Intelligent optimization method for boiler air and smoke system

Abstract

The invention provides an intelligent optimization method for a boiler air and smoke system, and relates to the technical field of boiler energy conservation, and the method comprises the steps: carrying out the anomaly detection and combustion response analysis of a system through first target data obtained after the collection and preprocessing of key data of a target boiler air and smoke system, and obtaining an anomaly detection result and a combustion response result; matching an undetermined exception handling strategy according to the exception detection result; obtaining an undetermined parameter optimization strategy based on the combustion response result; and combining the to-be-determined exception handling strategy with the to-be-determined parameter optimization strategy to generate an intelligent optimization strategy. First target data are obtained by collecting and preprocessing key data of a target boiler air and smoke system in real time, and anomaly detection and combustion response analysis are conducted on the target boiler air and smoke system based on the first target data. And then an intelligent optimization strategy is generated according to the obtained anomaly detection result and the combustion response result to optimize the current target boiler air and smoke system, effective control over the boiler air and smoke system is achieved, and the boiler combustion efficiency is improved.

Description

Technical Field

[0001] The invention relates to the technical field of boiler energy saving, and in particular to an intelligent optimization method for a boiler air and smoke system. Background Art

[0002] As an important heat energy conversion equipment, boilers play a vital role in industrial production. However, the control of traditional boiler air and smoke systems faces many challenges, such as uneven air volume distribution and incomplete combustion, which leads to low boiler combustion efficiency, high exhaust temperature and serious energy waste. Therefore, how to effectively control the boiler air and smoke system and improve the boiler combustion efficiency has become one of the current research focuses.

[0003] Therefore, the present invention provides an intelligent optimization method for a boiler air and smoke system. Summary of the invention

[0004] The present invention provides an intelligent optimization method for a boiler air and smoke system, which is used to obtain first target data by real-time acquisition and preprocessing of key data of a target boiler air and smoke system, perform abnormality detection and combustion response analysis on the target boiler air and smoke system based on the first target data, and then generate an intelligent optimization strategy according to the obtained abnormality detection results and combustion response results to optimize the current target boiler air and smoke system, thereby achieving effective control of the boiler air and smoke system and improving boiler combustion efficiency.

[0005] The present invention provides an intelligent optimization method for a boiler air and smoke system, comprising:

[0006] Step 1: collect key data of the target boiler air and smoke system in real time and perform data preprocessing to obtain the first target data;

[0007] Step 2: using the first target data to perform abnormality detection and combustion response analysis on the target boiler air and smoke system, and obtain abnormality detection results and combustion response results accordingly;

[0008] Step 3: According to the anomaly detection result, match the anomaly existing in the target boiler air and smoke system with a pending anomaly handling strategy;

[0009] Step 4: Based on the combustion response result, intelligent optimization analysis is performed on the target control parameters in the current target boiler air and smoke system to obtain an optimization strategy for the undetermined parameters;

[0010] Step 5: Combine the pending exception handling strategy with the pending parameter optimization strategy to generate an intelligent optimization strategy to optimize the current target boiler air and smoke system.

[0011] Preferably, the key data of the target boiler air and smoke system is collected in real time and preprocessed to obtain the first target data, including:

[0012] Utilize the preset sensors pre-installed at the set key positions of the current target boiler air and smoke system to collect the key data of the target boiler air and smoke system in real time;

[0013] The acquired key data is divided according to data types to obtain wind and smoke related data, and the data types are marked on the wind and smoke related data;

[0014] The wind-smoke correlation data is preprocessed to obtain first target data.

[0015] Preferably, the wind and smoke related data refers to the combustion parameter data, air volume parameter data, exhaust gas temperature data, flue gas composition, oxygen content, pressure and flow rate of the target boiler.

[0016] Preferably, the first target data is used to perform abnormality detection and combustion response analysis on the target boiler air and smoke system, and corresponding abnormality detection results and combustion response results are obtained, including:

[0017] Extracting corresponding data of the set abnormality detection index from the first target data and marking it as first detection index data;

[0018] Compare the first detection index data with the corresponding set detection threshold range, and mark the first detection index data that does not belong to the set detection threshold range as abnormal index data;

[0019] By analyzing the acquisition of abnormal index data, the abnormal degree of the current target boiler air and smoke system is determined, and the first abnormal coefficient and the system abnormal coefficient are generated;

[0020] The abnormal index data, the first abnormal coefficient and the system abnormal coefficient are combined to output as an abnormal detection result;

[0021] Extracting corresponding data for setting a combustion response index from the first target data and marking the data as first response index data;

[0022] The first response index data is input into a pre-established response generation model to generate first combustion response evaluation parameters of the current target boiler air and smoke system in different response directions, and output as a combustion response result.

[0023] Preferably, by analyzing the acquisition of abnormal index data, determining the abnormal degree of the current target boiler air and smoke system, and generating a first abnormal coefficient and a system abnormal coefficient, including:

[0024] If there is no abnormal indicator data, it is determined that the current target boiler air and smoke system is operating normally;

[0025] If there is abnormal index data, it is determined that the current target boiler air and smoke system is operating abnormally, and a first abnormal coefficient is determined according to the deviation between the value of the abnormal index data and the corresponding set detection threshold range;

[0026] The calculation formula of the first abnormal coefficient is as follows:

[0027] Where Y1 represents the first abnormal coefficient of the current abnormal index data; x max Indicates the upper limit of the detection threshold range corresponding to the current abnormal indicator data; x min Indicates the lower limit of the detection threshold range corresponding to the current abnormal indicator data; x 0 It is represented by the value of the current abnormal indicator data; e is represented by a constant, and its value is 2.7;

[0028] A comprehensive analysis is performed on the first abnormal coefficients of all abnormal index data to obtain the system abnormal coefficient of the current target boiler air and smoke system.

[0029] Preferably, according to the abnormality detection result, the abnormality matching pending abnormality handling strategy for the target boiler air and smoke system includes:

[0030] According to the system abnormality coefficient in the abnormality detection result, determining the abnormality level of the current target boiler air and smoke system from a set abnormality level mapping table;

[0031] If the exception level is a special exception, the corresponding processing strategy of the special exception level is adopted as the pending exception processing strategy;

[0032] If the abnormality level is a general abnormality or a slight abnormality, a first processing strategy corresponding to the abnormal indicator data is obtained according to the first abnormality coefficient in the abnormality detection result;

[0033] If there is only a single first processing strategy for the current abnormal indicator data, the current first processing strategy is output as the pending abnormal processing strategy;

[0034] If there are multiple first processing strategies in the current abnormal indicator data, the historical usage frequencies and abnormal repair scores of all existing first processing strategies within a preset time period are extracted;

[0035] The reliability coefficient of each first processing strategy is calculated using the historical usage frequency and the exception repair score, and then the first processing strategy with the largest reliability coefficient is used as the current pending exception processing strategy and output.

[0036] Preferably, based on the combustion response result, an intelligent optimization analysis is performed on the target control parameters in the current target boiler air and smoke system to obtain an optimization strategy for the undetermined parameters, including:

[0037] comparing the first combustion response evaluation parameter in the combustion response result with the corresponding expected combustion response;

[0038] If all first combustion response evaluation parameters meet the corresponding expected combustion response, it is determined that the combustion response of the current boiler air and smoke system meets the expected response requirement;

[0039] If there is a first combustion response evaluation parameter that does not meet the corresponding expected combustion response, the first combustion response evaluation parameter is regarded as a parameter to be optimized, and it is determined that the combustion response of the current boiler air and smoke system does not meet the expected response requirement;

[0040] The parameters to be optimized and the first response indicator data are input into a parameter optimization model established in advance based on the objective function and the objective constraint conditions to obtain an optimization strategy for the parameters to be determined.

[0041] Preferably, the pending exception handling strategy is combined with the pending parameter optimization strategy to generate an intelligent optimization strategy to optimize the current target boiler air and smoke system, including:

[0042] If the current target boiler air and smoke system performs normally and the combustion response does not meet the expected response requirements, the undetermined parameter optimization strategy is used as an intelligent optimization strategy;

[0043] If the current target boiler air and smoke system operates abnormally and the combustion response meets the expected response requirements, the pending abnormality handling strategy is used as the intelligent optimization strategy;

[0044] If the current target boiler air and smoke system operates abnormally and the combustion response does not meet the expected response requirements, the pending abnormality handling strategy is combined with the pending parameter optimization strategy for analysis to determine the intelligent optimization strategy.

[0045] Preferably, the pending exception handling strategy is combined with the pending parameter optimization strategy for analysis to determine an intelligent optimization strategy, including:

[0046] If there is no policy content for adjusting the target control parameter in the current pending exception handling strategy, then each target control parameter in the currently acquired pending parameter optimization strategy is taken as the middle value of the corresponding optimization value range as the optimized parameter value;

[0047] The first designated control parameter optimization strategy is generated by using all the optimized parameter values ​​obtained, and after being summarized and sorted with the currently pending exception handling strategy, it is output as an intelligent optimization strategy;

[0048] If there is a policy content for adjusting the target control parameter in the currently pending exception handling policy, the existing policy content for adjusting the target control parameter is marked as the first analysis content; and the target control parameter existing in the first analysis content is marked as the first control parameter;

[0049] Analyze the target control parameter involved in the current undetermined parameter optimization strategy and the first control parameter to obtain overlapping control parameters;

[0050] If there is a coincidence control parameter, obtaining a first parameter adjustment value of the coincidence control parameter in the first analysis content, and determining an adjustment abnormality target for each coincidence control parameter;

[0051] Obtaining the optimization value range of the overlap control parameter and the optimization evaluation direction of the overlap control parameter from the current undetermined parameter optimization strategy;

[0052] Based on the first parameter adjustment value, the adjustment abnormality target, the optimization value range and the optimization evaluation direction, the value of the current coincidence control parameter is determined to obtain the optimized coincidence parameter value;

[0053] For the remaining target control parameters except the overlap control parameters in the currently obtained pending parameter optimization strategy, the middle values ​​of the corresponding optimization value range are taken as the optimized parameter values;

[0054] Generate a second designated control parameter optimization strategy using all the obtained optimized parameter values ​​and optimized coincident parameter values;

[0055] After replacing the parameter adjustment value of the corresponding coincidence control parameter in the currently pending exception handling strategy with the optimized coincidence parameter value, a specified exception handling strategy is generated;

[0056] The second designated control parameter optimization strategy and the designated exception handling strategy are summarized and collated and output as an intelligent optimization strategy.

[0057] Preferably, based on the first parameter adjustment value, the abnormal adjustment target, the optimization value range and the optimization evaluation direction, the value of the current coincidence control parameter is determined to obtain the optimized coincidence parameter value, including:

[0058] The first correlation coefficient is obtained by evaluating the correlation between the abnormal target and the optimization evaluation direction by using the set correlation evaluation index;

[0059] Calculating the optimized overlap parameter value of the front overlap control parameter by using the first correlation coefficient;

[0060] The calculation formula of the optimized coincidence parameter value is as follows:

[0061] Wherein, H1 represents the optimized coincidence parameter value of the current coincidence control parameter; h0 represents the first parameter adjustment value of the current coincidence control parameter; f1 represents the first correlation coefficient; q max It represents the upper limit of the value within the corresponding optimization range of the current coincidence control parameter; q min It represents the lower limit of the value within the corresponding optimization range of the current coincidence control parameter; ln represents the natural logarithm; e represents a constant with a value of 2.7.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] The first target data is obtained by real-time collection and preprocessing of key data of the target boiler air and smoke system. Based on the first target data, abnormality detection and combustion response analysis are performed on the target boiler air and smoke system. Then, an intelligent optimization strategy is generated according to the obtained abnormal detection results and combustion response results to optimize the current target boiler air and smoke system, thereby achieving effective control of the boiler air and smoke system and improving the boiler combustion efficiency.

[0064] 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.

[0065] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] 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:

[0067] Figure 1 The present invention is a flowchart of an intelligent optimization method for a boiler air and smoke system in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] 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.

[0069] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system. Figure 1 As shown, including:

[0070] Step 1: collect key data of the target boiler air and smoke system in real time and perform data preprocessing to obtain the first target data;

[0071] Step 2: using the first target data to perform abnormality detection and combustion response analysis on the target boiler air and smoke system, and obtain abnormality detection results and combustion response results accordingly;

[0072] Step 3: According to the abnormality detection result, match the abnormality existing in the target boiler air and smoke system with the pending abnormality handling strategy;

[0073] Step 4: Based on the combustion response result, intelligent optimization analysis is performed on the target control parameters in the current target boiler air and smoke system to obtain an optimization strategy for the undetermined parameters;

[0074] Step 5: Combine the pending exception handling strategy with the pending parameter optimization strategy to generate an intelligent optimization strategy to optimize the current target boiler air and smoke system.

[0075] In this embodiment, the target boiler air and smoke system refers to the boiler air and smoke system that currently needs to be monitored and optimized. The boiler air and smoke system refers to a collection of a series of equipment and pipelines in the boiler used to transport air, fuel and exhaust flue gas, including blowers, induced draft fans, air preheaters, flues, dust removal equipment and other components; key data refers to important parameters collected by preset sensors that can reflect the operating status and performance of the target boiler air and smoke system, such as temperature, pressure, and flow rate; data preprocessing refers to data cleaning and normalization processing of key data; the first target data is obtained by preprocessing the air and smoke related data.

[0076] In this embodiment, the abnormality detection result is composed of abnormal indicator data, a first abnormality coefficient and a system abnormality coefficient; the combustion response result is composed of a first combustion response evaluation parameter; the pending abnormality handling strategy refers to a pre-selected solution that may be used to handle the abnormality, such as adjusting equipment parameters, repairing or replacing faulty parts, etc.; the pending parameter optimization strategy refers to a possible parameter adjustment solution; the intelligent optimization strategy is to form a comprehensive optimization solution by combining and analyzing the pending abnormality handling strategy and the pending parameter optimization strategy.

[0077] The beneficial effects of the above technical solution are: first target data is obtained by real-time collection and preprocessing of key data of the target boiler air and smoke system, abnormality detection and combustion response analysis of the target boiler air and smoke system is performed based on the first target data, and then an intelligent optimization strategy is generated according to the obtained abnormal detection results and combustion response results to optimize the current target boiler air and smoke system, thereby achieving effective control of the boiler air and smoke system and improving the boiler combustion efficiency.

[0078] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system, which collects key data of a target boiler air and smoke system in real time and performs data preprocessing to obtain first target data, including:

[0079] Utilize the preset sensors pre-installed at the set key positions of the current target boiler air and smoke system to collect the key data of the target boiler air and smoke system in real time;

[0080] The acquired key data is divided according to data types to obtain wind and smoke related data, and the data types are marked on the wind and smoke related data;

[0081] The wind-smoke correlation data is preprocessed to obtain first target data.

[0082] In this embodiment, setting key parts refers to pre-determined important positions that can reflect the operating status and performance of the target boiler's air and smoke system, such as the boiler's flue and air duct; preset sensors are used to collect key data of the current target boiler's air and smoke system, including temperature sensors, pressure sensors, flow rate sensors, humidity sensors, and flue gas composition analyzers; key data refers to important parameters collected by preset sensors that can reflect the operating status and performance of the target boiler's air and smoke system, such as temperature, pressure, and flow rate; data type refers to the specific classification of key data; air and smoke related data refers to the target boiler's combustion parameter data, air volume parameter data, exhaust temperature data, flue gas composition, oxygen content, pressure or flow rate; data preprocessing refers to data cleaning and normalization of key data; the first target data is obtained by preprocessing the air and smoke related data.

[0083] The beneficial effect of the above technical solution is: by real-time collection of key data of the target boiler air and smoke system and data preprocessing to obtain the first target data, the accuracy and real-time performance of data collection can be improved, and data support can be provided for the subsequent intelligent optimization of the boiler air and smoke system.

[0084] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system, which uses the first target data to perform abnormality detection and combustion response analysis on the target boiler air and smoke system, and correspondingly obtains abnormality detection results and combustion response results, including:

[0085] Extracting corresponding data of the set abnormality detection index from the first target data and marking it as first detection index data;

[0086] Compare the first detection index data with the corresponding set detection threshold range, and mark the first detection index data that does not belong to the set detection threshold range as abnormal index data;

[0087] By analyzing the acquisition of abnormal index data, the abnormal degree of the current target boiler air and smoke system is determined, and a first abnormal coefficient and a system abnormal coefficient are generated;

[0088] The abnormal index data, the first abnormal coefficient and the system abnormal coefficient are combined to output as an abnormality detection result;

[0089] Extracting corresponding data for setting a combustion response index from the first target data and marking the data as first response index data;

[0090] The first response index data is input into a pre-established response generation model to generate first combustion response evaluation parameters of the current target boiler air and smoke system in different response directions, and output as a combustion response result.

[0091] In this embodiment, the set abnormality detection index refers to a pre-set index for detecting and defining abnormalities, including temperature, pressure, flow, vibration, etc.; the first detection index data refers to the measured data extracted from the first target data and corresponding to the set abnormality detection index; the set detection threshold range is pre-set, and different first detection index data may correspond to different set detection threshold ranges; the abnormal index data refers to the first detection index data that does not belong to the set detection threshold range.

[0092] In this embodiment, the first abnormality coefficient is used to characterize the degree of abnormality of the current target boiler air and smoke system under different set abnormality detection indicators; the system abnormality coefficient is used to represent the degree of abnormality of the entire target boiler air and smoke system; the set combustion response indicator refers to a pre-set indicator for evaluating the combustion performance of the system; the first response indicator data refers to the measured data extracted from the first target data and corresponding to the set combustion response indicator; the response generation model refers to a model pre-established based on a neural network, which is used to generate a combustion response evaluation parameter based on the input first response indicator data; the response direction includes combustion efficiency, emission concentration, fuel consumption rate, etc.; the first combustion response evaluation parameter refers to the result output by the response generation model, which is used to evaluate the performance of the current target boiler air and smoke system in a specific response direction.

[0093] The beneficial effect of the above technical solution is: by using the first target data to perform abnormal detection and combustion response analysis on the target boiler air and smoke system, corresponding abnormal detection results and combustion response results are obtained, which can provide data support for the subsequent intelligent optimization of the boiler air and smoke system.

[0094] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system, which determines the abnormality degree of the current target boiler air and smoke system by analyzing the acquisition of abnormal index data, and generates a first abnormality coefficient and a system abnormality coefficient, including:

[0095] If there is no abnormal indicator data, it is determined that the current target boiler air and smoke system is operating normally;

[0096] If there is abnormal index data, it is determined that the current target boiler air and smoke system is operating abnormally, and a first abnormal coefficient is determined according to the deviation between the value of the abnormal index data and the corresponding set detection threshold range;

[0097] The calculation formula of the first abnormal coefficient is as follows:

[0098] Where Y1 represents the first abnormal coefficient of the current abnormal index data; x max Indicates the upper limit of the detection threshold range corresponding to the current abnormal indicator data; x min Indicates the lower limit of the detection threshold range corresponding to the current abnormal indicator data; x 0 It is represented by the value of the current abnormal indicator data; e is represented by a constant, and its value is 2.7;

[0099] A comprehensive analysis is performed on the first abnormal coefficients of all abnormal index data to obtain the system abnormal coefficient of the current target boiler air and smoke system.

[0100] In this embodiment, the first abnormality coefficient is used to characterize the abnormality degree of the current target boiler air and smoke system under different set abnormality detection indicators; the system abnormality coefficient is obtained by weighted average calculation of the first abnormality coefficients of all abnormal indicator data, and is used to represent the abnormality degree of the entire target boiler air and smoke system. The weight assigned to the abnormal indicator data is obtained by solving the matrix constructed after pairwise comparison and relative importance scoring using the hierarchical analysis method.

[0101] The beneficial effect of the above technical solution is: by analyzing the acquisition of abnormal indicator data, determining the abnormal degree of the current target boiler air and smoke system, generating a first abnormal coefficient and a system abnormal coefficient, and providing data support for the subsequent intelligent optimization of the boiler air and smoke system.

[0102] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system, and matches a pending exception handling strategy to an exception existing in a target boiler air and smoke system according to the exception detection result, including:

[0103] According to the system abnormality coefficient in the abnormality detection result, determining the abnormality level of the current target boiler air and smoke system from a set abnormality level mapping table;

[0104] If the exception level is a special exception, the corresponding processing strategy of the special exception level is adopted as the pending exception processing strategy;

[0105] If the abnormality level is a general abnormality or a slight abnormality, a first processing strategy corresponding to the abnormal indicator data is obtained according to the first abnormality coefficient in the abnormality detection result;

[0106] If there is only a single first processing strategy for the current abnormal indicator data, the current first processing strategy is output as the pending abnormal processing strategy;

[0107] If there are multiple first processing strategies in the current abnormal indicator data, the historical usage frequencies and abnormal repair scores of all existing first processing strategies within a preset time period are extracted;

[0108] The reliability coefficient of each first processing strategy is calculated using the historical usage frequency and the exception repair score, and then the first processing strategy with the largest reliability coefficient is used as the current pending exception processing strategy and output.

[0109] In this embodiment, the abnormal level mapping table is set to consist of the range of system abnormality coefficients and the corresponding abnormality levels; the abnormality levels include three levels: slight abnormality, general abnormality and special abnormality; the first processing strategy refers to the preliminary processing suggestions for abnormal indicator data, such as adjusting equipment parameters and replacing faulty parts; the historical usage frequency refers to the number of times the first processing strategy is used to process similar abnormal indicator data within a preset time period, wherein the preset time period is predetermined; the abnormal repair score is used to evaluate the effect of a certain first processing strategy after processing specific abnormal indicator data, and is determined based on the repair success rate and processing time;

[0110] The formula of the reliability coefficient is expressed as Wherein, k1 represents the reliability coefficient of the current first processing strategy; It is represented by the abnormal repair score of the current first processing strategy; ω1 is represented by the influence weight of the abnormal repair score on the calculation reliability coefficient; s1 is represented by the historical usage frequency of the current first processing strategy; ω2 is represented by the influence weight of the historical usage frequency on the calculation reliability coefficient; δ1 is represented by the contribution weight of the repair success rate to the calculation abnormal repair score; C 1 It is expressed as the historical repair success rate of the current first processing strategy; t avg It is expressed as the average value of the historical processing time of the current first processing strategy; t 0 It is represented as the average of the maximum and minimum historical processing time of the current first processing strategy; δ2 is represented as the contribution weight of the strategy processing time to the calculation of the anomaly repair score.

[0111] The beneficial effect of the above technical solution is that: through the abnormal detection results, the abnormal matching pending abnormal handling strategy for the abnormality existing in the target boiler air and smoke system can provide data basis for the determination of subsequent intelligent optimization strategy, thereby helping to improve the boiler combustion efficiency.

[0112] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system. Based on the combustion response result, an intelligent optimization analysis is performed on the target control parameters in the current target boiler air and smoke system to obtain an optimization strategy for the undetermined parameters, including:

[0113] comparing the first combustion response evaluation parameter in the combustion response result with the corresponding expected combustion response;

[0114] If all first combustion response evaluation parameters meet the corresponding expected combustion response, it is determined that the combustion response of the current boiler air and smoke system meets the expected response requirement;

[0115] If there is a first combustion response evaluation parameter that does not meet the corresponding expected combustion response, the first combustion response evaluation parameter is regarded as a parameter to be optimized, and it is determined that the combustion response of the current boiler air and smoke system does not meet the expected response requirement;

[0116] The parameters to be optimized and the first response indicator data are input into a parameter optimization model established in advance based on the objective function and the objective constraint conditions to obtain an optimization strategy for the parameters to be determined.

[0117] In this embodiment, the first combustion response evaluation parameter is used to reflect the combustion performance of the current target boiler air and smoke system in different response directions, such as combustion efficiency, emission concentration and fuel consumption rate; the parameter to be optimized refers to the first combustion response evaluation parameter that does not meet the expected combustion response; the parameter optimization model is a mathematical model established based on the objective function and the objective constraint conditions, wherein the objective function is used to define the optimization target of the system performance, generally to maximize the combustion efficiency, and the objective constraint conditions are determined by considering the physical limitations and safety requirements of the system, such as boiler temperature range and flow rate restrictions; the first response index data refers to the key data extracted from the first target data for evaluating the system combustion response, such as combustion temperature and oxygen concentration; the undetermined parameter optimization strategy is the output of the parameter optimization model, which is composed of the optimization value range of the target control parameters, wherein the target control parameters include air volume, oxygen content, secondary air door opening, burner swing angle, etc.

[0118] The beneficial effect of the above technical solution is: by performing intelligent optimization analysis on the target control parameters in the current target boiler air and smoke system based on the combustion response results, an optimization strategy for the pending parameters is obtained, which can provide a data basis for determining the subsequent intelligent optimization strategy, thereby helping to improve the boiler combustion efficiency.

[0119] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system, which combines the pending exception handling strategy with the pending parameter optimization strategy to generate an intelligent optimization strategy to optimize the current target boiler air and smoke system, including:

[0120] If the current target boiler air and smoke system performs normally and the combustion response does not meet the expected response requirements, the undetermined parameter optimization strategy is used as an intelligent optimization strategy;

[0121] If the current target boiler air and smoke system operates abnormally and the combustion response meets the expected response requirements, the pending abnormality handling strategy is used as the intelligent optimization strategy;

[0122] If the current target boiler air and smoke system operates abnormally and the combustion response does not meet the expected response requirements, the pending abnormality handling strategy is combined with the pending parameter optimization strategy for analysis to determine the intelligent optimization strategy.

[0123] In this embodiment, the intelligent optimization strategy is a pending parameter optimization strategy (the current target boiler air and smoke system performs normally, and the combustion response does not meet the expected response requirements), a pending exception handling strategy (the current target boiler air and smoke system operates abnormally, and the combustion response meets the expected response requirements), or a strategy obtained by combining and analyzing the pending exception handling strategy with the intelligent optimization strategy (the current target boiler air and smoke system operates abnormally, and the combustion response does not meet the expected response requirements).

[0124] The beneficial effect of the above technical solution is: by combining the pending exception handling strategy with the pending parameter optimization strategy, an intelligent optimization strategy is generated to optimize the current target boiler air and smoke system, which can improve the flexibility and reliability of system operation, achieve effective control of the boiler air and smoke system, and thus improve the boiler combustion efficiency.

[0125] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system, which combines and analyzes the pending exception handling strategy with the pending parameter optimization strategy to determine the intelligent optimization strategy, including:

[0126] If there is no policy content for adjusting the target control parameter in the current pending exception handling strategy, then each target control parameter in the currently acquired pending parameter optimization strategy is taken as the middle value of the corresponding optimization value range as the optimized parameter value;

[0127] The first designated control parameter optimization strategy is generated by using all the optimized parameter values ​​obtained, and after being summarized and sorted with the currently pending exception handling strategy, it is output as an intelligent optimization strategy;

[0128] If there is a policy content for adjusting the target control parameter in the currently pending exception handling policy, the existing policy content for adjusting the target control parameter is marked as the first analysis content; and the target control parameter existing in the first analysis content is marked as the first control parameter;

[0129] Analyze the target control parameter involved in the current undetermined parameter optimization strategy and the first control parameter to obtain overlapping control parameters;

[0130] If there is a coincidence control parameter, obtaining a first parameter adjustment value of the coincidence control parameter in the first analysis content, and determining an adjustment abnormality target for each coincidence control parameter;

[0131] Obtaining the optimization value range of the overlap control parameter and the optimization evaluation direction of the overlap control parameter from the current undetermined parameter optimization strategy;

[0132] Based on the first parameter adjustment value, the adjustment abnormality target, the optimization value range and the optimization evaluation direction, the value of the current coincidence control parameter is determined to obtain the optimized coincidence parameter value;

[0133] For the remaining target control parameters except the overlap control parameters in the currently obtained pending parameter optimization strategy, the middle values ​​of the corresponding optimization value range are taken as the optimized parameter values;

[0134] Generate a second designated control parameter optimization strategy using all the obtained optimized parameter values ​​and optimized coincident parameter values;

[0135] After replacing the parameter adjustment value of the corresponding coincidence control parameter in the currently pending exception handling strategy with the optimized coincidence parameter value, a specified exception handling strategy is generated;

[0136] The second designated control parameter optimization strategy and the designated exception handling strategy are summarized and collated and output as an intelligent optimization strategy.

[0137] In this embodiment, the optimized parameter value refers to the target control parameters in the pending parameter optimization strategy being taken as the middle value of the corresponding optimization value range; the first designated control parameter optimization strategy refers to the parameter optimization strategy formed by taking all the target control parameters in the pending parameter optimization strategy as the middle value of the optimization value range when the pending exception handling strategy does not contain the content of adjusting the target control parameters; the first analysis content refers to the part of the content in the pending exception handling strategy that involves adjusting the target control parameters; the first control parameter refers to the target control parameter that needs to be adjusted mentioned in the first analysis content.

[0138] In this embodiment, the overlap control parameter refers to the target control parameter involved in both the pending parameter optimization strategy and the pending exception handling strategy; the first parameter adjustment value refers to the adjustment value recommended for the overlap control parameter in the pending exception handling strategy; the adjustment exception target refers to the abnormal problem to be solved by adjusting the overlap control parameter; the optimization evaluation direction refers to the response direction optimized by adjusting the overlap control parameter, such as combustion efficiency, emission concentration, and fuel consumption rate, etc.; the optimized overlap parameter value refers to the optimal value of the overlap control parameter determined by analysis; the second designated control parameter optimization strategy refers to a parameter optimization strategy formed by taking the middle value of the optimized value range for other target control parameters except the overlap control parameter after considering the optimized value of the overlap control parameter; the designated exception handling strategy refers to a strategy formed by replacing the parameter adjustment value of the corresponding overlap control parameter in the pending exception handling strategy with the optimized overlap parameter value.

[0139] The beneficial effect of the above technical solution is: when the target boiler air and smoke system operates abnormally and the combustion response does not meet the expected response requirements, the pending exception handling strategy is combined with the intelligent optimization strategy for analysis to determine the intelligent optimization strategy, which can provide data basis for the intelligent optimization of the boiler air and smoke system, realize effective control of the boiler air and smoke system, and thus improve the boiler combustion efficiency.

[0140] The embodiment of the present invention provides an intelligent optimization method for a boiler air and smoke system, which determines the value of a current coincidence control parameter based on a first parameter adjustment value, an adjustment abnormality target, an optimization value range, and an optimization evaluation direction, and obtains an optimized coincidence parameter value, including:

[0141] The first correlation coefficient is obtained by evaluating the correlation between the abnormal target and the optimization evaluation direction by using the set correlation evaluation index;

[0142] Calculating the optimized overlap parameter value of the front overlap control parameter by using the first correlation coefficient;

[0143] The calculation formula of the optimized coincidence parameter value is as follows:

[0144] Wherein, H1 represents the optimized coincidence parameter value of the current coincidence control parameter; h0 represents the first parameter adjustment value of the current coincidence control parameter; f1 represents the first correlation coefficient; q max It represents the upper limit of the value within the corresponding optimization range of the current coincidence control parameter; q min It represents the lower limit of the value within the corresponding optimization range of the current coincidence control parameter; ln represents the natural logarithm; e represents a constant with a value of 2.7.

[0145] In this embodiment, the set association evaluation index is a pre-set quantitative index used to evaluate the correlation between the adjustment abnormal target and the optimization evaluation direction, generally refers to the correlation index between the adjustment abnormal target and the optimization evaluation direction (calculated using statistical methods such as the Pearson correlation coefficient and the Spearman rank correlation coefficient) and the causal correlation index (obtained based on a pre-established causal reasoning model analysis, wherein the causal reasoning model is obtained by collecting a large amount of historical key data of the boiler air and smoke system and processing it to obtain first data, and then using the first data and the causal graph between the variables established according to the operating principle and empirical knowledge of the boiler air and smoke system as training data to train the neural network); the first correlation coefficient is obtained by weighted average calculation of the index coefficient obtained by evaluating using the set association evaluation index, reflecting the degree of correlation between the adjustment abnormal target and the optimization evaluation direction, and the value range is [0, 1]; the optimized overlap parameter value refers to the optimal value of the overlap control parameter determined by analysis.

[0146] The beneficial effect of the above technical solution is: by determining the value of the current overlap control parameter based on the first parameter adjustment value, adjusting the abnormal target, optimizing the value range and optimizing the evaluation direction, and obtaining the optimized overlap parameter value, it can help generate accurate intelligent optimization strategies and provide strong data basis for the intelligent optimization of the boiler air and smoke system.

[0147] 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 intelligent optimization method for a boiler air and smoke system, characterized in that: include: Step 1: collect key data of the target boiler air and smoke system in real time and perform data preprocessing to obtain the first target data; Step 2: using the first target data to perform abnormality detection and combustion response analysis on the target boiler air and smoke system, and obtain abnormality detection results and combustion response results accordingly; Step 3: According to the anomaly detection result, match the anomaly existing in the target boiler air and smoke system with a pending anomaly handling strategy; Step 4: Based on the combustion response result, intelligent optimization analysis is performed on the target control parameters in the current target boiler air and smoke system to obtain an optimization strategy for the undetermined parameters; Step 5: Combine the pending exception handling strategy with the pending parameter optimization strategy to generate an intelligent optimization strategy to optimize the current target boiler air and smoke system.

2. The intelligent optimization method for a boiler air and smoke system according to claim 1 is characterized in that: Real-time collection of key data of the target boiler air and smoke system and data preprocessing to obtain the first target data, including: Utilize the preset sensors pre-installed at the set key positions of the current target boiler air and smoke system to collect the key data of the target boiler air and smoke system in real time; The acquired key data is divided according to data types to obtain wind and smoke related data, and the data types are marked on the wind and smoke related data; The wind-smoke correlation data is preprocessed to obtain first target data.

3. The intelligent optimization method for a boiler air and smoke system according to claim 2 is characterized in that: The wind and smoke related data refers to the combustion parameter data, air volume parameter data, exhaust gas temperature data, flue gas composition, oxygen content, pressure and flow rate of the target boiler.

4. The intelligent optimization method for a boiler air and smoke system according to claim 1 is characterized in that: The first target data is used to perform abnormality detection and combustion response analysis on the target boiler air and smoke system, and corresponding abnormality detection results and combustion response results are obtained, including: Extracting corresponding data of the set abnormality detection index from the first target data and marking it as first detection index data; Compare the first detection index data with the corresponding set detection threshold range, and mark the first detection index data that does not belong to the set detection threshold range as abnormal index data; By analyzing the acquisition of abnormal index data, the abnormal degree of the current target boiler air and smoke system is determined, and the first abnormal coefficient and the system abnormal coefficient are generated; The abnormal index data, the first abnormal coefficient and the system abnormal coefficient are combined to output as an abnormal detection result; Extracting corresponding data for setting a combustion response index from the first target data and marking the data as first response index data; The first response index data is input into a pre-established response generation model to generate first combustion response evaluation parameters of the current target boiler air and smoke system in different response directions, and output as a combustion response result.

5. The intelligent optimization method for a boiler air and smoke system according to claim 4 is characterized in that: By analyzing the acquisition of abnormal index data, the abnormal degree of the current target boiler air and smoke system is determined, and the first abnormal coefficient and the system abnormal coefficient are generated, including: If there is no abnormal indicator data, it is determined that the current target boiler air and smoke system is operating normally; If there is abnormal index data, it is determined that the current target boiler air and smoke system is operating abnormally, and a first abnormal coefficient is determined according to the deviation between the value of the abnormal index data and the corresponding set detection threshold range; The calculation formula of the first abnormal coefficient is as follows: Y1 represents the first abnormal coefficient of the current abnormal index data; x max Indicates the upper limit of the detection threshold range corresponding to the current abnormal indicator data; x min It represents the lower limit of the detection threshold range corresponding to the current abnormal index data; x0 represents the value of the current abnormal index data; e represents a constant with a value of 2.7; A comprehensive analysis is performed on the first abnormal coefficients of all abnormal index data to obtain the system abnormal coefficient of the current target boiler air and smoke system.

6. The intelligent optimization method for a boiler air and smoke system according to claim 1, characterized in that: According to the anomaly detection result, the anomaly matching pending anomaly handling strategy for the target boiler air and smoke system includes: According to the system abnormality coefficient in the abnormality detection result, determining the abnormality level of the current target boiler air and smoke system from a set abnormality level mapping table; If the exception level is a special exception, the corresponding processing strategy of the special exception level is adopted as the pending exception processing strategy; If the abnormality level is a general abnormality or a slight abnormality, a first processing strategy corresponding to the abnormal indicator data is obtained according to the first abnormality coefficient in the abnormality detection result; If there is only a single first processing strategy for the current abnormal indicator data, the current first processing strategy is output as the pending abnormal processing strategy; If there are multiple first processing strategies in the current abnormal indicator data, the historical usage frequencies and abnormal repair scores of all existing first processing strategies within a preset time period are extracted; The reliability coefficient of each first processing strategy is calculated using the historical usage frequency and the exception repair score, and then the first processing strategy with the largest reliability coefficient is used as the current pending exception processing strategy and output.

7. The intelligent optimization method for a boiler air and smoke system according to claim 1, characterized in that: Based on the combustion response results, the target control parameters in the current target boiler air and smoke system are intelligently optimized and analyzed to obtain the optimization strategy for the undetermined parameters, including: comparing the first combustion response evaluation parameter in the combustion response result with the corresponding expected combustion response; If all first combustion response evaluation parameters meet the corresponding expected combustion response, it is determined that the combustion response of the current boiler air and smoke system meets the expected response requirement; If there is a first combustion response evaluation parameter that does not meet the corresponding expected combustion response, the first combustion response evaluation parameter is regarded as a parameter to be optimized, and it is determined that the combustion response of the current boiler air and smoke system does not meet the expected response requirement; The parameters to be optimized and the first response indicator data are input into a parameter optimization model established in advance based on the objective function and the objective constraint conditions to obtain an optimization strategy for the parameters to be determined.

8. The intelligent optimization method for a boiler air and smoke system according to claim 1 is characterized in that: The pending exception handling strategy is combined with the pending parameter optimization strategy to generate an intelligent optimization strategy to optimize the current target boiler air and smoke system, including: If the current target boiler air and smoke system performs normally and the combustion response does not meet the expected response requirements, the undetermined parameter optimization strategy is used as an intelligent optimization strategy; If the current target boiler air and smoke system operates abnormally and the combustion response meets the expected response requirements, the pending abnormality handling strategy is used as the intelligent optimization strategy; If the current target boiler air and smoke system operates abnormally and the combustion response does not meet the expected response requirements, the pending abnormality handling strategy is combined with the pending parameter optimization strategy for analysis to determine the intelligent optimization strategy.

9. The intelligent optimization method for a boiler air and smoke system according to claim 8, characterized in that: The pending exception handling strategy is combined with the pending parameter optimization strategy for analysis to determine an intelligent optimization strategy, including: If there is no policy content for adjusting the target control parameter in the current pending exception handling strategy, then each target control parameter in the currently acquired pending parameter optimization strategy is taken as the middle value of the corresponding optimization value range as the optimized parameter value; The first designated control parameter optimization strategy is generated by using all the optimized parameter values ​​obtained, and after being summarized and sorted with the currently pending exception handling strategy, it is output as an intelligent optimization strategy; If there is a policy content for adjusting the target control parameter in the currently pending exception handling policy, the existing policy content for adjusting the target control parameter is marked as the first analysis content; and the target control parameter existing in the first analysis content is marked as the first control parameter; Analyze the target control parameter involved in the current undetermined parameter optimization strategy and the first control parameter to obtain overlapping control parameters; If there is a coincidence control parameter, obtaining a first parameter adjustment value of the coincidence control parameter in the first analysis content, and determining an adjustment abnormality target for each coincidence control parameter; Obtaining the optimization value range of the overlap control parameter and the optimization evaluation direction of the overlap control parameter from the current undetermined parameter optimization strategy; Based on the first parameter adjustment value, the adjustment abnormality target, the optimization value range and the optimization evaluation direction, the value of the current coincidence control parameter is determined to obtain the optimized coincidence parameter value; For the remaining target control parameters except the overlap control parameters in the currently obtained pending parameter optimization strategy, the middle values ​​of the corresponding optimization value range are taken as the optimized parameter values; Generate a second designated control parameter optimization strategy using all the obtained optimized parameter values ​​and optimized coincident parameter values; After replacing the parameter adjustment value of the corresponding coincidence control parameter in the currently pending exception handling strategy with the optimized coincidence parameter value, a specified exception handling strategy is generated; The second designated control parameter optimization strategy and the designated exception handling strategy are summarized and collated and output as an intelligent optimization strategy.

10. The intelligent optimization method for a boiler air and smoke system according to claim 9, characterized in that: Based on the first parameter adjustment value, the abnormal adjustment target, the optimization value range and the optimization evaluation direction, the value of the current coincidence control parameter is determined to obtain the optimized coincidence parameter value, including: The first correlation coefficient is obtained by evaluating the correlation between the abnormal target and the optimization evaluation direction by using the set correlation evaluation index; Calculating the optimized overlap parameter value of the front overlap control parameter by using the first correlation coefficient; The calculation formula of the optimized coincidence parameter value is as follows: Wherein, H1 represents the optimized coincidence parameter value of the current coincidence control parameter; h0 represents the first parameter adjustment value of the current coincidence control parameter; f1 represents the first correlation coefficient; q max It represents the upper limit of the value within the corresponding optimization range of the current coincidence control parameter; q min It represents the lower limit of the value within the corresponding optimization range of the current coincidence control parameter; ln represents the natural logarithm; e represents a constant with a value of 2.7.

Images