance and the second combustion performance to obtain a comprehensive prediction result of the target boiler;

[0030] Step 35: According to the comprehensive prediction results, the combustion parameters are optimized one by one in combination with the optimization algorithm to obtain the combustion parameter combination of each optimization result, thereby obtaining the combustion parameter set of the target boiler;

[0031] Step 36: performing comparison based on each combustion parameter combination in the combustion parameter set of the target boiler, thereby extracting the optimal combustion parameter combination in the combustion parameter set as the initial combustion parameter combination;

[0032] Step 37: Combining the initial combustion parameter combination with the boiler parameters of the target boiler to perform combustion simulation, thereby determining the corresponding combustion performance;

[0033] If the combustion performance is higher than the preset minimum combustion performance, the initial combustion parameter combination is used as the optimal combustion parameter combination;

[0034] Step 38: Send the combustion parameters in the optimal combustion parameter combination to the boiler DCS system, thereby achieving parameter optimization of the combustion parameters of the target boiler and achieving optimized control of the target boiler.

[0035] The fuel control parameters provided by the present invention include: supply quantity, air supply quantity, and air door opening parameters.

[0036] According to the present invention, a combustion parameter in an optimal combustion parameter combination is sent to a boiler DCS system, thereby optimizing the combustion parameters of a target boiler and optimizing the control of the target boiler, including:

[0037] Step 381: converting each parameter value in the optimal combustion parameter combination into a format and range that can be recognized by the DCS system;

[0038] Step 382: Sending the optimal combustion parameters to the boiler DCS system based on a preset communication interface;

[0039] Step 383: Update the corresponding parameter setting values ​​in the DCS system to achieve parameter optimization of the combustion parameters.

[0040] According to the present invention, the boiler operation data of the target boiler after real-time monitoring and optimization control is determined, thereby determining the optimization control effect of the target boiler and performing control optimization, including:

[0041] Step 41: Using the DCS system to monitor the boiler operation data of the target boiler in real time to obtain real-time monitoring data;

[0042] Step 42: Compare the real-time monitoring data with the data before optimization to analyze whether the optimization control effect reaches the expected goal;

[0043] Step 43: If the optimization control effect is not ideal, the combustion optimization model is adjusted and optimized according to the real-time monitoring data to improve the accuracy and adaptability of the model;

[0044] Step 44: Re-optimize the combustion parameters according to the adjusted combustion optimization model, and implement optimization control again.

[0045] If the optimization control effect provided by the present invention is not ideal, the combustion optimization model is adjusted and optimized according to the real-time monitoring data to improve the accuracy and adaptability of the model, including:

[0046] Step 431: If the optimization control effect is not ideal, based on the construction process of the combustion optimization model, the risk type of each sub-construction process is obtained to obtain a first risk set;

[0047] Step 432: Determine the first construction risk of the combustion optimization model based on the real-time monitoring data corresponding to each risk type in the first risk set, so as to adjust and optimize the combustion optimization model:

[0048] Step 433: verifying and testing the adjusted combustion optimization model;

[0049] If the adjusted model still cannot achieve the expected results, it is necessary to reconsider the model construction method or seek other optimization strategies.

[0050] According to the combustion optimization model after adjustment provided by the present invention, the combustion parameters are optimized again, and the optimization control is implemented again, including:

[0051] Step 441: resetting the target value or range of the combustion parameter based on the adjusted combustion optimization model;

[0052] Step 442: Optimize and calculate the combustion parameters using an optimization algorithm or tool to obtain an optimal combination of combustion parameters;

[0053] Step 443: applying the optimized combustion parameters to the target boiler and performing real-time monitoring;

[0054] If the performance index of the target boiler corresponding to the optimized combustion parameters is higher than the performance index of the target boiler corresponding to the combustion parameters before optimization, the operating parameters corresponding to the adjusted combustion optimization model are used as new operating parameters;

[0055] If the performance index of the target boiler corresponding to the optimized combustion parameters is not higher than the performance index of the target boiler corresponding to the combustion parameters before optimization, continue to adjust and optimize.

[0056] Compared with the prior art, the beneficial effects of the present invention are as follows: a boiler combustion intelligent optimization control method provided by the present invention optimizes the combustion optimization model of the target boiler in combination with the boiler characteristics, thereby making the combustion optimization control of the target boiler more timely and effective, and improving the optimization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 It is a flow chart of a boiler combustion intelligent optimization control method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] Embodiment 1:

[0061] The embodiment of the present invention provides a method for intelligent optimization control of boiler combustion, such as Figure 1 As shown, including:

[0062] Step 1: Collect boiler operation data of a target boiler based on a boiler DCS system, and process the data to obtain first processed data;

[0063] Step 2: Determine an optimization learning algorithm based on the boiler characteristics of the target boiler, thereby obtaining a combustion optimization model of the target boiler;

[0064] Step 3: Optimizing the combustion parameters of the target boiler based on the combustion optimization model, and optimizing the control of the target boiler based on the optimized combustion parameters;

[0065] Step 4: Monitor the boiler operation data of the target boiler after optimized control in real time to determine the optimized control effect of the target boiler and perform control optimization.

[0066] In this embodiment, the boiler DCS system is a distributed control system. In boiler control, the boiler DCS is used to centrally or distributedly monitor and control various operating parameters of the boiler, such as temperature, pressure, flow, liquid level, etc., to ensure safe, efficient and stable operation of the boiler.

[0067] In this embodiment, boiler operation data refers to various parameters and status information generated by the boiler during operation, including but not limited to fuel consumption, feed water flow, steam pressure, steam temperature, flue gas composition, exhaust temperature, furnace temperature, etc.

[0068] In this embodiment, data processing refers to the process of cleaning, sorting, converting and analyzing boiler operation data to obtain first processed data that can be used for subsequent analysis and optimization. Data processing includes steps such as data denoising, data smoothing, and data standardization.

[0069] In this embodiment, boiler characteristics refer to various characteristics and performances exhibited by the boiler during the design, manufacture and use process, such as the capacity, thermal efficiency, fuel adaptability, load regulation capability, etc. of the boiler.

[0070] In this embodiment, the optimization learning algorithm refers to an algorithm that improves system performance through learning and optimization processes. In boiler combustion optimization, the optimization learning algorithm can automatically adjust combustion parameters according to the operating data and characteristics of the boiler to achieve the goals of maximizing combustion efficiency and minimizing pollutant emissions. Common optimization learning algorithms include genetic algorithms, particle swarm algorithms, neural networks, etc.

[0071] In this embodiment, the combustion optimization model is established based on boiler characteristics and optimization learning algorithms. It is a mathematical model used to describe the relationship between various parameters in the boiler combustion process. The combustion optimization model can predict and optimize combustion parameters based on the boiler's operating data and optimization objectives to achieve efficient and clean operation of the boiler.

[0072] In this embodiment, the combustion parameters refer to various parameters that affect the combustion efficiency and pollutant emissions of the boiler, such as fuel supply, air flow, furnace temperature, flue gas oxygen content, etc. By adjusting these parameters, the combustion process of the boiler can be optimized and the thermal efficiency and environmental performance can be improved.

[0073] In this embodiment, optimization control refers to the process of adjusting and controlling the combustion parameters of the boiler in real time based on the combustion optimization model and optimization algorithm. Optimization control can ensure that the boiler can maintain the best operating state under various working conditions and achieve efficient, stable and environmentally friendly combustion.

[0074] In this embodiment, real-time monitoring refers to the process of collecting and analyzing boiler operation data in real time to obtain the current operation status and performance parameters of the boiler.

[0075] In this embodiment, control optimization refers to the process of further adjusting and improving the combustion parameters and control strategies based on the real-time monitored boiler operation data and optimized control effects. Control optimization can continuously improve the operation efficiency and environmental performance of the boiler and reduce the operation cost.

[0076] The beneficial effect of the above technical solution is that by optimizing the combustion optimization model of the target boiler in combination with the boiler characteristics, the combustion optimization control of the target boiler can be made more timely and effective, thereby improving the optimization efficiency.

[0077] Embodiment 2:

[0078] Based on Example 1, the boiler operation data of the target boiler is collected based on the boiler DCS system, and the data is processed to obtain first processed data, including:

[0079] Step 11: extracting key operating data of the target boiler based on the boiler DCS system to obtain first operating data;

[0080] Step 12: Perform data cleaning and data standardization on the first operation data to obtain initial processed data;

[0081] Step 13: Determine the data structure of the initial processed data, and adjust the structure so that the data structure of each data in the initial processed data is consistent, thereby obtaining the first processed data.

[0082] In this embodiment, key operation data refers to data that has a significant impact on boiler performance, safety and economy during boiler operation. Key operation data generally include boiler fuel consumption, feed water flow, steam pressure, steam temperature, flue gas composition (such as oxygen content, carbon dioxide content, nitrogen oxide content, etc.), exhaust gas temperature, furnace temperature, etc.

[0083] In this embodiment, the first operating data refers to an original set of key operating data extracted from the boiler DCS system.

[0084] In this embodiment, data cleaning refers to the process of processing raw data to remove or correct errors, outliers or missing values. Data cleaning is an important step in data preprocessing, which can ensure the accuracy and reliability of data and provide a good foundation for subsequent data analysis or model building.

[0085] In this embodiment, data standardization processing refers to converting data according to certain rules so that the data has a unified dimension or range for comparison and analysis. Data standardization processing can eliminate the dimensional differences between different data, making the data easier to process and analyze.

[0086] In this embodiment, the initial processing data is a data set obtained after data cleaning and data standardization. These data have been free of noise, outliers and missing values, and have a unified dimension or range, and can be used for subsequent data analysis or model building.

[0087] In this embodiment, data structure refers to the format and organization used when data is stored and represented. The data structure determines the access speed, storage efficiency and maintainability of the data. In data analysis and processing, choosing a suitable data structure is crucial to improving processing efficiency and accuracy.

[0088] In this embodiment, structural adjustment refers to adjusting the storage or representation format of data to make it more suitable for subsequent analysis or processing steps.

[0089] In this embodiment, the first processed data refers to a data set obtained after data cleaning, data standardization and structure adjustment. The first processed data has a consistent data structure and has removed noise, outliers and missing values, and can be used for subsequent optimization control or data analysis.

[0090] The beneficial effect of the above technical solution is: by processing the boiler operation data, building a combustion optimization model in combination with the boiler characteristics, and optimizing the model, the combustion optimization control of the target boiler can be made more accurate and the optimization efficiency can be improved.

[0091] Embodiment 3:

[0092] Based on Example 2, an optimization learning algorithm is determined based on the boiler characteristics of the target boiler, thereby obtaining a combustion optimization model of the target boiler, including:

[0093] Step 21: analyzing the boiler characteristics of the target boiler, and extracting the optimization learning algorithm with the highest degree of consistency with the target boiler from the algorithm database based on the boiler characteristics as the first learning algorithm;

[0094] Step 22: using the first processed data as a training sample to perform algorithm training on the first learning algorithm, thereby constructing an initial combustion optimization model of the target boiler based on the training result;

[0095] Step 23: Randomly extract historical operation data of the target boiler in the historical operation process, and perform model verification on the initial combustion optimization model, and obtain the combustion optimization model of the target boiler based on the verification optimization results.

[0096] In this embodiment, the boiler characteristics of the target boiler refer to various characteristics and performances exhibited by the target boiler during the design, manufacturing and use process, including but not limited to the boiler capacity, thermal efficiency, fuel adaptability, load regulation capability, combustion mode, flue gas emission characteristics, etc. These characteristics are important bases for selecting and optimizing the combustion control algorithm.

[0097] In this embodiment, the algorithm database refers to a database storing a variety of optimization learning algorithms. The algorithm database may include genetic algorithms, particle swarm algorithms, neural networks, support vector machines, etc., which are used to solve optimization problems in different fields. In boiler combustion optimization, the algorithm database provides a variety of possible optimization algorithms for selection.

[0098] In this embodiment, the optimization learning algorithm refers to an algorithm that improves system performance through learning and optimization processes. In boiler combustion optimization, the optimization learning algorithm can automatically adjust combustion parameters according to the operating data and characteristics of the boiler to achieve goals such as maximizing combustion efficiency and minimizing pollutant emissions.

[0099] In this embodiment, the degree of fit refers to the degree of match between the algorithm and the target boiler characteristics. When selecting an optimization learning algorithm, it is necessary to consider whether the algorithm is suitable for the characteristics and requirements of the target boiler, and how the algorithm performs in similar scenarios. The higher the degree of fit, the better the optimization effect of the algorithm on the target boiler may be.

[0100] In this embodiment, the first learning algorithm refers to an optimization learning algorithm extracted from an algorithm database that has the highest degree of consistency with the target boiler, and is used for subsequent training and construction of a combustion optimization model.

[0101] In this embodiment, the training sample is a data set used to train a machine learning or optimization algorithm. In boiler combustion optimization, the training sample usually includes boiler operation data (such as fuel consumption, feed water flow, steam pressure, steam temperature, etc.) and corresponding combustion parameters (such as fuel supply, air flow, etc.).

[0102] In this embodiment, algorithm training refers to the process of training the optimization learning algorithm using training samples. Through training, the algorithm can learn the associations and rules between data, so as to predict or optimize the output according to new input data.

[0103] In this embodiment, the initial combustion optimization model refers to a combustion optimization model that is constructed based on training results and has not yet been verified. The model can predict or optimize combustion parameters based on input operating data, but may have certain errors or deficiencies.

[0104] In this embodiment, the historical operation data refers to the data recorded during the past operation of the target boiler. These data can be used to verify the accuracy and reliability of the combustion optimization model and evaluate the performance of the model in actual application.

[0105] In this embodiment, model verification refers to the process of verifying the initial combustion optimization model using historical operation data. Through verification, the prediction accuracy, optimization effect and applicability of the model in practical applications can be evaluated.

[0106] In this embodiment, the validation optimization result refers to the optimization result obtained during the model validation process, including the comparison between the combustion parameters predicted by the model and the actual operation data, the degree of improvement of the model on combustion efficiency and pollutant emissions, etc.

[0107] In this embodiment, the combustion optimization model of the target boiler refers to a combustion optimization model that is obtained after verification and optimization and is suitable for the target boiler. It can accurately predict or optimize combustion parameters based on input operating data to improve the combustion efficiency and environmental performance of the boiler.

[0108] The beneficial effect of the above technical solution is that by optimizing the combustion optimization model of the target boiler in combination with the boiler characteristics, the combustion optimization control of the target boiler can be made more timely and effective.

[0109] Embodiment 4:

[0110] Based on Example 3, a combustion optimization model of a target boiler is obtained based on the verification optimization results, including:

[0111] Step 231: randomly extracting historical operation data of the target boiler during its historical operation, which data should include operating parameters and combustion performance indicators of the boiler under different operating conditions;

[0112] Step 232: Use historical operation data to verify the initial combustion optimization model and compare the difference between the model prediction results and the actual operation results;

[0113] Step 233: Based on the verification results, evaluate the accuracy, stability and applicability of the model;

[0114] If the model performs poorly, the first learning algorithm is retrained to obtain an initial combustion optimization model, and the model is revalidated;

[0115] If the model performs well, it will be used as the combustion optimization model of the target boiler.

[0116] In this embodiment, historical operation data refers to data recorded during the operation of the target boiler in the past period of time (may be several months or years). These data include operating parameters (such as fuel consumption, feed water flow, steam pressure, steam temperature, etc.) and combustion performance indicators (such as combustion efficiency, pollutant emission concentration, etc.) of the boiler under different working conditions (such as load changes, fuel type changes, etc.).

[0117] In this embodiment, the initial combustion optimization model refers to a combustion optimization model that is obtained based on machine learning or optimization algorithm training and has not been fully verified. The model can predict or optimize combustion performance indicators based on input operating parameters.

[0118] In this embodiment, model validation refers to the process of testing and evaluating the initial combustion optimization model using historical operating data. By comparing the differences between the model prediction results and the actual operating results, the accuracy, stability and applicability of the model can be evaluated.

[0119] In this embodiment, accuracy refers to the degree of consistency between the model prediction results and the actual operation results. The higher the accuracy, the more accurate the model's description of the boiler combustion process is.

[0120] In this embodiment, stability refers to the degree of consistency of the model under different working conditions and conditions. A model with good stability can maintain stable prediction and optimization effects under various working conditions.

[0121] In this embodiment, applicability refers to the feasibility and effectiveness of the model in practical applications. A model with good applicability can be flexibly adjusted and optimized according to actual operation data and requirements.

[0122] The beneficial effect of the above technical solution is: by performing performance evaluation on the combustion optimization model of the target boiler and optimizing parameters based on the evaluation verification, the combustion optimization model of the target boiler can be made more accurate, thereby making the combustion optimization control of the target boiler more timely and effective, thereby improving the optimization efficiency.

[0123] Embodiment 5:

[0124] Based on Example 3, the combustion parameters of the target boiler are optimized based on the combustion optimization model, and the target boiler is optimized and controlled based on the optimized combustion parameters, including:

[0125] Step 31: Input the real-time boiler operation data into the combustion optimization model to obtain the combustion efficiency and emission performance predicted by the model;

[0126] Step 32: performing a first comparison between the combustion efficiency predicted by the model and the minimum combustion efficiency of the target boiler, thereby determining a first combustion performance of the combustion optimization model based on the first comparison result;

[0127] Step 33: performing a second comparison between the emission performance predicted by the model and the standard emission performance of the target boiler, thereby determining a second combustion performance of the combustion optimization model based on the second comparison result;

[0128] Step 34: combining the first combustion performance and the second combustion performance to obtain a comprehensive prediction result of the target boiler;

[0129] Step 35: According to the comprehensive prediction results, the combustion parameters are optimized one by one in combination with the optimization algorithm to obtain the combustion parameter combination of each optimization result, thereby obtaining the combustion parameter set of the target boiler;

[0130] Step 36: performing comparison based on each combustion parameter combination in the combustion parameter set of the target boiler, thereby extracting the optimal combustion parameter combination in the combustion parameter set as the initial combustion parameter combination;

[0131] Step 37: Combining the initial combustion parameter combination with the boiler parameters of the target boiler to perform combustion simulation, thereby determining the corresponding combustion performance;

[0132] If the combustion performance is higher than the preset minimum combustion performance, the initial combustion parameter combination is used as the optimal combustion parameter combination;

[0133] Step 38: Send the combustion parameters in the optimal combustion parameter combination to the boiler DCS system, thereby achieving parameter optimization of the combustion parameters of the target boiler and achieving optimized control of the target boiler.

[0134] In this embodiment, the combustion efficiency refers to the ratio of the useful energy released during the fuel combustion process to the energy that can be released theoretically when the fuel is completely burned. The higher the combustion efficiency, the higher the fuel utilization rate and the higher the thermal efficiency of the boiler.

[0135] In this embodiment, the emission performance refers to the amount of pollutants emitted during the combustion process of the boiler, such as nitrogen oxides (NOx), sulfur oxides (SOx), particulate matter, etc. The emission performance directly reflects the environmental performance of the boiler.

[0136] In this embodiment, the optimization algorithm can automatically adjust the combustion parameters according to the prediction results and the set optimization goals to achieve the optimal combustion state.

[0137] In this embodiment, the combustion parameters refer to various parameters that affect the combustion process of the boiler, such as fuel flow, air flow, burner angle, furnace pressure, etc. By adjusting these parameters, the combustion state and performance of the boiler can be changed.

[0138] The beneficial effect of the above technical solution is that by optimizing the parameters of the combustion optimization model of the target boiler, the combustion optimization model of the target boiler can be made more accurate, thereby making the combustion optimization control of the target boiler more timely and effective, thereby improving the optimization efficiency.

[0139] Embodiment 6:

[0140] Based on the fifth embodiment, the fuel control parameters include: supply quantity, air supply quantity, and air door opening parameters.

[0141] The beneficial effect of the above technical solution is that by optimizing the parameters of the combustion optimization model of the target boiler, the combustion optimization model of the target boiler can be made more accurate, thereby making the combustion optimization control of the target boiler more timely and effective, thereby improving the optimization efficiency.

[0142] Embodiment 7:

[0143] Based on Example 5, the combustion parameters in the optimal combustion parameter combination are sent to the boiler DCS system, thereby realizing parameter optimization of the combustion parameters of the target boiler and realizing optimal control of the target boiler, including:

[0144] Step 381: converting each parameter value in the optimal combustion parameter combination into a format and range that can be recognized by the DCS system;

[0145] Step 382: Sending the optimal combustion parameters to the boiler DCS system based on a preset communication interface;

[0146] Step 383: Update the corresponding parameter setting values ​​in the DCS system to achieve parameter optimization of the combustion parameters.

[0147] The beneficial effect of the above technical solution is that by optimizing the combustion optimization model of the target boiler, the combustion optimization model of the target boiler can be made more accurate, thereby making the combustion optimization control of the target boiler more timely and effective.

[0148] Embodiment 8:

[0149] Based on Example 5, the boiler operation data of the target boiler after the optimized control is monitored in real time, so as to judge the optimized control effect of the target boiler, and then perform control optimization, including:

[0150] Step 41: Using the DCS system to monitor the boiler operation data of the target boiler in real time to obtain real-time monitoring data;

[0151] Step 42: Compare the real-time monitoring data with the data before optimization to analyze whether the optimization control effect reaches the expected goal;

[0152] Step 43: If the optimization control effect is not ideal, the combustion optimization model is adjusted and optimized according to the real-time monitoring data to improve the accuracy and adaptability of the model;

[0153] Step 44: Re-optimize the combustion parameters according to the adjusted combustion optimization model, and implement optimization control again.

[0154] The beneficial effect of the above technical solution is: by real-time monitoring of the boiler operation data of the target boiler, the optimization control result can be timely judged, and the control optimization of the target boiler can be performed, which can make the combustion optimization control of the target boiler more timely and effective, and improve the optimization efficiency.

[0155] Embodiment 9:

[0156] Based on Example 8, if the optimization control effect is not ideal, the combustion optimization model is adjusted and optimized according to the real-time monitoring data to improve the accuracy and adaptability of the model, including:

[0157] Step 431: If the optimization control effect is not ideal, based on the construction process of the combustion optimization model, the risk type of each sub-construction process is obtained to obtain a first risk set;

[0158] Step 432: Determine the first construction risk of the combustion optimization model based on the real-time monitoring data corresponding to each risk type in the first risk set, so as to adjust and optimize the combustion optimization model:

[0159] Step 433: verifying and testing the adjusted combustion optimization model;

[0160] If the adjusted model still cannot achieve the expected results, it is necessary to reconsider the model construction method or seek other optimization strategies.

[0161] The beneficial effect of the above technical solution is: by real-time monitoring of the boiler operation data of the target boiler, the optimization control result can be timely judged, and the control optimization of the target boiler can be performed, which can make the combustion optimization control of the target boiler more timely and effective, and improve the optimization efficiency.

[0162] Embodiment 10:

[0163] Based on Example 8, the combustion parameters are optimized again according to the adjusted combustion optimization model, and the optimization control is implemented again, including:

[0164] Step 441: resetting the target value or range of the combustion parameter based on the adjusted combustion optimization model;

[0165] Step 442: Optimize and calculate the combustion parameters using an optimization algorithm or tool to obtain an optimal combination of combustion parameters;

[0166] Step 443: applying the optimized combustion parameters to the target boiler and performing real-time monitoring;

[0167] If the performance index of the target boiler corresponding to the optimized combustion parameters is higher than the performance index of the target boiler corresponding to the combustion parameters before optimization, the operating parameters corresponding to the adjusted combustion optimization model are used as new operating parameters;

[0168] If the performance index of the target boiler corresponding to the optimized combustion parameters is not higher than the performance index of the target boiler corresponding to the combustion parameters before optimization, continue to adjust and optimize.

[0169] The beneficial effect of the above technical solution is: by real-time monitoring of the boiler operation data of the target boiler, the optimization control result can be timely judged, and the control optimization of the target boiler can be performed, which can make the combustion optimization control of the target boiler more timely and effective, and improve the optimization efficiency.

[0170] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A boiler combustion intelligent optimization control method, characterized in that: include: Step 1: Collect boiler operation data of a target boiler based on a boiler DCS system, and process the data to obtain first processed data; Step 2: Determine an optimization learning algorithm based on the boiler characteristics of the target boiler, thereby obtaining a combustion optimization model of the target boiler; Step 3: Optimizing the combustion parameters of the target boiler based on the combustion optimization model, and optimizing the control of the target boiler based on the optimized combustion parameters; Step 4: Monitor the boiler operation data of the target boiler after optimized control in real time to determine the optimized control effect of the target boiler and perform control optimization.

2. The method for intelligent optimization control of boiler combustion according to claim 1, characterized in that: The boiler operation data of the target boiler is collected based on the boiler DCS system, and the data is processed to obtain first processed data, including: Step 11: extracting key operating data of the target boiler based on the boiler DCS system to obtain first operating data; Step 12: Perform data cleaning and data standardization on the first operation data to obtain initial processed data; Step 13: Determine the data structure of the initial processed data, and adjust the structure so that the data structure of each data in the initial processed data is consistent, thereby obtaining the first processed data.

3. The method for intelligent optimization control of boiler combustion according to claim 2 is characterized in that: The optimization learning algorithm is determined based on the boiler characteristics of the target boiler, thereby obtaining the combustion optimization model of the target boiler, including: Step 21: analyzing the boiler characteristics of the target boiler, and extracting the optimization learning algorithm with the highest degree of consistency with the target boiler from the algorithm database based on the boiler characteristics as the first learning algorithm; Step 22: using the first processed data as a training sample to perform algorithm training on the first learning algorithm, thereby constructing an initial combustion optimization model of the target boiler based on the training result; Step 23: Randomly extract historical operation data of the target boiler in the historical operation process, and perform model verification on the initial combustion optimization model, and obtain the combustion optimization model of the target boiler based on the verification optimization results.

4. The method for intelligent optimization control of boiler combustion according to claim 3 is characterized in that: Randomly extract historical operation data of the target boiler in the historical operation process, and perform model verification on the initial combustion optimization model. Based on the verification optimization results, the combustion optimization model of the target boiler is obtained, including: Step 231: randomly extracting historical operation data of the target boiler during its historical operation, which data should include operating parameters and combustion performance indicators of the boiler under different operating conditions; Step 232: Use historical operation data to verify the initial combustion optimization model and compare the difference between the model prediction results and the actual operation results; Step 233: Based on the verification results, evaluate the accuracy, stability and applicability of the model; If the model performs poorly, the first learning algorithm is retrained to obtain an initial combustion optimization model, and the model is revalidated; If the model performs well, it will be used as the combustion optimization model of the target boiler.

5. The method for intelligent optimization control of boiler combustion according to claim 3 is characterized in that: The combustion parameters of the target boiler are optimized based on the combustion optimization model, and the target boiler is optimized and controlled based on the optimized combustion parameters, including: Step 31: Input the real-time boiler operation data into the combustion optimization model to obtain the combustion efficiency and emission performance predicted by the model; Step 32: performing a first comparison between the combustion efficiency predicted by the model and the minimum combustion efficiency of the target boiler, thereby determining a first combustion performance of the combustion optimization model based on the first comparison result; Step 33: performing a second comparison between the emission performance predicted by the model and the standard emission performance of the target boiler, thereby determining a second combustion performance of the combustion optimization model based on the second comparison result; Step 34: combining the first combustion performance and the second combustion performance to obtain a comprehensive prediction result of the target boiler; Step 35: According to the comprehensive prediction results, the combustion parameters are optimized one by one in combination with the optimization algorithm to obtain the combustion parameter combination of each optimization result, thereby obtaining the combustion parameter set of the target boiler; Step 36: performing comparison based on each combustion parameter combination in the combustion parameter set of the target boiler, thereby extracting the optimal combustion parameter combination in the combustion parameter set as the initial combustion parameter combination; Step 37: Combining the initial combustion parameter combination with the boiler parameters of the target boiler to perform combustion simulation, thereby determining the corresponding combustion performance; If the combustion performance is higher than the preset minimum combustion performance, the initial combustion parameter combination is used as the optimal combustion parameter combination; Step 38: Send the combustion parameters in the optimal combustion parameter combination to the boiler DCS system, thereby achieving parameter optimization of the combustion parameters of the target boiler and achieving optimized control of the target boiler.

6. The method for intelligent optimization control of boiler combustion according to claim 5 is characterized in that: Fuel control parameters include: supply quantity, air supply volume, and air door opening parameters.

7. The method for intelligent optimization control of boiler combustion according to claim 5, characterized in that: The combustion parameters in the optimal combustion parameter combination are sent to the boiler DCS system, thereby optimizing the combustion parameters of the target boiler and achieving optimal control of the target boiler, including: Step 381: converting each parameter value in the optimal combustion parameter combination into a format and range that can be recognized by the DCS system; Step 382: Sending the optimal combustion parameters to the boiler DCS system based on a preset communication interface; Step 383: Update the corresponding parameter setting values ​​in the DCS system to achieve parameter optimization of the combustion parameters.

8. The method for intelligent optimization control of boiler combustion according to claim 5, characterized in that: Real-time monitoring of the boiler operation data of the target boiler after optimized control, so as to judge the optimized control effect of the target boiler and perform control optimization, including: Step 41: Using the DCS system to monitor the boiler operation data of the target boiler in real time to obtain real-time monitoring data; Step 42: Compare the real-time monitoring data with the data before optimization to analyze whether the optimization control effect reaches the expected goal; Step 43: If the optimization control effect is not ideal, the combustion optimization model is adjusted and optimized according to the real-time monitoring data to improve the accuracy and adaptability of the model; Step 44: Re-optimize the combustion parameters according to the adjusted combustion optimization model, and implement optimization control again.

9. The method for intelligent optimization control of boiler combustion according to claim 8, characterized in that: If the optimization control effect is not ideal, the combustion optimization model will be adjusted and optimized according to the real-time monitoring data to improve the accuracy and adaptability of the model, including: Step 431: If the optimization control effect is not ideal, based on the construction process of the combustion optimization model, the risk type of each sub-construction process is obtained to obtain a first risk set; Step 432: Determine the first construction risk of the combustion optimization model based on the real-time monitoring data corresponding to each risk type in the first risk set, so as to adjust and optimize the combustion optimization model: Step 433: verifying and testing the adjusted combustion optimization model; If the adjusted model still cannot achieve the expected results, it is necessary to reconsider the model construction method or seek other optimization strategies.

10. The method for intelligent optimization control of boiler combustion according to claim 8, characterized in that: According to the adjusted combustion optimization model, the combustion parameters are re-optimized and the optimization control is implemented again, including: Step 441: resetting the target value or range of the combustion parameter based on the adjusted combustion optimization model; Step 442: Optimize and calculate the combustion parameters using an optimization algorithm or tool to obtain an optimal combination of combustion parameters; Step 443: applying the optimized combustion parameters to the target boiler and performing real-time monitoring; If the performance index of the target boiler corresponding to the optimized combustion parameters is higher than the performance index of the target boiler corresponding to the combustion parameters before optimization, the operating parameters corresponding to the adjusted combustion optimization model are used as new operating parameters; If the performance index of the target boiler corresponding to the optimized combustion parameters is not higher than the performance index of the target boiler corresponding to the combustion parameters before optimization, continue to adjust and optimize.

Images