Combustion energy efficiency optimization method and system, electronic equipment and storage medium

Abstract

The invention provides a combustion energy efficiency optimization method and system, electronic equipment and a storage medium. The combustion energy efficiency optimization method comprises the following steps: inputting first state time sequence data into a control variable determination model, and outputting a control variable matched with the first state time sequence data; the control variable is used for energy efficiency optimization of the boiler; by means of the energy efficiency evaluation model, whether expected state time sequence data obtained after energy efficiency optimization is conducted on the boiler according to the control variables meet energy efficiency conditions or not is determined; in response to the condition that the expected state time sequence data does not meet the energy efficiency condition, adjusting a parameter value of at least one target parameter in the expected state time sequence data; and in response to the condition that the adjusted expected state time sequence data meets the energy efficiency condition, inputting the adjusted expected state time sequence data into a control variable determination model, and outputting a matched control variable. The effectiveness of the control variable can be ensured, it is ensured that the expected state time sequence data can meet the requirement after the boiler is optimized according to the control variable, and the effectiveness of energy efficiency optimization is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler optimization control, and in particular to a combustion energy efficiency optimization method, system, electronic equipment and storage medium. Background Art

[0002] Boilers are important equipment for industrial production and power supply. In order to improve the combustion efficiency of boilers and reduce the generation and emission of harmful substances, energy efficiency optimization is required. Boiler combustion is a complex process involving the interaction of multiple factors. The state data of the boiler combustion process is relatively complex. It is impossible to accurately understand the performance characteristics and operating rules of the boiler, and thus it is impossible to accurately optimize energy efficiency, resulting in poor effectiveness of energy efficiency optimization. Summary of the invention

[0003] The technical problem to be solved by the present disclosure is to overcome the defect of poor effectiveness of energy efficiency optimization in the prior art, and to provide a combustion energy efficiency optimization method, system, electronic device and storage medium.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] In a first aspect, a combustion energy efficiency optimization method is provided, the combustion energy efficiency optimization method comprising:

[0006] Acquire first state time series data of the boiler;

[0007] Inputting the first state time series data into a control variable determination model, and outputting a control variable matching the first state time series data; wherein the control variable determination model is obtained by training a neural network with training samples, the training samples include state time series data marked with control variables, and the control variables are used for energy efficiency optimization of the boiler;

[0008] Using an energy efficiency evaluation model, determining whether the expected state time series data after the energy efficiency of the boiler is optimized according to the control variable meets the energy efficiency conditions; wherein the energy efficiency evaluation model is based on the business mechanism knowledge of the boiler and is obtained by the counter-balance method;

[0009] In response to the expected state time series data satisfying the energy efficiency condition, optimizing the energy efficiency of the boiler based on the control variable;

[0010] In response to the expected state time series data not satisfying the energy efficiency condition, adjusting a parameter value of at least one target parameter in the expected state time series data, and re-determining whether the adjusted expected state time series data satisfies the energy efficiency condition by using the energy efficiency evaluation model;

[0011] In response to the adjusted expected state time series data satisfying the energy efficiency condition, the adjusted expected state time series data is input into the control variable determination model, and a control variable matching the expected state time series data is output.

[0012] Optionally, the step of adjusting a parameter value of at least one target parameter in the expected state time series data includes:

[0013] Determine a target parameter from the expected state time series data; wherein the target parameter includes: a target parameter whose parameter value has the greatest impact on the energy efficiency score, or a target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold;

[0014] Iteratively adjust the parameter value of the target parameter in the expected state time series data, and input the expected state time series data after each adjustment into the energy efficiency evaluation model, and re-determine whether the adjusted expected state time series data meets the energy efficiency condition, and stop the iteration until the adjusted expected state time series data meets the energy efficiency condition.

[0015] Optionally, the step of determining target parameters from the expected state time series data includes:

[0016] Each time, the parameter value of one parameter in the expected state time series data is changed to a reference value and input into the energy efficiency evaluation model;

[0017] According to the energy efficiency scores before and after the parameter value is changed, the target parameter whose parameter value has the greatest impact on the energy efficiency score is screened out, or the target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold is screened out.

[0018] Optionally, in response to the adjusted expected state time series data satisfying the energy efficiency condition, the combustion energy efficiency optimization method further includes:

[0019] storing the adjusted expected state time series data and the corresponding control variables in an associated manner;

[0020] The expected state time series data and the corresponding control variables stored in the association are used to optimize the energy efficiency of the boiler in combination with the second state time series data; wherein the similarity between the second state time series data and the expected state time series data stored in the association is greater than a similarity threshold.

[0021] In a second aspect, a combustion energy efficiency optimization system is provided, the combustion energy efficiency optimization system comprising:

[0022] An acquisition module, used for acquiring first state time series data of the boiler;

[0023] a first control variable determination module, configured to input the first state time series data into a control variable determination model, and output a control variable matching the first state time series data; wherein the control variable determination model is obtained by training a neural network with training samples, the training samples comprising state time series data marked with control variables, and the control variables are used for energy efficiency optimization of the boiler;

[0024] An energy efficiency evaluation module, used to determine whether the expected state time series data after the energy efficiency optimization of the boiler according to the control variable meets the energy efficiency condition by using an energy efficiency evaluation model; wherein the energy efficiency evaluation model is based on the business mechanism knowledge of the boiler and is obtained by using a counter-balance method;

[0025] an energy efficiency optimization module, configured to optimize the energy efficiency of the boiler based on the control variable in response to the expected state time series data satisfying the energy efficiency condition;

[0026] an adjustment module, configured to adjust a parameter value of at least one target parameter in the expected state time series data in response to the expected state time series data not satisfying the energy efficiency condition, and re-determine whether the adjusted expected state time series data satisfies the energy efficiency condition by using the energy efficiency evaluation model;

[0027] The second control variable determination module is used for inputting the adjusted expected state time series data into the control variable determination model in response to the adjusted expected state time series data satisfying the energy efficiency condition, and outputting the control variable matching the expected state time series data.

[0028] Optionally, the adjustment module includes:

[0029] A target parameter determination unit, configured to determine a target parameter from the expected state time series data; wherein the target parameter includes: a target parameter whose parameter value has the greatest impact on the energy efficiency score, or a target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold;

[0030] An iterative adjustment unit is used to iteratively adjust the parameter value of the target parameter in the expected state time series data, and input the expected state time series data after each adjustment into the energy efficiency evaluation model, and re-determine whether the adjusted expected state time series data meets the energy efficiency condition, until the iteration is stopped when the adjusted expected state time series data meets the energy efficiency condition.

[0031] Optionally, the target parameter determination unit is specifically used to:

[0032] Each time, the parameter value of one parameter in the expected state time series data is changed to a reference value and input into the energy efficiency evaluation model;

[0033] According to the energy efficiency scores before and after the parameter value is changed, the target parameter whose parameter value has the greatest impact on the energy efficiency score is screened out, or the target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold is screened out.

[0034] Optionally, the combustion energy efficiency optimization system further includes:

[0035] A storage module, configured to store the adjusted expected state time series data and the corresponding control variable in association with each other in response to the adjusted expected state time series data satisfying the energy efficiency condition;

[0036] The expected state time series data and the corresponding control variables stored in the association are used to optimize the energy efficiency of the boiler in combination with the second state time series data; wherein the similarity between the second state time series data and the expected state time series data stored in the association is greater than a similarity threshold.

[0037] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein the processor implements any of the above-mentioned combustion energy efficiency optimization methods when executing the computer program.

[0038] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the combustion energy efficiency optimization method described in any one of the above items is implemented.

[0039] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.

[0040] The positive progressive effect of the present disclosure lies in: the present disclosure utilizes the control variable determination model to obtain the control variable matching the first state time series data, and then utilizes the energy efficiency evaluation model to determine whether the expected state time series data after optimizing the boiler according to the control variable meets the energy efficiency conditions. If the energy efficiency conditions are not met, the expected state time series data is adjusted, and the control variable determination model is re-entered to obtain the control variable matching the adjusted expected state time series data, which is used for the energy efficiency optimization of the boiler to ensure the effectiveness of the control variable, that is, to ensure that after optimizing the boiler according to the control variable, the expected state time series data can meet the demand, thereby improving the effectiveness of energy efficiency optimization. Energy efficiency optimization can reduce slag generation, extend boiler life, reduce equipment maintenance costs, and improve boiler production efficiency, thereby effectively improving combustion efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flow chart of a combustion energy efficiency optimization method provided for an exemplary embodiment of the present disclosure;

[0042] Figure 2 Another flow chart of a combustion energy efficiency optimization method provided for an exemplary embodiment of the present disclosure;

[0043] Figure 3 A graph showing a trend of changes in inlet NOx concentration in a combustion energy efficiency optimization method provided by an exemplary embodiment of the present disclosure;

[0044] Figure 4 A graph showing a trend of ammonia injection amount in a combustion energy efficiency optimization method provided by an exemplary embodiment of the present disclosure;

[0045] Figure 5 A schematic diagram of a module of an optimization system of a combustion energy efficiency optimization system provided by an exemplary embodiment of the present disclosure;

[0046] Figure 6 The present invention provides a schematic structural diagram of an electronic device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0047] The present disclosure is further described below by way of examples, but the present disclosure is not limited to the scope of the examples.

[0048] Prefixes such as "first" and "second" are used in the embodiments of the present disclosure only to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of prefixes such as ordinal numbers to distinguish description objects in the embodiments of the present disclosure does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and no unnecessary limitation should be constituted due to the use of such prefixes. In addition, in the description of the present embodiment, unless otherwise specified, the meaning of "plurality" is two or more.

[0049] Example 1

[0050] Figure 1 A flowchart of a combustion energy efficiency optimization method provided by an exemplary embodiment of the present disclosure, the combustion energy efficiency optimization method comprising:

[0051] Step S101, obtaining first state time series data of the boiler.

[0052] The first state time series data represents the state time series data within a preset time period traced back from the current moment. The preset time period can be set according to actual conditions and is not particularly limited here.

[0053] Step S102: input the first state time series data into a control variable determination model, and output a control variable matching the first state time series data.

[0054] The control variable determination model is obtained by training a neural network with training samples. The training samples include state time series data labeled with control variables. The control variables are used for optimizing the energy efficiency of the boiler.

[0055] The input of the control variable determination model is the state time series data, and the output is the control variable.

[0056] The control variables are usually adjustable operating parameters, such as coal feed, pulverizer outlet mixture temperature, fan inlet air volume, etc. The expression of the control variables is as follows:

[0057] X={x1,x2,x3,…,x m}

[0058] State time series data is usually the result of detection, which is used to monitor the combustion state of the boiler. For example, the economizer outlet subcooling, exhaust gas temperature, nitrogen oxide concentration, etc. The expression of state time series data is as follows:

[0059] Y={y1,y2,y3,…,y m}

[0060] The first state time series data represents the state time series data within a preset time period traced back from the current moment. The preset time period can be set according to actual conditions and is not particularly limited here.

[0061] By adjusting the control variable X, the combustion state of the boiler can be changed, and the change of the combustion state of the boiler can be directly reflected in the state time series data Y, that is, the state time series data Y will change with the change of the control variable X. Accordingly, in order to make the subsequent state time series data of the boiler meet the energy efficiency conditions, the currently collected first state time series data can be used to predict the appropriate control variable.

[0062] Step S103: using the energy efficiency evaluation model, determine whether the expected state time series data after the boiler is optimized for energy efficiency according to the control variables meets the energy efficiency conditions.

[0063] The energy efficiency evaluation model is based on the business mechanism knowledge of the boiler and is obtained by using the counter-balance method.

[0064] The independent variable of the energy efficiency evaluation model is the state time series data, and the dependent variable is the energy efficiency score. The method of acquiring business mechanism knowledge can be selected according to the actual situation and is not specifically limited here.

[0065] Meeting the energy efficiency condition is specifically manifested in that the energy efficiency score output by the energy efficiency evaluation model is greater than the energy efficiency threshold. The energy efficiency threshold can be set according to actual conditions and is not particularly limited here.

[0066] By setting appropriate energy efficiency thresholds, energy efficiency evaluation is performed on the expected state time series data input into the energy efficiency evaluation model, and an energy efficiency score is output.

[0067] Step S104: In response to the expected state time series data satisfying the energy efficiency condition, the energy efficiency of the boiler is optimized based on the control variable; in response to the expected state time series data not satisfying the energy efficiency condition, the parameter value of at least one target parameter in the expected state time series data is adjusted, and the energy efficiency evaluation model is used to re-determine whether the adjusted expected state time series data satisfies the energy efficiency condition.

[0068] Step S105 , in response to the adjusted expected state time series data satisfying the energy efficiency condition, inputting the adjusted expected state time series data into a control variable determination model, and outputting control variables matching the expected state time series data.

[0069] In this embodiment, the control variable determination model is used to obtain the control variable matching the first state time series data, and then the energy efficiency evaluation model is used to determine whether the expected state time series data after optimizing the boiler according to the control variable meets the energy efficiency conditions. If the energy efficiency conditions are not met, the expected state time series data is adjusted, and the control variable determination model is re-entered to obtain the control variable matching the adjusted expected state time series data, which is used for energy efficiency optimization of the boiler to ensure the effectiveness of the control variable, that is, to ensure that after optimizing the boiler according to the control variable, the expected state time series data can meet the demand, thereby improving the effectiveness of energy efficiency optimization. Energy efficiency optimization can ensure that the boiler operates in the best state, can reduce slag generation, extend boiler life, reduce equipment maintenance costs, and improve the production efficiency of the boiler, thereby effectively improving combustion efficiency.

[0070] In one embodiment, historical state time series data of the boiler is mined, and quasi-steady-state data in the state time series data is extracted as training samples.

[0071] The training samples include state time series data labeled with control variables, and each state time series data is labeled with a corresponding control variable.

[0072] Select multiple parameters from the control variables that can be used to judge whether the boiler is in steady-state combustion as the quasi-steady-state judgment criteria. The number of selected parameters can be selected according to the actual situation and is not specifically limited here. According to the parameter values ​​of multiple parameters in the control variables, filter the quasi-steady-state data from the state time series data. Quasi-steady-state data refers to state time series data with small fluctuations. The quasi-steady-state data screening conditions are as follows:

[0073]

[0074] Among them, p1, p2, …, p l for x1,x2,…,x l The corresponding fluctuation threshold, x l,t For x lThe parameter value of the control variable at time t, x l,t-1 For x l The parameter value of the control variable at time t-1, l is the number of control variables, x l is the lth control variable.

[0075] The number of control variables can be set according to actual conditions and is not particularly limited here.

[0076] After obtaining N consecutive control variables that meet the above-mentioned quasi-steady-state data screening conditions, the control variables and their corresponding state time series data are combined into N quasi-steady-state data as training samples for the control variable determination model, and the training samples include state time series data marked with control variables. The value of N is selected according to the actual situation and is not particularly limited here.

[0077] The state time series data is used as feature data, the control scalar is used as label data, and the artificial intelligence algorithm is used to train the control variable determination model, which can predict the combustion state of the boiler. The input of the control variable determination model is the state time series data, and the output is the control variable.

[0078] Compared with the traditional extraction of boiler combustion data, quasi-steady-state data is obtained based on the state analysis of several important individual measuring points. It not only ensures the extraction of relatively stable state data to eliminate or reduce data fluctuations and errors caused by non-steady-state conditions, making the analysis more accurate, but also avoids the problem of traditional data extraction leading to overly complex calculations or overly harsh conditions resulting in too little data to meet the requirements.

[0079] In this embodiment, the training samples are screened based on the parameter values ​​of multiple control variables, so as to reduce the complexity of the data of the training samples and improve the accuracy of the control variable determination model constructed based on the training samples.

[0080] In one embodiment, step S104 includes: determining a target parameter from the expected state time series data, iteratively adjusting the parameter value of the target parameter in the expected state time series data, and inputting the expected state time series data after each adjustment into the energy efficiency evaluation model, and re-determining whether the adjusted expected state time series data meets the energy efficiency condition, until the adjusted expected state time series data meets the energy efficiency condition, the iteration is stopped.

[0081] The target parameters include: target parameters whose parameter values ​​have the greatest impact on the energy efficiency score, or target parameters whose parameter values ​​have an impact on the energy efficiency score that is greater than an energy efficiency impact threshold.

[0082] After the target parameters are determined, the adjustment step size s is set. i , adjust the target parameters according to the following formula:

[0083]

[0084] Among them, y i ' is the parameter value of the adjusted target parameter, y i is the parameter value of the target parameter before adjustment, s i The adjustment step length can be set according to the actual situation and is not particularly limited here.

[0085] The iterative adjustment step can be implemented using an intelligent search algorithm. Specifically, an intelligent search algorithm suitable for combustion energy efficiency optimization methods can be reconstructed based on a swarm intelligence algorithm with stronger global search capabilities (such as a genetic algorithm or a particle swarm algorithm). The intelligent search algorithm approaches the optimal solution through continuous iteration and search, rather than obtaining the optimal solution through mathematical deduction. Therefore, it can handle complex, difficult-to-model multi-objective optimization problems, and is more suitable for handling boiler energy efficiency optimization problems.

[0086] The optimal solution can be the adjusted expected state time series data that meets the energy efficiency conditions, or the adjusted expected state time series data with the highest energy efficiency score. The specific solution can be selected according to actual conditions and is not particularly limited here.

[0087] Based on the swarm intelligence algorithm with stronger global search capability, a multi-level and multi-step search intelligent search algorithm suitable for combustion energy efficiency optimization method is reconstructed. It has good global search capability, can avoid falling into the local optimal solution, and can perform iterative adjustments in parallel. Parallelism greatly improves the solution speed of the intelligent search algorithm, especially when there are sufficient computing resources. The intelligent search algorithm can find multiple approximate optimal solutions more quickly.

[0088] Input the expected state time series data after each adjustment into the energy efficiency evaluation model to re-determine whether the adjusted expected state time series data meets the energy efficiency conditions. The expression of the adjusted expected state time series data is as follows:

[0089] Y'={y1,y2,y3,…, y i ',…,y m}

[0090] In this embodiment, the parameter values ​​of the target parameters in the expected state time series data are iteratively adjusted until the adjusted expected state time series data meets the energy efficiency conditions, thereby obtaining expected state time series data with better energy efficiency, thereby facilitating obtaining more effective control variables and improving the effectiveness of energy efficiency optimization.

[0091] In one embodiment, the step of determining the target parameter from the expected state time series data includes: changing the parameter value of a parameter in the expected state time series data to a baseline value each time and inputting it into an energy efficiency evaluation model; based on the energy efficiency scores corresponding to the parameter value before and after the change, screening out the target parameter with the greatest impact on the energy efficiency score.

[0092] In this embodiment, the parameter with the greatest impact on the energy efficiency score is determined as the target parameter, thereby determining the object to be adjusted in the expected state time series data, which facilitates adjustment of the expected state time series data so that the expected state time series data has a higher energy efficiency score.

[0093] In one embodiment, the step of determining the target parameter from the expected state time series data includes: changing the parameter value of a parameter in the expected state time series data to a baseline value each time and inputting it into an energy efficiency evaluation model; based on the energy efficiency scores before and after the parameter value is changed, screening out the target parameter whose parameter value has an impact on the energy efficiency score that is greater than the energy efficiency impact threshold.

[0094] In this embodiment, the target parameters are determined according to the influence of the energy efficiency score, so as to determine the object to be adjusted in the expected state time series data, so as to facilitate the adjustment of the expected state time series data so that the expected state time series data has a higher energy efficiency score.

[0095] In one embodiment, in response to the adjusted expected state time series data satisfying the energy efficiency condition, the combustion energy efficiency optimization method further includes: associatively storing the adjusted expected state time series data and the corresponding control variables.

[0096] The associated stored expected state time series data and the corresponding control variables are used to optimize the energy efficiency of the boiler in combination with the second state time series data, wherein the similarity between the second state time series data and the associated stored expected state time series data is greater than a similarity threshold.

[0097] Meeting the energy efficiency condition is specifically manifested as the energy efficiency score output by the energy efficiency evaluation model being greater than the energy efficiency threshold. The energy efficiency threshold can be set according to actual conditions and is not specifically limited here. By setting a suitable energy efficiency threshold, the expected state time series data that meets the energy efficiency condition is associated with the corresponding control variable and stored.

[0098] By associating and storing the expected state time series data that meet the energy efficiency conditions and the corresponding control variables, an optimal operating condition database is constructed to provide support for subsequent real-time optimization. When the optimal operating condition database is constructed, multiple expected state time series data that meet the energy efficiency conditions and the corresponding control variables are associated and stored in the optimal operating condition database, and the energy efficiency of the second state time series data of the coal-fired boiler can be optimized later.

[0099] After obtaining the second state time series data of the boiler, the expected state time series data and the corresponding control variables that are most similar to the second state time series data can be selected from the optimal operating condition database, and the control variables can be used to optimize the energy efficiency of the boiler. The more similar the expected state time series data and the second state time series data are, the more suitable the control variables corresponding to the expected state data are for optimizing the energy efficiency of the second state time series data of the boiler.

[0100] Specifically, data that can characterize the real-time combustion state of the boiler is obtained at every preset time interval Δt, and the second state time series data at a certain moment is screened out from the data. The expected state time series data that is most similar to the second state time series data is screened out from the optimal operating condition database, and the control variables corresponding to the expected state time series data that is most similar to the second state time series data are returned. The obtained control variables can be used as operating parameters for optimizing the energy efficiency of the boiler, and can also be used to guide on-site operators to adjust the boiler.

[0101] The preset time interval Δt can be selected according to actual conditions and is not particularly limited here.

[0102] In this embodiment, offline data is used to construct an optimal operating condition database, and the control variables that best match the real-time state time series data are screened online for real-time optimization to improve the timeliness of energy efficiency optimization, effectively solving the problem of limited real-time optimization of combustion energy efficiency, and thereby improving the effectiveness of energy efficiency optimization.

[0103] Combine the following Figure 2 The combustion energy efficiency optimization method is further explained.

[0104] See also Figure 2 First, historical combustion data is obtained. The historical combustion data is divided into state time series data and corresponding control variables. Quasi-steady-state analysis is performed through the control variables, and quasi-steady-state data is obtained by screening from the state time series data. Among them, the quasi-steady-state data is the state time series data that meets the above-mentioned quasi-steady-state data screening conditions.

[0105] The state time series data that meets the quasi-steady-state data screening conditions are input into the evaluation model M2, which is also the energy efficiency evaluation model. According to the energy efficiency score output by the evaluation model M2, it is judged whether the input state time series data meets the evaluation threshold. Meeting the evaluation threshold means meeting the energy efficiency conditions. If the input state time series data meets the evaluation threshold, the state time series data and the control variables obtained by inputting the state time series data into the prediction model M1 are used as the optimal operating condition data, and are associated and stored in the optimal operating condition database to provide support for subsequent real-time optimization.

[0106] If the input state time series data does not meet the evaluation threshold, a multi-level and multi-step energy efficiency optimization is performed, that is, the parameter value of at least one target parameter in the expected state time series data is adjusted, and the energy efficiency evaluation model is used to re-determine whether the adjusted expected state time series data meets the energy efficiency condition. Specifically, the target parameter is determined from the expected state time series data, the parameter value of the target parameter in the expected state time series data is iteratively adjusted, and the expected state time series data after each adjustment is input into the energy efficiency evaluation model, and the adjusted expected state time series data is re-determined whether it meets the energy efficiency condition, until the adjusted expected state time series data meets the energy efficiency condition and the iteration is stopped.

[0107] Acquire the real-time data of the boiler, which is also the real-time state time series data. Search the state time series data with high similarity to the real-time data in the optimal operating condition database. When the similarity between the real-time and the associated stored state time series data is greater than the similarity threshold, the associated stored state time series data and the corresponding control variables can be used to optimize the real-time energy efficiency of the boiler.

[0108] Therefore, the obtained associated stored state time series data and the corresponding control variables can be sent to the control system, and the control system optimizes the energy efficiency of the boiler according to the control variables so that the device is in the optimal operating state. The device usually represents a boiler.

[0109] The combustion energy efficiency optimization method is further explained below through specific application scenarios.

[0110] The historical combustion data of the past three months are obtained from the boiler control system of a power plant. The historical combustion data are divided into state time series data and corresponding control variables.

[0111] The primary air flow rate, outlet air-powder mixture temperature, rotary separator inverter speed, total primary air volume, oxygen content, and outlet superheat of the steam-water separator corresponding to the coal mills A, B, C, D, and E of the boiler are selected as control variables, and the NOx (nitrogen oxide) concentration at the inlet and outlet of the reactor of the boiler unit, the ammonia leakage concentration, and the outlet flue gas sulfur dioxide concentration are selected as state time series data. The preset time interval Δt=30s. It should be noted that the selection of the above control variables, state time series data, and preset time interval Δt is only for example, and can be selected according to actual conditions, and is not particularly limited here.

[0112] The main steam pressure, total coal quantity, main steam temperature and power generation of the generator set in the control variables are selected as the quasi-steady-state evaluation criteria, and 5 minutes is a quasi-steady-state stage. According to the quasi-steady-state data screening conditions, the quasi-steady-state data are screened from the state time series data to determine the training samples.

[0113] The xgboost (Gradient Boosting Decision Tree, an ensemble learning algorithm based on gradient boosting decision tree) in the machine learning algorithm is adopted, and the state time series data marked with control variables are used as training samples to build a control variable determination model.

[0114] Based on the business mechanism knowledge of the boiler, the energy efficiency evaluation model is obtained by the reverse balance method. By setting a suitable energy efficiency threshold, the expected state time series data input into the energy efficiency evaluation model is evaluated for energy efficiency, and an energy efficiency score is output. If the energy efficiency score is greater than the energy efficiency threshold, it means that the expected state time series data input into the energy efficiency evaluation model meets the energy efficiency conditions, and the control variables corresponding to the expected state time series data can be used to optimize the energy efficiency of the boiler. The energy efficiency threshold can be set according to the actual situation and is not specifically limited here.

[0115] When the expected state time series data meets the energy efficiency conditions, the expected state time series data and the corresponding control variables that meet the energy efficiency conditions are stored in an associated manner. By storing the expected state time series data and the corresponding control variables that meet the energy efficiency conditions in an associated manner, an optimal operating condition database is constructed to provide support for subsequent real-time optimization.

[0116] When the expected state time series data does not meet the energy efficiency condition, the parameter value of at least one target parameter in the expected state time series data is adjusted, and the energy efficiency evaluation model is used to re-determine whether the adjusted expected state time series data meets the energy efficiency condition.

[0117] Specifically, each time the parameter value of a parameter in the expected state time series data is changed to the baseline value and input into the energy efficiency evaluation model, the target parameter with the greatest impact on the energy efficiency score is selected according to the energy efficiency scores before and after the parameter value is changed. After the target parameter is determined, the adjustment step s is set. i , adjust the target parameters according to the following formula:

[0118]

[0119] Among them, y i ' is the parameter value of the adjusted target parameter, y i is the parameter value of the target parameter before adjustment, s i The adjustment step length can be set according to the actual situation and is not particularly limited here.

[0120] Input the expected state time series data after each adjustment into the energy efficiency evaluation model to re-determine whether the adjusted expected state time series data meets the energy efficiency conditions. If the adjusted expected state time series data meets the energy efficiency conditions, the expected state time series data and the corresponding control variables that meet the energy efficiency conditions are stored in association. If the adjusted expected state time series data does not meet the energy efficiency conditions, the target parameters are re-determined and the re-determined target parameters are adjusted until the expected state time series data that meets the energy efficiency conditions is obtained.

[0121] The expected state time series data is input into the control variable determination model to obtain the control variables corresponding to the input expected state time series data, and the expected state time series data and the corresponding control variables that meet the energy efficiency conditions are stored in an associated manner.

[0122] After obtaining the second state time series data, the Wasserstein distance is used as a measurement method to find the expected state time series data and corresponding control variables that are most similar to the second state time series data from the optimal working condition database, that is, to find the expected state time series data that is closest to the second state time series data, and return the control variables corresponding to the expected state time series data that is most similar to the second state time series data.

[0123] The most representative data in the second state time series data, such as the inlet NOx concentration and the ammonia injection amount, are selected to analyze the changes of the inlet NOx concentration and the ammonia injection amount during the iterative optimization process, such as Figure 3 and Figure 4 As shown, Figure 3 and Figure 4 The horizontal axis is the number of energy efficiency optimizations, and the vertical axis is the energy efficiency score. Figure 3 The red line in the figure shows the changing trend of inlet NOx concentration, and the blue line shows the relationship between the number of energy efficiency iterations and the energy efficiency score. Figure 4 The red line in the figure shows the changing trend of the amount of ammonia injection, and the blue line shows the relationship between the number of energy efficiency iterations and the energy efficiency score. During the combustion energy efficiency optimization process, the inlet NOx concentration and the amount of ammonia injection were significantly reduced, that is, the combustion energy efficiency of the boiler was significantly improved.

[0124] In this embodiment, the control variable determination model is used to obtain the control variable matching the first state time series data, and then the energy efficiency evaluation model is used to determine whether the expected state time series data after optimizing the boiler according to the control variable meets the energy efficiency conditions. If the energy efficiency conditions are not met, the expected state time series data is adjusted, and the control variable determination model is re-entered to obtain the control variable matching the adjusted expected state time series data, which is used for energy efficiency optimization of the boiler to ensure the effectiveness of the control variable, that is, to ensure that after optimizing the boiler according to the control variable, the expected state time series data can meet the demand, thereby improving the effectiveness of energy efficiency optimization. Energy efficiency optimization can ensure that the boiler operates in the best state, can reduce slag generation, extend boiler life, reduce equipment maintenance costs, and improve the production efficiency of the boiler, thereby effectively improving combustion efficiency.

[0125] Example 2

[0126] Corresponding to the aforementioned combustion energy efficiency optimization method embodiment, the present disclosure also provides an embodiment of a combustion energy efficiency optimization system.

[0127] Figure 5 A module schematic diagram of a combustion energy efficiency optimization system provided by an exemplary embodiment of the present disclosure, the combustion energy efficiency optimization system comprising:

[0128] An acquisition module 21 is used to acquire first state time series data of the boiler;

[0129] A first control variable determination module 22 is used to input the first state time series data into a control variable determination model, and output a control variable matching the first state time series data; wherein the control variable determination model is obtained by training a neural network with training samples, the training samples include state time series data marked with control variables, and the control variables are used for energy efficiency optimization of the boiler;

[0130] The energy efficiency evaluation module 23 is used to determine whether the expected state time series data after the energy efficiency optimization of the boiler according to the control variables meets the energy efficiency conditions by using the energy efficiency evaluation model; wherein the energy efficiency evaluation model is based on the business mechanism knowledge of the boiler and is obtained by the counter-balance method;

[0131] An energy efficiency optimization module 24, for optimizing the energy efficiency of the boiler based on the control variables in response to the expected state time series data satisfying the energy efficiency condition;

[0132] An adjustment module 25, configured to adjust a parameter value of at least one target parameter in the expected state time series data in response to the expected state time series data not satisfying the energy efficiency condition, and re-determine whether the adjusted expected state time series data satisfies the energy efficiency condition by using the energy efficiency evaluation model;

[0133] The second control variable determination module 26 is used for inputting the adjusted expected state time series data into the control variable determination model in response to the adjusted expected state time series data satisfying the energy efficiency condition, and outputting the control variable matching the expected state time series data.

[0134] In one embodiment, the adjustment module 25 includes:

[0135] A target parameter determination unit, used to determine the target parameter from the expected state time series data; wherein the target parameter includes: a target parameter whose parameter value has the greatest impact on the energy efficiency score, or a target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold;

[0136] The iterative adjustment unit is used to iteratively adjust the parameter value of the target parameter in the expected state time series data, and input the expected state time series data after each adjustment into the energy efficiency evaluation model, and re-determine whether the adjusted expected state time series data meets the energy efficiency condition, until the adjusted expected state time series data meets the energy efficiency condition, the iteration is stopped.

[0137] In one embodiment, the target parameter determination unit is specifically used to:

[0138] Each time, the parameter value of one parameter in the expected state time series data is changed to a reference value and input into the energy efficiency evaluation model;

[0139] According to the energy efficiency scores before and after the parameter value is changed, the target parameter whose parameter value has the greatest impact on the energy efficiency score is screened out, or the target parameter whose parameter value has an impact on the energy efficiency score greater than the energy efficiency impact threshold is screened out.

[0140] In one embodiment, the combustion energy efficiency optimization system further includes:

[0141] A storage module, configured to store the adjusted expected state time series data and the corresponding control variable in association with each other in response to the adjusted expected state time series data satisfying the energy efficiency condition;

[0142] The associated stored expected state time series data and the corresponding control variables are used to optimize the energy efficiency of the boiler in combination with the second state time series data; wherein the similarity between the second state time series data and the associated stored expected state time series data is greater than a similarity threshold.

[0143] In this embodiment, the control variable determination model is used to obtain the control variable matching the first state time series data, and then the energy efficiency evaluation model is used to determine whether the expected state time series data after optimizing the boiler according to the control variable meets the energy efficiency conditions. If the energy efficiency conditions are not met, the expected state time series data is adjusted, and the control variable determination model is re-entered to obtain the control variable matching the adjusted expected state time series data, which is used for energy efficiency optimization of the boiler to ensure the effectiveness of the control variable, that is, to ensure that after optimizing the boiler according to the control variable, the expected state time series data can meet the demand, thereby improving the effectiveness of energy efficiency optimization. Energy efficiency optimization can ensure that the boiler operates in the best state, can reduce slag generation, extend boiler life, reduce equipment maintenance costs, and improve the production efficiency of the boiler, thereby effectively improving combustion efficiency.

[0144] As for the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The system embodiment described above is only illustrative, wherein the units described as separate components may or may not be physically separated, and the components as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the disclosed solution.

[0145] Example 3

[0146] Figure 6 This is a structural schematic diagram of an electronic device shown in an example embodiment of the present disclosure, the electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor, and when the processor executes the computer program, the combustion energy efficiency optimization method described in any of the above embodiments is implemented. Figure 6 The electronic device 30 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0147] like Figure 6 As shown, the electronic device 30 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including the memory 32 and the processor 31).

[0148] The bus 33 includes a data bus, an address bus, and a control bus.

[0149] The memory 32 may include a volatile memory, such as a random access memory (RAM) 321 and / or a cache memory 322 , and may further include a read only memory (ROM) 323 .

[0150] The memory 32 may also include a program tool 325 (or utility) having a set (at least one) of program modules 324, such program modules 324 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0151] The processor 31 executes various functional applications and data processing by running the computer program stored in the memory 32, such as the combustion energy efficiency optimization method provided in any of the above embodiments.

[0152] The electronic device 30 may also communicate with one or more external devices 34 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 35. Furthermore, the electronic device 30 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 36. As shown, the network adapter 36 communicates with other modules of the electronic device 30 via a bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems, etc.

[0153] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0154] Example 4

[0155] The embodiments of the present disclosure further provide a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the combustion energy efficiency optimization method provided in any of the above embodiments is implemented.

[0156] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0157] Although the specific embodiments of the present disclosure are described above, those skilled in the art should understand that this is only an example, and the protection scope of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, but these changes and modifications all fall within the protection scope of the present disclosure.

Claims

1. A method for optimizing combustion energy efficiency, characterized in that: The combustion energy efficiency optimization method comprises: Acquire first state time series data of the boiler; Inputting the first state time series data into a control variable determination model, and outputting a control variable matching the first state time series data; wherein the control variable determination model is obtained by training a neural network with training samples, the training samples include state time series data marked with control variables, and the control variables are used for energy efficiency optimization of the boiler; Using an energy efficiency evaluation model, determining whether the expected state time series data after the energy efficiency of the boiler is optimized according to the control variable meets the energy efficiency conditions; wherein the energy efficiency evaluation model is based on the business mechanism knowledge of the boiler and is obtained by the counter-balance method; In response to the expected state time series data satisfying the energy efficiency condition, optimizing the energy efficiency of the boiler based on the control variable; In response to the expected state time series data not satisfying the energy efficiency condition, adjusting a parameter value of at least one target parameter in the expected state time series data, and re-determining whether the adjusted expected state time series data satisfies the energy efficiency condition by using the energy efficiency evaluation model; In response to the adjusted expected state time series data satisfying the energy efficiency condition, the adjusted expected state time series data is input into the control variable determination model, and a control variable matching the expected state time series data is output.

2. The combustion energy efficiency optimization method according to claim 1, characterized in that: The step of adjusting the parameter value of at least one target parameter in the expected state time series data comprises: Determine a target parameter from the expected state time series data; wherein the target parameter includes: a target parameter whose parameter value has the greatest impact on the energy efficiency score, or a target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold; Iteratively adjust the parameter value of the target parameter in the expected state time series data, and input the expected state time series data after each adjustment into the energy efficiency evaluation model, and re-determine whether the adjusted expected state time series data meets the energy efficiency condition, and stop the iteration until the adjusted expected state time series data meets the energy efficiency condition.

3. The combustion energy efficiency optimization method according to claim 2, characterized in that: The step of determining the target parameter from the expected state time series data comprises: Each time, the parameter value of one parameter in the expected state time series data is changed to a reference value and input into the energy efficiency evaluation model; According to the energy efficiency scores before and after the parameter value is changed, the target parameter whose parameter value has the greatest impact on the energy efficiency score is screened out, or the target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold is screened out.

4. The combustion energy efficiency optimization method according to claim 1, characterized in that: In response to the adjusted expected state time series data satisfying the energy efficiency condition, the combustion energy efficiency optimization method further includes: storing the adjusted expected state time series data and the corresponding control variables in an associated manner; The expected state time series data and the corresponding control variables stored in the association are used to optimize the energy efficiency of the boiler in combination with the second state time series data; wherein the similarity between the second state time series data and the expected state time series data stored in the association is greater than a similarity threshold.

5. A combustion energy efficiency optimization system, characterized in that: The combustion energy efficiency optimization system comprises: An acquisition module, used for acquiring first state time series data of the boiler; a first control variable determination module, configured to input the first state time series data into a control variable determination model, and output a control variable matching the first state time series data; wherein the control variable determination model is obtained by training a neural network with training samples, the training samples comprising state time series data marked with control variables, and the control variables are used for energy efficiency optimization of the boiler; An energy efficiency evaluation module, used to determine whether the expected state time series data after the energy efficiency optimization of the boiler according to the control variable meets the energy efficiency condition by using an energy efficiency evaluation model; wherein the energy efficiency evaluation model is based on the business mechanism knowledge of the boiler and is obtained by using a counter-balance method; an energy efficiency optimization module, configured to optimize the energy efficiency of the boiler based on the control variable in response to the expected state time series data satisfying the energy efficiency condition; an adjustment module, configured to adjust a parameter value of at least one target parameter in the expected state time series data in response to the expected state time series data not satisfying the energy efficiency condition, and re-determine whether the adjusted expected state time series data satisfies the energy efficiency condition by using the energy efficiency evaluation model; The second control variable determination module is used for inputting the adjusted expected state time series data into the control variable determination model in response to the adjusted expected state time series data satisfying the energy efficiency condition, and outputting the control variable matching the expected state time series data.

6. The combustion energy efficiency optimization system according to claim 5, characterized in that: The adjustment module comprises: A target parameter determination unit, configured to determine a target parameter from the expected state time series data; wherein the target parameter includes: a target parameter whose parameter value has the greatest impact on the energy efficiency score, or a target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold; An iterative adjustment unit is used to iteratively adjust the parameter value of the target parameter in the expected state time series data, and input the expected state time series data after each adjustment into the energy efficiency evaluation model, and re-determine whether the adjusted expected state time series data meets the energy efficiency condition, until the iteration is stopped when the adjusted expected state time series data meets the energy efficiency condition.

7. The combustion energy efficiency optimization system according to claim 6, characterized in that: The target parameter determination unit is specifically used for: Each time, the parameter value of one parameter in the expected state time series data is changed to a reference value and input into the energy efficiency evaluation model; According to the energy efficiency scores before and after the parameter value is changed, the target parameter whose parameter value has the greatest impact on the energy efficiency score is screened out, or the target parameter whose parameter value has an impact on the energy efficiency score greater than an energy efficiency impact threshold is screened out.

8. The combustion energy efficiency optimization system according to claim 5, characterized in that: The combustion energy efficiency optimization system also includes: A storage module, configured to store the adjusted expected state time series data and the corresponding control variable in association with each other in response to the adjusted expected state time series data satisfying the energy efficiency condition; The expected state time series data and the corresponding control variables stored in the association are used to optimize the energy efficiency of the boiler in combination with the second state time series data; wherein the similarity between the second state time series data and the expected state time series data stored in the association is greater than a similarity threshold.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, characterized in that: When the processor executes the computer program, the combustion energy efficiency optimization method according to any one of claims 1 to 4 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the combustion energy efficiency optimization method according to any one of claims 1 to 4 is implemented.

Images