A power plant coal-fired big data analysis method based on adaptive support vector machine

Abstract

The application discloses a power plant coal-fired big data analysis method based on an adaptive support vector machine, which comprises the following steps: using a pre-determined coal quality data analysis model to perform real-time analysis on a coal type to be analyzed, and obtaining a coal quality analysis result; wherein the coal quality data analysis model comprises a plurality of optimal support vector machines for analyzing coal quality data of different coal types; according to coal quality historical data, using a genetic algorithm to train and optimize the support vector machines, obtaining the optimal support vector machines, using the obtained optimal support vector machines to perform real-time analysis on a new coal type, according to the deviation between the real-time analysis value of the new coal type and experimental analysis data, determining whether the optimal support vector machines need to be self-updated, and according to the number of sample sets, determining whether a new optimal support vector machine is obtained by training and optimization. The application can perform real-time analysis on various coal types to obtain coal quality data, and can improve the data analysis function and information intelligent level of a fuel management system.

Description

TECHNICAL FIELD

[0001] The present application relates to a power plant coal big data analysis method based on an adaptive support vector machine, and belongs to the technical field of unit fuel big data analysis and management. BACKGROUND

[0002] Fuel expenditure accounts for more than 70% of the cost of thermal power generating units, and effective management thereof is a key link for thermal power enterprises to reduce production and operation costs, improve efficiency, and resist market risks. Influenced by the coal supply market, the coal type of thermal power generating units is variable, and the instability of the coal quality of the coal supply has become an inevitable reality. The traditional fuel management system has already been unable to adapt to the current coal supply changes. It is of certain practical significance to establish a new intelligent fuel management system within a regional power grid and further expand the functions and applications of the fuel management system, with the units connected to the regional power grid as the target.

[0003] Fuel management involves many aspects, mainly including planning, procurement, inspection, in-furnace coal quality detection, consumption management, statistical analysis, and ledger management. The data involved are large in quantity and numerous in type, and the management and statistical analysis of the data are complicated. Therefore, data processing and application analysis in fuel management are particularly important. The main functions of the current fuel management system are still conventional statistics, input, query, and report, and a large amount of data resources has not been fully mined and utilized. At present, in-furnace coal quality detection still relies on laboratory detection, which can reliably and comprehensively obtain coal quality data, but the detection period is long, and the coal quality of the in-furnace coal type cannot be obtained in real time, which cannot meet the needs of establishing a new intelligent fuel management system within a regional power grid and is not conducive to further expanding the functions and applications of the fuel management system. SUMMARY

[0004] The present application aims to overcome the deficiencies in the prior art and provide a power plant coal big data analysis method based on an adaptive support vector machine, which can obtain the coal quality of the in-furnace coal type in real time. To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0005] In a first aspect, the present application provides a power plant coal big data analysis method based on an adaptive support vector machine, which comprises:

[0006] real-time analysis of the coal type to be analyzed by using a predetermined coal quality data analysis model to obtain a coal quality analysis result;

[0007] The coal quality data analysis model comprises a plurality of optimal support vector machines for analyzing the coal quality data of different coal types, and the coal quality data analysis model is obtained by the following steps:

[0008] obtaining coal quality historical data and experimental analysis data of all new coal types;

[0009] Based on the acquired historical coal quality data, a genetic algorithm is used to train and optimize a pre-established support vector machine to obtain the optimal support vector machine.

[0010] The optimal support vector machine is used to perform real-time analysis on the new coal type. Based on the deviation between the real-time analysis values ​​and the experimental analysis data, it is determined whether the optimal support vector machine needs to be updated.

[0011] When the deviation exceeds the preset deviation threshold E cs The optimal support vector machine continues to run and stores the data for this new coal type in the sample set;

[0012] When the deviation is less than the preset deviation threshold E cs The best support vector machine performs self-update;

[0013] Determine whether to train and optimize to obtain a new optimal support vector machine based on the size of the sample set:

[0014] When the number of samples exceeds a set threshold q Based on the sample set, a pre-established support vector machine is trained and optimized to obtain a new optimal support vector machine, which is then analyzed in real time and updated automatically for new coal types.

[0015] When the number of samples is less than a set threshold q The existing best support vector machine continues to run.

[0016] In conjunction with the first aspect, furthermore, the coal quality data obtained from coal quality analysis is the received basis lower heating value - Q. net,ar Ash content on dry basis - A d Moisture - M ar Dry, ash-free, volatile matter-free - V daf and dry basis sulfur content - S t,d .

[0017] Building upon the first aspect, further, a genetic algorithm is used to train and optimize a pre-established support vector machine prototype to obtain the optimal support vector machine, including:

[0018] The acquired historical coal quality data was divided into training samples and test samples;

[0019] The training sample data is divided into input variables. x and output variables y Substitute the pre-built support vector machine;

[0020] The parameters ε and C in the support vector machine are used as optimization variables, and the root mean square error of the support vector machine test is used as the optimization objective function.

[0021] Based on the optimization variables and objective function, a genetic algorithm is used to optimize the training process of a pre-established support vector machine, including:

[0022] Initialize the population size to m ;

[0023] Training the Support Vector Machine: Based on the training samples, a genetic algorithm is used to perform selection, crossover, and mutation genetic calculations to obtain... m The optimized parameters ε and C are then input into the support vector machine prototype for regression training, yielding... m One regression support vector machine;

[0024] Testing Support Vector Machines: Using test sample pairs m Test each regression support vector machine and select the optimal regression support vector machine.

[0025] Based on whether the set number of generations of inheritance has been reached. p The algorithm determines whether the genetic algorithm has terminated before the set number of generations has been reached. p Continue training and testing the support vector machine until the set number of generations is reached. p Generation; when the number of generations of inheritance reaches a set value. p The genetic algorithm is terminated at the next generation.

[0026] After the genetic algorithm terminates, the optimal regression support vector machine with the smallest regression error during testing is selected as the best support vector machine.

[0027] In conjunction with the first aspect, further, the training samples and test samples are: the obtained... n The top 2 of historical coal quality data n Three samples were divided into training samples, and then... n Three samples were divided into test samples.

[0028] In conjunction with the first aspect, furthermore, the baseline low-order calorific value in the training samples is used as the output variable of the pre-established support vector machine. y The ash content, moisture content, volatile matter on a dry basis, and sulfur content on a dry basis were used as input variables for a pre-established support vector machine. x .

[0029] In conjunction with the first aspect, the calculation method for the deviation between the real-time analysis value and the experimental analysis data of the new coal type is as follows:

[0030] Using optimal support vector machines for new coal types Real-time analysis is performed; the optimal support vector machine is represented by the following formula:

[0031] (1)

[0032] In equation (1), , For the Lagrange multipliers in the support vector machine model, For kernel function, This is the deviation coefficient;

[0033] The deviation between the real-time analysis value and the experimental analysis data of the new coal type is calculated using the following formula:

[0034] (2)

[0035] in, This indicates the deviation between the real-time analysis value and the experimental analysis data for the new coal type; This represents experimental analysis data; This indicates the analytical values ​​calculated by the model.

[0036] In conjunction with the first aspect, the sample set further includes three sets obtained by further dividing the training samples obtained from the acquired historical coal quality data: a data vector set E representing deviations exceeding the boundary, a data vector set S representing deviations near the boundary, and a data vector set R representing deviations within the boundary;

[0037] The data vector set E is represented as: (3)

[0038] The data vector set S is represented as: (4)

[0039] The data vector set R is represented as: (5)

[0040] In equations (3), (4), and (5), The deviation values ​​of the Lagrange multipliers. ; To account for the discrepancy between real-time analysis values ​​and experimental analysis data for the new coal type, .

[0041] In conjunction with the first aspect, the rule for the optimal support vector machine to perform self-updates is further as follows:

[0042] like ,calculate ,like Then the data for the new coal type will be stored in the data vector set E;

[0043] like Then the data for the new coal type will be stored in the data vector set S;

[0044] For samples in the original data vector set S, if If so, then move the sample to the data vector set E; if If so, then move the sample to the data vector set R;

[0045] For a sample in the original data vector set R, if If so, then move the sample to the data vector set S;

[0046] For samples in the original data vector set E, if If so, then the sample is moved to the data vector set S.

[0047] In conjunction with the first aspect, preferably, if the data vector set S remains unchanged, then each data vector set is modified; if the data vector set S changes, then it is determined whether to train and optimize to obtain a new optimal support vector machine based on the number of samples.

[0048] Secondly, the present invention provides a power plant coal-fired big data analysis device based on adaptive support vector machine, including a processor and a storage medium;

[0049] The storage medium is used to store instructions;

[0050] The processor is configured to operate according to the instructions to perform the steps of the method described in the first aspect.

[0051] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0052] Compared with existing technologies, the beneficial effects achieved by the big data analysis method for coal-fired power plants based on adaptive support vector machines provided in this invention include:

[0053] This invention utilizes a pre-determined coal quality data analysis model to perform real-time analysis of the coal type to be analyzed, thereby obtaining coal quality analysis results. The coal quality data analysis model of this invention employs multiple optimal support vector machines for analyzing coal quality data of different coal types, which can adapt to various coal types and enable real-time analysis to obtain coal quality data of each coal type.

[0054] This invention uses historical coal quality data and a genetic algorithm to train and optimize a support vector machine (SVM) to obtain an optimal SVM. This optimal SVM is then used for real-time analysis of new coal types. Based on the deviation between the real-time analysis values ​​and experimental data for the new coal type, it determines whether the optimal SVM needs self-updating. The number of samples determines whether to train and optimize to obtain a new optimal SVM. This invention achieves real-time evaluation and adaptive updating of the model, resulting in more reliable and comprehensive analysis results. It is beneficial for further improving the intelligent data analysis function of the fuel management system, expanding its functions and applications, and meeting the needs of establishing a new intelligent fuel management system within a regional power grid. Attached Figure Description

[0055] Figure 1 This is a flowchart of a big data analysis method for coal-fired power plants based on adaptive support vector machines, provided in Embodiment 1 of the present invention.

[0056] Figure 2 This is a flowchart of obtaining the optimal support vector machine in a big data analysis method for coal-fired power plants based on adaptive support vector machines, provided in Embodiment 1 of the present invention.

[0057] Figure 3 This is the optimal support vector machine self-updating rule in the big data analysis method for coal-fired power plants based on adaptive support vector machines provided in Embodiment 1 of the present invention;

[0058] Figure 4 This is a calculation effect diagram of the method described in the present invention in Embodiment 1 of the present invention. Detailed Implementation

[0059] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0060] Example 1:

[0061] like Figure 1 As shown, this embodiment of the invention provides a method for analyzing big data on coal-fired power plant combustion based on adaptive support vector machines, including:

[0062] A pre-determined coal quality data analysis model is used to perform real-time analysis on the coal type to be analyzed, and the coal quality analysis results are obtained.

[0063] The coal quality data analysis model includes multiple optimal support vector machines for analyzing coal quality data of different coal types. The coal quality data analysis model is obtained through the following steps:

[0064] Acquire historical coal quality data and experimental analysis data for all new coal types;

[0065] Based on the acquired historical coal quality data, a genetic algorithm is used to train and optimize a pre-established support vector machine to obtain the optimal support vector machine.

[0066] The optimal support vector machine is used to perform real-time analysis on the new coal type. Based on the deviation between the real-time analysis values ​​and the experimental analysis data, it is determined whether the optimal support vector machine needs to be updated.

[0067] When the deviation exceeds the preset deviation threshold E cs The optimal support vector machine continues to run and stores the data for this new coal type in the sample set;

[0068] When the deviation is less than the preset deviation threshold E cs The best support vector machine performs self-update;

[0069] Determine whether to train and optimize to obtain a new optimal support vector machine based on the size of the sample set:

[0070] When the number of samples exceeds a set threshold q Based on the sample set, a pre-established support vector machine is trained and optimized to obtain a new optimal support vector machine, which is then analyzed in real time and updated automatically for new coal types.

[0071] When the number of samples is less than a set threshold q The existing best support vector machine continues to run.

[0072] The invention utilizes access data from the fuel management system of a thermal power unit established within a regional power grid as an example.

[0073] Step 1: Obtain historical coal quality data and experimental analysis data for all new coal types.

[0074] Ninety historical data points for coal quality composition analysis of different coal types were collected from the fuel management system of thermal power units, and experimental analysis data for new coal types were obtained. Specifically, the coal quality data obtained from the coal quality analysis is the received basis lower heating value - Q. net,ar Ash content on dry basis - A d Moisture - M ar Dry, ash-free, volatile matter-free - V daf and dry basis sulfur content - S t,d .

[0075] Step 2: Based on the obtained historical coal quality data, use a genetic algorithm to train and optimize the pre-established support vector machine to obtain the optimal support vector machine.

[0076] Specific steps are as follows Figure 2 As shown:

[0077] Step 2.1: Divide the first 60 samples of the 90 historical coal quality data points obtained into training samples and the last 30 samples into test samples.

[0078] Step 2.2: Use the baseline calorific value in the training samples as the output variable of the pre-established support vector machine. y The ash content, moisture content, volatile matter on a dry basis, and sulfur content on a dry basis were used as input variables for a pre-established support vector machine. x Input variables x and output variables y Substitute the data into the pre-built support vector machine.

[0079] Step 2.3: Use the parameters ε and C in the support vector machine as optimization variables, and use the root mean square error of the support vector machine test. , which serves as the objective function for optimization.

[0080] Step 2.4: Based on the optimization variables and the optimization objective function, the training process of the pre-established support vector machine is optimized using a genetic algorithm. The specific steps are as follows:

[0081] Step 2.4.1: Initialize the population size as follows m .

[0082] Step 2.4.2: Training the Support Vector Machine: Based on the training samples, a genetic algorithm is used to perform selection, crossover, and mutation genetic calculations to obtain... m The optimized parameters ε and C are then input into the support vector machine prototype for regression training, yielding... m A regression support vector machine.

[0083] Step 2.4.3: Testing the Support Vector Machine: Using test sample pairs m Test several regression support vector machines and select the optimal one.

[0084] Step 2.4.4: Determine whether the genetic algorithm should terminate based on whether the number of generations has reached the set number of generations 200. If the number of generations has not reached 200, return to step 2.4.2 to continue training and testing the support vector machine until the number of generations has reached the set number of generations 200. If the number of generations has reached the set number of generations, terminate the genetic algorithm.

[0085] Step 2.5: After the genetic algorithm terminates, select the optimal regression support vector machine with the smallest regression error during testing as the best support vector machine.

[0086] Step 3: Use the optimal support vector machine to perform real-time analysis on the new coal type, predict the lower heating value of the new coal type, and determine whether the optimal support vector machine needs to be updated based on the deviation between the real-time analysis value of the lower heating value of the new coal type and the experimental analysis data.

[0087] When the deviation exceeds the preset deviation threshold E cs The optimal support vector machine continues to run and stores the data for this new coal type in the sample set;

[0088] When the deviation is less than the preset deviation threshold E cs The best support vector machine performs self-updates.

[0089] Specifically, the calculation method for the deviation between the real-time analysis value and the experimental analysis data of the new coal type is as follows:

[0090] Using optimal support vector machines for new coal types Real-time analysis is performed; the optimal support vector machine is represented by the following formula:

[0091] (1)

[0092] In equation (1), , For the Lagrange multipliers in the support vector machine model, For kernel function, This is the deviation coefficient;

[0093] The deviation between the real-time analysis value and the experimental analysis data of the new coal type is calculated using the following formula:

[0094] (2)

[0095] in, This indicates the deviation between the real-time analysis value and the experimental analysis data for the new coal type; This represents experimental analysis data; This indicates the analytical values ​​calculated by the model.

[0096] The sample set includes three sets obtained by further dividing the acquired historical coal quality data into training samples: E, representing data vector set E indicating deviations exceeding the boundary; S, representing data vector set S indicating deviations near the boundary; and R, representing data vector set R indicating deviations within the boundary.

[0097] The data vector set E is represented as: (3)

[0098] The data vector set S is represented as: (4)

[0099] The data vector set R is represented as: (5)

[0100] In equations (3), (4), and (5), The deviation values ​​of the Lagrange multipliers. ; To account for the discrepancy between real-time analysis values ​​and experimental analysis data for the new coal type, .

[0101] like Figure 3 As shown, the rule for self-updating of the optimal support vector machine is:

[0102] like ,calculate ,like Then the data for the new coal type will be stored in the data vector set E;

[0103] like Then the data for the new coal type will be stored in the data vector set S;

[0104] For samples in the original data vector set S, if If so, then move the sample to the data vector set E; if If so, then move the sample to the data vector set R;

[0105] For a sample in the original data vector set R, if If so, then move the sample to the data vector set S;

[0106] For samples in the original data vector set E, if If so, then the sample is moved to the data vector set S.

[0107] In this embodiment, a pre-set deviation threshold E cs =10%.

[0108] Step 4: Determine whether to train and optimize to obtain a new optimal support vector machine based on the number of data vectors in the data vector set S.

[0109] When the quantity exceeds the set threshold q Based on the sample set, a pre-established support vector machine is trained and optimized to obtain a new optimal support vector machine, which is then analyzed in real time and updated automatically for new coal types.

[0110] When the quantity is less than the set threshold q The existing best support vector machine continues to run.

[0111] like Figure 4The figures show the laboratory analysis values ​​of coal quality, the regression values ​​of the proposed scheme, and the regression values ​​of the traditional support vector machine. A total of 30 types of coal were selected for testing. It can be seen that the adaptive support vector machine model predicts calorific values ​​very close to the laboratory analysis values ​​for most coal types, especially avoiding large regression errors at certain points, indicating better regression performance. Applying it to a fuel management system can improve the system's data analysis capabilities and information intelligence level.

[0112] Example 2:

[0113] This invention provides a big data analysis device for coal-fired power plants based on adaptive support vector machines, including a processor and a storage medium;

[0114] The storage medium is used to store instructions;

[0115] The processor is configured to operate according to the instructions to execute the steps of the method described in Embodiment 1.

[0116] Example 3:

[0117] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0118] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0119] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0120] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0121] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0122] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for analyzing big data on coal-fired power plant combustion based on adaptive support vector machines, characterized in that, The method includes: A pre-determined coal quality data analysis model is used to perform real-time analysis on the coal type to be analyzed, and the coal quality analysis results are obtained. The coal quality data analysis model includes multiple optimal support vector machines for analyzing coal quality data of different coal types. The coal quality data analysis model is obtained through the following steps: Acquire historical coal quality data and experimental analysis data for all new coal types; Based on the acquired historical coal quality data, a genetic algorithm is used to train and optimize a pre-established support vector machine to obtain the optimal support vector machine; wherein, the process of using a genetic algorithm to train and optimize a pre-established support vector machine prototype to obtain the optimal support vector machine includes: The acquired historical coal quality data is divided into training samples and test samples; wherein, the training samples and test samples are: the acquired... n The top 2 of historical coal quality data n Three samples were divided into training samples, and then... n Three samples were divided into test samples; The training sample data is divided into input variables. x and output variables y Substitute the values ​​into a pre-established support vector machine; where the baseline heat generation from the training samples is used as the output variable of the pre-established support vector machine. y The ash content, moisture content, volatile matter on a dry basis, and sulfur content on a dry basis were used as input variables for a pre-established support vector machine. x ; The parameters ε and C in the support vector machine are used as optimization variables, and the root mean square error of the support vector machine test is used as the optimization objective function. Based on the optimization variables and objective function, a genetic algorithm is used to optimize the training process of a pre-established support vector machine, including: Initialize the population size to m ; Training the Support Vector Machine: Based on the training samples, a genetic algorithm is used to perform selection, crossover, and mutation genetic calculations to obtain... m The optimized parameters ε and C are then input into the support vector machine prototype for regression training, yielding... m One regression support vector machine; Testing Support Vector Machines: Using test sample pairs m Test each regression support vector machine and select the optimal regression support vector machine. Based on whether the set number of generations of inheritance has been reached. p The algorithm determines whether the genetic algorithm has terminated before the set number of generations has been reached. p Continue training and testing the support vector machine until the set number of generations is reached. p Generation; when the number of generations of inheritance reaches a set value. p The genetic algorithm is terminated at the next generation. After the genetic algorithm terminates, the optimal regression support vector machine with the smallest regression error during testing is selected as the best support vector machine. The optimal support vector machine (SVM) is used to perform real-time analysis on the new coal type. Based on the deviation between the real-time analysis values ​​and the experimental analysis data, it is determined whether the optimal SVM needs to be updated. The calculation method for the deviation between the real-time analysis values ​​and the experimental analysis data is as follows: Using optimal support vector machines for new coal types Real-time analysis is performed; the optimal support vector machine is represented by the following formula: （1） In equation (1), , For the Lagrange multipliers in the support vector machine model, For kernel function, This is the deviation coefficient; The deviation between the real-time analysis value and the experimental analysis data of the new coal type is calculated using the following formula: （2） in, This indicates the deviation between the real-time analysis value and the experimental analysis data for the new coal type; This represents experimental analysis data; This indicates that the analytical values ​​are calculated by the model; When the deviation exceeds the preset deviation threshold E cs The optimal support vector machine continues to run and stores the data for the new coal type in the sample set; When the deviation is less than the preset deviation threshold E cs The best support vector machine performs self-update; Determine whether to train and optimize to obtain a new optimal support vector machine based on the size of the sample set: When the number of samples exceeds a set threshold q Based on the sample set, a pre-established support vector machine is trained and optimized to obtain a new optimal support vector machine, which is then analyzed in real time and updated automatically for new coal types. When the number of samples is less than a set threshold q The existing best support vector machine continues to run.

2. The method for analyzing big data on coal-fired power plant combustion based on adaptive support vector machine according to claim 1, characterized in that, The coal quality data obtained from the coal quality analysis are the net calorific value on the received basis, ash content on the dry basis, moisture content, volatile matter on the dry ash-free basis, and sulfur content on the dry basis.

3. The method for analyzing big data on coal-fired power plant combustion based on adaptive support vector machine according to claim 1, characterized in that, The sample set includes three sets obtained by further dividing the acquired historical coal quality data into training samples: a data vector set representing deviations exceeding the boundary. E , representing a data vector set where the deviation is near the boundary. S And represent the data vector set whose deviation is within the boundary. R ; Data vector set E Represented as: (3) Data vector set S Represented as: (4) Data vector set R Represented as: (5) In equations (3), (4), and (5), The deviation values ​​of the Lagrange multipliers. ; To account for the discrepancy between real-time analysis values ​​and experimental analysis data for the new coal type, .

4. The method for analyzing big data on coal-fired power plant combustion based on adaptive support vector machine according to claim 3, characterized in that, The rules for self-updating the optimal support vector machine are as follows: like ,calculate ,like The data for the new coal type will then be stored in a data vector set. E middle; like The data for the new coal type will then be stored in a data vector set. S middle; For the original data vector set S The samples in, if Then move the sample to the data vector set. E In the middle; if Then move the sample to the data vector set. R middle; For the original data vector set R If the samples in the sample, Then move the sample to the data vector set. S middle; For the original data vector set E If the samples in the sample, Then move the sample to the data vector set. S middle.

5. A power plant coal-fired big data analysis device based on adaptive support vector machine, characterized in that, Including processor and storage media; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1 to 4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1 to 4.

Images