Boiler operation combustion diagnosis method and system based on multi-source heterogeneous information mining

Abstract

The invention discloses a boiler operation combustion diagnosis method and system based on multi-source heterogeneous information mining. The method comprises the steps that boiler thermal parameters, a hearth combustion flame image and a hearth numerical physical field are obtained; respectively carrying out feature extraction on the boiler thermal parameter data, the hearth combustion flame image data and the hearth numerical physical field data to obtain thermal parameter features, flame parameter features and numerical physical field features; fusing the thermal parameter characteristics, the flame parameter characteristics and the numerical physical field characteristic data to construct a multi-source heterogeneous characteristic parameter set; performing quantitative evaluation on the multi-source heterogeneous characteristic parameter set by adopting a combustion state diagnosis grade evaluation quantitative strategy, calculating a combustion state diagnosis comprehensive evaluation index, and calculating a boiler operation combustion state diagnosis category through the combustion state diagnosis comprehensive evaluation index; according to the method, the uncertainty of judgment of the boiler operation combustion state caused by human factors can be overcome, and the accuracy of the judgment result of the boiler operation combustion state is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of thermal power generation, and in particular relates to a boiler operation combustion diagnosis method and system based on multi-source heterogeneous information mining. Background Art

[0002] As one of the core equipment in thermal power generating units, the combustion performance of coal-fired boilers directly affects the overall power generation efficiency of thermal power generating units. In particular, as thermal power generating units play an increasingly prominent role in peak load regulation of power grids, thermal power generating units need to adjust the operating status of the boilers according to the power generation load instructions of the power grid to maintain the balance of the power system. The combustion status of the boiler directly affects the operation adjustment of the boiler. Therefore, the diagnosis of the combustion status of coal-fired boilers becomes particularly important.

[0003] In the prior art, the monitoring of boiler combustion status mainly relies on the power operator to obtain a large amount of thermal measurement point data and flame monitoring industrial television images from the distributed control system (DCS) or SIS system configured by the boiler unit itself, combined with their own professional experience, to judge the boiler combustion status and control the operation of the boiler. However, due to the complexity and variability of the combustion state in the boiler furnace, the current judgment of the boiler operation combustion state based on human experience has great uncertainty, which directly affects the accuracy of the judgment results of the boiler operation combustion state, and ultimately affects the boiler's control and production efficiency. Summary of the invention

[0004] The first objective of the present invention is to overcome the deficiencies of the prior art and to provide a boiler operation combustion diagnosis method and system based on multi-source heterogeneous information mining. The method obtains three types of multi-source heterogeneous data from different sources and with different data structure characteristics, namely, thermal parameter data, furnace combustion flame image data and furnace numerical physical field data, and calculates the boiler's combustion state diagnosis comprehensive evaluation index based on the multi-source heterogeneous data combined with a pre-constructed combustion state diagnosis grade evaluation quantification strategy. Finally, the current operating combustion state diagnosis category of the boiler is accurately diagnosed based on the combustion state diagnosis comprehensive evaluation index. This overcomes the large uncertainty in the judgment of the boiler's operating combustion state due to human experience, improves the accuracy of the boiler's operating combustion state judgment result, and also improves the boiler's operation intelligence and diagnosis level.

[0005] The second object of the present invention is to provide a boiler operation combustion diagnosis system based on multi-source heterogeneous information mining.

[0006] A third object of the present invention is to provide a computer device.

[0007] A fourth object of the present invention is to provide a storage medium.

[0008] The first object of the present invention can be achieved by adopting the following technical scheme:

[0009] A boiler operation combustion diagnosis method based on multi-source heterogeneous information mining includes the following steps:

[0010] S1. Obtain the current thermal parameter data of the boiler, the furnace combustion flame image data and the furnace numerical physical field data;

[0011] S2. Acquire corresponding thermal parameter characteristics according to the thermal parameter data;

[0012] S3, preprocessing the furnace combustion flame image data, performing feature extraction on the preprocessed furnace combustion flame image, and obtaining flame image features corresponding to the furnace combustion flame image data; and calculating flame parameter features of the boiler according to the flame image features;

[0013] S4, using the flame image feature obtained in step S3 to correct the furnace numerical physical field data, to obtain the furnace numerical physical field distribution parameters, and to calculate the numerical physical field features according to the furnace numerical physical field distribution parameters;

[0014] S5. Data fusion is performed on the thermal parameter characteristics, flame parameter characteristics and numerical physical field characteristics to construct a multi-source heterogeneous characteristic parameter set; a pre-constructed combustion state diagnosis grade evaluation quantification strategy is used to quantitatively evaluate and assign corresponding values ​​to each data in the multi-source heterogeneous characteristic parameter set to obtain a multi-source heterogeneous characteristic parameter set after quantitative evaluation and assignment, and a combustion state diagnosis comprehensive evaluation index is calculated based on the multi-source heterogeneous characteristic parameter set after quantitative evaluation and assignment, and the current operating combustion state diagnosis category of the boiler is calculated through the combustion state diagnosis comprehensive evaluation index;

[0015] S6. Use the KNN fast classification algorithm to verify the current operating combustion state diagnosis category of the boiler. When the verification is successful, the boiler operating combustion diagnosis result is obtained. If the verification fails, jump to step S1 to re-diagnose.

[0016] Preferably, the thermal parameter data and furnace combustion flame image data are acquired through an online monitoring system of the boiler; and the furnace numerical physical field data are acquired using a CFD numerical simulation method.

[0017] Preferably, the thermal parameter characteristics include nitrogen oxide generation and boiler efficiency; the boiler efficiency is calculated by collecting furnace coal parameters, steam-water parameters, and flue gas parameters from the thermal parameter data using a counter-balance method.

[0018] Preferably, the flame parameter characteristics include flame combustion stability and flame offset distance; the flame combustion stability is obtained by collecting flame images through 30 consecutive dynamic windows to extract the inter-frame brightness change rate, area change rate and center displacement, and the variance of the image frame features in the dynamic window interval is calculated, and the average value of the inter-frame feature variance is used as the evaluation criterion for flame combustion stability. The average value of the inter-frame feature variance is expressed as follows:

[0019]

[0020] Among them, σ i 2 is the feature variance of the image frame in the dynamic window interval, represents the characteristic parameter value of image frame i, μ i Indicates the average value of the current feature parameter;

[0021] The combustion flame offset distance is expressed as follows:

[0022]

[0023] Among them, F(x,y) is the feature profile vector function, P and Q are the two components of the F(x,y) vector;

[0024] Then the coordinates of the center of the burning flame are calculated using the following formula:

[0025]

[0026] Among them, x is the center x coordinate, O y is the center y coordinate;

[0027] Finally, the coordinates of the center of the combustion flame are compared with the geometric center of the boiler furnace to obtain the combustion flame offset distance.

[0028] Preferably, the specific process of step S4 is as follows:

[0029] First, the furnace numerical physical field data and the flame image features obtained in step S3 are synchronized and isomorphized to form numerical temperature fields and flame image frame data of a unified working condition step and a unified regular grid; then, the flame image features are calculated by feature extraction to obtain a flame image feature matrix; the flame image feature matrix is ​​subjected to matrix operation to correct and calculate each grid parameter of the numerical temperature field, and finally the furnace numerical physical field distribution parameters are obtained; then, the furnace numerical physical field distribution parameters are subjected to feature extraction to finally obtain numerical physical field features;

[0030] Since the numerical physical field is mainly used to reflect the temperature of each heating surface inside the boiler, the numerical physical field feature uses the proportion of the over-temperature area of ​​the heating surface in the boiler furnace as a characteristic indicator. The process of obtaining the proportion of the over-temperature area of ​​the heating surface in the boiler furnace is as follows:

[0031] Firstly, the temperature of each grid position in the furnace is obtained through the numerical physical field distribution parameters of the furnace, and the grids corresponding to the water-cooled wall, the full-screen superheater wall, and the rear-screen superheater wall are screened out through the grid number, and the near-wall temperature parameters of the corresponding grids are obtained. By comparing with the metal heating threshold, the positions of each over-temperature grid are obtained, and finally the proportion of over-temperature areas of the heating surface in the boiler furnace is obtained by counting the over-temperature grids.

[0032] Preferably, the combustion state diagnosis comprehensive evaluation index in step S5 is expressed as follows:

[0033] Q=α 1 S+α 2 C+α 3 P+α 4 η+α 5 O

[0034] Among them, Q represents the comprehensive evaluation index of combustion state diagnosis, S represents combustion stability, C represents combustion center offset, P represents nitrogen oxide generation, η represents boiler efficiency, and O represents the quantitative evaluation of local overtemperature; α 1 , α 2 , α 3 , α 4 and α 5 They represent the corresponding weights of the parameters respectively.

[0035] Preferably, the specific process of step S6 is as follows:

[0036] Since the performance indicators in the multi-source heterogeneous characteristic parameter set corresponding to the current operating combustion state diagnosis category of the boiler obtained in step S5 are greatly different, the multi-source heterogeneous characteristic parameter set is normalized, which is specifically expressed as follows:

[0037]

[0038] Among them, z is the sample feature, ΔZ is the corresponding feature value range, is the normalized feature result;

[0039] Then, the distance between each performance index in the normalized multi-source heterogeneous characteristic parameter set and each performance index in the preset combustion classification sample set is calculated. The preset combustion classification sample set is constructed by obtaining historical thermal parameter characteristics, flame parameter characteristics and numerical physical field characteristics. The distance is specifically expressed as follows:

[0040]

[0041] Among them, X i and Y i Represents two data points in N-dimensional space, N represents the number of features contained in each data point, and i represents the i-th feature;

[0042] Finally, the corresponding k nearest neighbor results are obtained by calculating the N-dimensional Euclidean space distance, and the k nearest neighbor results are based on the preset combustion classification sample set and classified by a weighted majority voting scheme, so as to obtain the corresponding KNN combustion state classification result;

[0043] When the KNN combustion state classification result is consistent with the current operating combustion state diagnosis category of the boiler, it is considered that the current operating combustion state diagnosis category of the boiler is accurate and reliable, and the current operating combustion state diagnosis category of the boiler is used as the boiler operating combustion diagnosis result.

[0044] A boiler operation combustion diagnosis system based on multi-source heterogeneous information mining is used to implement the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining, and the system includes:

[0045] The boiler operation data acquisition module is used to obtain the current thermal parameter data of the boiler and the furnace combustion flame image data;

[0046] A furnace numerical physical field module is used to calculate the current thermal parameter data of the boiler and the furnace combustion flame image data obtained by the boiler operation data acquisition module using a CFD numerical simulation method, and finally obtain furnace numerical physical field data;

[0047] A flame image preprocessing module preprocesses the furnace combustion flame image data to obtain preprocessed furnace combustion flame image data;

[0048] A flame image feature extraction module is used to extract features from the preprocessed furnace combustion flame image to obtain flame image features corresponding to the furnace combustion flame image data;

[0049] A furnace numerical physical field correction module, used to correct the furnace numerical physical field data according to the flame image characteristics to obtain furnace numerical physical field distribution parameters;

[0050] A thermal parameter characteristic calculation module, used to calculate the current thermal parameter characteristics of the boiler according to the current thermal parameter data of the boiler, wherein the thermal parameter characteristics include nitrogen oxide generation and boiler efficiency;

[0051] A flame parameter feature calculation module, used to calculate the current flame parameter features of the boiler according to the flame image features, wherein the flame parameter features include the average value of the combustion flame offset distance and the inter-frame feature variance;

[0052] A numerical physical field characteristic calculation module, used to calculate the numerical physical field characteristics according to the numerical physical field distribution parameters of the furnace, wherein the numerical physical field characteristics are the proportion of the over-temperature area of ​​the heating surface in the furnace;

[0053] A multi-source feature data fusion module is used to fuse the thermal parameter features, flame parameter features and numerical physical field features respectively obtained by the thermal parameter feature calculation module, the flame parameter feature calculation module and the numerical physical field feature calculation module to construct a multi-source heterogeneous feature parameter set;

[0054] A combustion state diagnosis level evaluation quantitative strategy building module is used to set corresponding weights for each indicator in the multi-source heterogeneous characteristic parameter set according to the three perspectives of boiler safety, environmental protection and economy combined with boiler historical operation indicators to form a combustion state diagnosis level evaluation quantitative strategy;

[0055] A combustion state diagnosis level evaluation quantification strategy module is used to perform quantitative evaluation and corresponding value assignment on each data in the multi-source heterogeneous characteristic parameter set through the combustion state diagnosis level evaluation quantification strategy, so as to obtain the multi-source heterogeneous characteristic parameter set after quantitative evaluation and value assignment;

[0056] A boiler operation combustion state diagnosis module is used to calculate a combustion state diagnosis comprehensive evaluation index according to the multi-source heterogeneous characteristic parameter set after the quantitative evaluation assignment, and calculate the current operation combustion state category of the boiler through the combustion state diagnosis comprehensive evaluation index;

[0057] The combustion state verification module is used to verify the current operation combustion state category of the boiler by using the KNN fast classification algorithm. When the verification is successful, the boiler operation combustion diagnosis result output is obtained;

[0058] The combustion diagnosis result display module is used to display the boiler operation combustion diagnosis result output by the combustion state verification module.

[0059] A computer device comprises a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining is implemented.

[0060] A storage medium stores a program, and when the program is executed by a processor, the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining is implemented.

[0061] The present invention has the following beneficial effects compared with the prior art:

[0062] (1) The boiler operation combustion diagnosis method based on multi-source heterogeneous information mining of the present invention obtains three types of multi-source heterogeneous data from different sources and with different data structure characteristics, namely thermal parameter data, furnace combustion flame image data and furnace numerical physical field data, and calculates the boiler's combustion state diagnosis comprehensive evaluation index based on the multi-source heterogeneous data combined with a pre-constructed combustion state diagnosis grade evaluation quantification strategy. Finally, based on the combustion state diagnosis comprehensive evaluation index, the current operation combustion state of the boiler can be effectively diagnosed from multiple angles of safety, environmental protection and economy; the inaccurate boiler operation combustion state diagnosis results caused by human factors are overcome, the accuracy of the boiler operation combustion state judgment results is improved, and the power operation personnel can better monitor the boiler and improve the boiler operation intelligence and diagnosis level.

[0063] (2) The boiler operation and combustion diagnosis system based on multi-source heterogeneous information mining of the present invention can accurately obtain the various characteristic parameters of the current operation of the boiler and accurately judge the boiler operation and combustion status judgment results through the effective combination of multiple modules. The various modules of the present invention can be embedded in the online monitoring system of power operation, so that power operation personnel can better monitor the boiler, thereby improving the safety of power production. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A flow chart of a boiler operation combustion diagnosis method based on multi-source heterogeneous information mining provided in Example 1 of the present invention;

[0065] Figure 2 This is a graph of the classification accuracy results of the KNN fast algorithm under different k values ​​provided in Example 1 of the present invention.

[0066] Figure 3 A structural diagram of a boiler operation combustion diagnosis system based on multi-source heterogeneous information mining provided in Example 2 of the present invention;

[0067] Figure 4 A comprehensive evaluation index result diagram output by the boiler operation combustion state diagnosis module provided in Example 2 of the present invention;

[0068] Figure 5 Embodiment 2 of the present invention provides a result graph generated by each module in a boiler operation combustion diagnosis system based on multi-source heterogeneous information mining. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0070] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0071] In the description of the present invention, it is necessary to understand that the orientations or positional relationships indicated by the terms "upper", "lower", "top", "bottom", "inside", "outside", etc. are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0072] Embodiment 1:

[0073] like Figure 1 As shown, a boiler operation combustion diagnosis method based on multi-source heterogeneous information mining includes the following steps:

[0074] S1. Obtain the current thermal parameter data of the boiler, the furnace combustion flame image data and the furnace numerical physical field data;

[0075] Specifically, in this embodiment, the thermal parameter data and the furnace combustion flame image data are obtained through the DCS online monitoring system of the boiler;

[0076] The furnace numerical physical field data is obtained by using CFD numerical simulation method, and the specific process is as follows:

[0077] First, the geometric model, mesh model and mathematical model describing the physical and chemical process of the boiler object are constructed; then, the initial conditions and boundary conditions are determined based on the actual historical operation data of the boiler, and the mesh model and the physical and chemical mathematical model are used to perform iterative calculations through the finite volume method until convergence results are obtained; finally, the physical state parameters of the converged results (such as temperature, velocity, component concentration, etc.) are compared and verified with the DCS measurement point data of the actual boiler object to ensure that the numerical model can accurately reflect the actual operation of the boiler object, and finally the numerical physical field data of the boiler furnace is obtained.

[0078] S2. Acquire corresponding thermal parameter characteristics according to the thermal parameter data;

[0079] Specifically, the thermal parameter characteristics include nitrogen oxide generation and boiler efficiency; the boiler efficiency is calculated by collecting furnace coal parameters, steam-water parameters, and flue gas parameters from the thermal parameter data using a counter-balance method.

[0080] The boiler efficiency is expressed as follows:

[0081] η 锅 =100-q 2 -q 3 -q 4 -q 5 -q 6

[0082] Among them, η 锅 represents boiler efficiency, q 2 represents the smoke exhaust loss, q 3 Indicates incomplete chemical combustion, q 4 is the mechanical incomplete combustion loss, q 5 is the heat dissipation loss, q 6 Heat loss of ash.

[0083] Specifically, q 2 ,q 3 ,q 4 ,q 5 and q 6 All are obtained through the thermal parameter data.

[0084] S3, preprocessing the furnace combustion flame image data, performing feature extraction on the preprocessed furnace combustion flame image, and obtaining flame image features corresponding to the furnace combustion flame image data; and calculating flame parameter features of the boiler according to the flame image features;

[0085] Specifically, the flame parameter characteristics include flame combustion stability and flame offset distance; the flame combustion stability is obtained by collecting flame images through 30 consecutive dynamic windows to extract the inter-frame brightness change rate, area change rate and center displacement, and the variance of the image frame features in the dynamic window interval is calculated, and the average value of the inter-frame feature variance is used as the evaluation criterion for flame combustion stability. The average value of the inter-frame feature variance is expressed as follows:

[0086]

[0087] Among them, σ i 2 is the feature variance of the image frame in the dynamic window interval, represents the characteristic parameter value of image frame i, μ i Indicates the average value of the current feature parameter;

[0088] The combustion flame offset distance is expressed as follows:

[0089]

[0090] Among them, F(x,y) is the feature profile vector function, P and Q are the two components of the F(x,y) vector;

[0091] Then the coordinates of the center of the burning flame are calculated using the following formula:

[0092]

[0093] Among them, x is the center x coordinate, O y is the center y coordinate;

[0094] Finally, the coordinates of the center of the combustion flame are compared with the geometric center of the boiler furnace to obtain the combustion flame offset distance.

[0095] S4, using the flame image feature obtained in step S3 to correct the furnace numerical physical field data, obtain the furnace numerical physical field distribution parameters, and calculate the numerical physical field features according to the furnace numerical physical field distribution parameters; the specific process is as follows:

[0096] First, the furnace numerical physical field data and the flame image features obtained in step S3 are synchronized and isomorphized to form numerical temperature fields and flame image frame data of a unified working condition step and a unified regular grid; then, the flame image features are calculated by feature extraction to obtain a flame image feature matrix; the flame image feature matrix is ​​subjected to matrix operation to correct and calculate each grid parameter of the numerical temperature field, and finally the furnace numerical physical field distribution parameters are obtained; then, the furnace numerical physical field distribution parameters are subjected to feature extraction to finally obtain numerical physical field features;

[0097] Since the numerical physical field is mainly used to reflect the temperature of each heating surface inside the boiler, the numerical physical field feature uses the proportion of the over-temperature area of ​​the heating surface in the boiler furnace as a characteristic indicator. The process of obtaining the proportion of the over-temperature area of ​​the heating surface in the boiler furnace is as follows:

[0098] Firstly, the temperature of each grid position in the furnace is obtained through the numerical physical field distribution parameters of the furnace, and the grids corresponding to the water-cooled wall, the full-screen superheater wall, and the rear-screen superheater wall are screened out through the grid number, and the near-wall temperature parameters of the corresponding grids are obtained. By comparing with the metal heating threshold, the positions of each over-temperature grid are obtained, and finally the proportion of over-temperature areas of the heating surface in the boiler furnace is obtained by counting the over-temperature grids.

[0099] S5. Data fusion is performed on the thermal parameter characteristics, flame parameter characteristics and numerical physical field characteristics to construct a multi-source heterogeneous characteristic parameter set; a pre-constructed combustion state diagnosis grade evaluation quantification strategy is used to quantitatively evaluate and assign corresponding values ​​to each data in the multi-source heterogeneous characteristic parameter set to obtain a multi-source heterogeneous characteristic parameter set after quantitative evaluation and assignment, and a combustion state diagnosis comprehensive evaluation index is calculated based on the multi-source heterogeneous characteristic parameter set after quantitative evaluation and assignment, and the current operating combustion state diagnosis category of the boiler is calculated through the combustion state diagnosis comprehensive evaluation index;

[0100] Specifically, the construction process of the combustion state diagnosis level evaluation quantification strategy is as follows:

[0101] First, the historical operation data of the boiler and the corresponding flame image data are collected, and then the historical operation data of the boiler and the corresponding flame image data are extracted and calculated using the above steps S2-S4. After obtaining the corresponding feature information, a multi-source heterogeneous feature parameter set is constructed, and combined with the corresponding operating procedures of the boiler and the safety, environmental protection and economic evaluation system of the boiler required by power production to form a statistical setting range and level of various performance indicators of the boiler, as shown in Table 1.

[0102]

[0103] Table 1 Setting range and level of status diagnosis performance indicators

[0104] Specifically, the process of performing quantitative evaluation and corresponding value assignment on each data in the multi-source heterogeneous feature parameter set to finally obtain the multi-source heterogeneous feature parameter set after quantitative evaluation and value assignment is as follows:

[0105] First, the various performance indicators in the multi-source heterogeneous feature parameter set are rated according to Table 1 above, and then the performance indicators are assigned values ​​according to the scores corresponding to the levels. The values ​​corresponding to the levels are 10 for good, 5 for medium, and 1 for poor, respectively. Finally, the multi-source heterogeneous feature parameter set after quantitative evaluation and assignment is obtained.

[0106] Specifically, the combustion state diagnosis comprehensive evaluation index is expressed as follows:

[0107] Q=α 1 S+α 2 C+α 3 P+α 4 η+α 5 O

[0108] Among them, Q represents the comprehensive evaluation index of combustion state diagnosis, S represents combustion stability, C represents combustion center offset, P represents nitrogen oxide generation, η represents boiler efficiency, and O represents the quantitative evaluation of local overtemperature; α 1, α 2 , α 3 , α 4 and α 5 They represent the corresponding weights of the parameters respectively.

[0109] In this embodiment, after many experiments, when α 1 , α 2 and α 5 The value of is 0.25, α 3 The value of is 0.15, α 4 When the value of is 0.1, the final combustion state diagnosis category is closer to the actual combustion condition, thereby improving the accuracy of the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining.

[0110] The process of determining the diagnostic category of the boiler's current operating combustion status is as follows:

[0111] The combustion state category is determined according to the combustion state diagnosis comprehensive evaluation index, wherein the combustion state categories include poor (Q∈[1,3]), medium (Q∈(3,6]), good (Q∈(6,8]), and excellent (Q∈(8,10]). Finally, the current operating combustion state diagnosis category of the boiler can be obtained according to the specific value of the combustion state diagnosis comprehensive evaluation index.

[0112] S6. Use the KNN fast classification algorithm to verify the current operating combustion state diagnosis category of the boiler. When the verification is successful, the boiler operating combustion diagnosis result is obtained. If the verification fails, jump to step S1 to re-diagnose.

[0113] The specific process of step S6 is as follows:

[0114] Since the performance indicators in the multi-source heterogeneous characteristic parameter set corresponding to the current operating combustion state diagnosis category of the boiler obtained in step S5 are greatly different, the multi-source heterogeneous characteristic parameter set is normalized, which is specifically expressed as follows:

[0115]

[0116] Among them, z is the sample feature, ΔZ is the corresponding feature value range, is the normalized feature result;

[0117] Then, the distance between each performance index in the normalized multi-source heterogeneous feature parameter set and each performance index in the preset combustion classification sample set is calculated. The distance is specifically expressed as follows:

[0118]

[0119] Among them, X i and Yi Represents two data points in N-dimensional space, N represents the number of features contained in each data point, and i represents the i-th feature;

[0120] Specifically, the preset combustion classification sample set is constructed by acquiring historical thermal parameter characteristics, flame parameter characteristics and numerical physical field characteristics, and each set of characteristic parameters of the preset combustion classification sample set corresponds to the corresponding combustion state classification result after calibration;

[0121] Finally, the corresponding k nearest neighbor results are obtained by calculating the N-dimensional Euclidean space distance, and the k nearest neighbor results are based on the preset combustion classification sample set and classified by a weighted majority voting scheme, so as to obtain the corresponding KNN combustion state classification result;

[0122] like Figure 2 As shown, in this embodiment, after multiple tests, when the K value is 12, the accuracy of diagnosing the current operating combustion state of the boiler using the KNN fast classification algorithm is the highest.

[0123] When the KNN combustion state classification result is consistent with the current operating combustion state diagnosis category of the boiler, it is considered that the current operating combustion state diagnosis category of the boiler is accurate and reliable, and the current operating combustion state diagnosis category of the boiler is used as the boiler operating combustion diagnosis result.

[0124] Example 2

[0125] like Figure 3-Figure 5 As shown, a boiler operation combustion diagnosis system based on multi-source heterogeneous information mining is used to implement the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining described in Example 1, and the system includes:

[0126] The boiler operation data acquisition module is used to obtain the current thermal parameter data of the boiler and the furnace combustion flame image data;

[0127] A furnace numerical physical field module is used to calculate the current thermal parameter data of the boiler and the furnace combustion flame image data obtained by the boiler operation data acquisition module using a CFD numerical simulation method, and finally obtain furnace numerical physical field data;

[0128] A flame image preprocessing module preprocesses the furnace combustion flame image data to obtain preprocessed furnace combustion flame image data;

[0129] A flame image feature extraction module is used to extract features from the preprocessed furnace combustion flame image to obtain flame image features corresponding to the furnace combustion flame image data;

[0130] A furnace numerical physical field correction module, used to correct the furnace numerical physical field data according to the flame image characteristics to obtain furnace numerical physical field distribution parameters;

[0131] A thermal parameter characteristic calculation module, used to calculate the current thermal parameter characteristics of the boiler according to the current thermal parameter data of the boiler, wherein the thermal parameter characteristics include nitrogen oxide generation and boiler efficiency;

[0132] A flame parameter feature calculation module, used to calculate the current flame parameter features of the boiler according to the flame image features, wherein the flame parameter features include the average value of the combustion flame offset distance and the inter-frame feature variance;

[0133] A numerical physical field characteristic calculation module, used to calculate the numerical physical field characteristics according to the numerical physical field distribution parameters of the furnace, wherein the numerical physical field characteristics are the proportion of the over-temperature area of ​​the heating surface in the furnace;

[0134] A multi-source feature data fusion module is used to fuse the thermal parameter features, flame parameter features and numerical physical field features respectively obtained by the thermal parameter feature calculation module, the flame parameter feature calculation module and the numerical physical field feature calculation module to construct a multi-source heterogeneous feature parameter set;

[0135] A combustion state diagnosis level evaluation quantitative strategy building module is used to set corresponding weights for each indicator in the multi-source heterogeneous characteristic parameter set according to the three perspectives of boiler safety, environmental protection and economy combined with boiler historical operation indicators to form a combustion state diagnosis level evaluation quantitative strategy;

[0136] A combustion state diagnosis level evaluation quantification strategy module is used to perform quantitative evaluation and corresponding value assignment on each data in the multi-source heterogeneous characteristic parameter set through the combustion state diagnosis level evaluation quantification strategy, so as to obtain the multi-source heterogeneous characteristic parameter set after quantitative evaluation and value assignment;

[0137] A boiler operation combustion state diagnosis module is used to calculate a combustion state diagnosis comprehensive evaluation index according to the multi-source heterogeneous characteristic parameter set after the quantitative evaluation assignment, and calculate the current operation combustion state category of the boiler through the combustion state diagnosis comprehensive evaluation index;

[0138] The combustion state verification module is used to verify the current operation combustion state category of the boiler by using the KNN fast classification algorithm. When the verification is successful, the boiler operation combustion diagnosis result output is obtained;

[0139] The combustion diagnosis result display module is used to display the boiler operation combustion diagnosis result output by the combustion state verification module.

[0140] Example 3

[0141] A storage medium stores a program, which, when executed by a processor, implements the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining described in Example 1.

[0142] Example 4

[0143] A computer device includes a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining described in Example 1 is implemented.

[0144] It should be noted that the computer-readable storage medium of the present embodiment may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0145] In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device, or device. In this embodiment, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable storage medium other than a computer-readable storage medium, which may send, propagate, or transmit a program used by or in combination with an instruction execution system, device, or device. The computer program contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0146] The computer readable storage medium can be written in one or more programming languages ​​or a combination thereof to execute the computer program of the present embodiment, and the programming language includes an object-oriented programming language, such as Java, Python, C++, and also includes a conventional procedural programming language, such as C language or a similar programming language. The program can be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., using an Internet service provider to connect through the Internet).

[0147] In summary, the present invention obtains three types of multi-source heterogeneous data from different sources and with different data structure characteristics, namely thermal parameter data, furnace combustion flame image data and furnace numerical physical field data, and calculates the boiler's combustion state diagnosis comprehensive evaluation index based on the multi-source heterogeneous data in combination with a pre-constructed combustion state diagnosis grade evaluation quantification strategy. Finally, according to the combustion state diagnosis comprehensive evaluation index, the current operating combustion state of the boiler can be effectively diagnosed from multiple angles of safety, environmental protection and economy; the inaccurate diagnosis results of the boiler's operating combustion state caused by human factors are overcome, the accuracy of the boiler's operating combustion state judgment results is improved, and the power operation personnel can better monitor the boiler and improve the intelligence and diagnosis level of the boiler operation.

[0148] The above specific implementation modes are preferred embodiments of the present invention and cannot be used to limit the present invention. Any other changes or other equivalent replacement methods that do not deviate from the technical solution of the present invention are included in the protection scope of the present invention.

Claims

1. A boiler operation combustion diagnosis method based on multi-source heterogeneous information mining, characterized in that: The following steps are involved: S1. Obtain the current thermal parameter data of the boiler, the furnace combustion flame image data and the furnace numerical physical field data; S2. Acquire corresponding thermal parameter characteristics according to the thermal parameter data; S3, preprocessing the furnace combustion flame image data, performing feature extraction on the preprocessed furnace combustion flame image, and obtaining flame image features corresponding to the furnace combustion flame image data; Calculating the flame parameter characteristics of the boiler according to the flame image characteristics; S4, using the flame image feature obtained in step S3 to correct the furnace numerical physical field data, to obtain the furnace numerical physical field distribution parameters, and to calculate the numerical physical field features according to the furnace numerical physical field distribution parameters; S5. Data fusion is performed on the thermal parameter characteristics, flame parameter characteristics and numerical physical field characteristics to construct a multi-source heterogeneous characteristic parameter set; a pre-constructed combustion state diagnosis grade evaluation quantification strategy is used to quantitatively evaluate and assign corresponding values ​​to each characteristic data in the multi-source heterogeneous characteristic parameter set to obtain a multi-source heterogeneous characteristic parameter set after quantitative evaluation and assignment, and then a combustion state diagnosis comprehensive evaluation index is calculated based on the multi-source heterogeneous characteristic parameter set after quantitative evaluation and assignment, and the current operating combustion state diagnosis category of the boiler is calculated through the combustion state diagnosis comprehensive evaluation index; S6. Use the KNN fast classification algorithm to verify the current operating combustion state diagnosis category of the boiler. When the verification is successful, the boiler operating combustion diagnosis result is obtained. If the verification fails, jump to step S1 to re-diagnose.

2. The boiler operation combustion diagnosis method based on multi-source heterogeneous information mining according to claim 1 is characterized in that: The thermal parameter data and furnace combustion flame image data are obtained through the boiler's online monitoring system; the furnace numerical physical field data is obtained using the CFD numerical simulation method.

3. The boiler operation combustion diagnosis method based on multi-source heterogeneous information mining according to claim 1 is characterized in that: The thermal parameter characteristics include nitrogen oxide generation and boiler efficiency; the boiler efficiency is calculated by collecting furnace coal parameters, steam-water parameters, and flue gas parameters from the thermal parameter data using a counter-balance method.

4. The boiler operation combustion diagnosis method based on multi-source heterogeneous information mining according to claim 1 is characterized in that: The flame parameter features include flame combustion stability and flame offset distance; the flame combustion stability is obtained by collecting flame images through 30 consecutive dynamic windows to extract the brightness change rate, area change rate and center displacement between frames, and calculating the variance of image frame features in the dynamic window interval, and taking the average value of the feature variance between frames as the evaluation criterion for flame combustion stability. The average value of the feature variance between frames is expressed as follows: Among them, σ i 2 is the feature variance of the image frame in the dynamic window interval, represents the characteristic parameter value of image frame i, μ i Indicates the average value of the current feature parameter; The combustion flame offset distance is expressed as follows: Among them, F(x,y) is the feature profile vector function, P and Q are the two components of the F(x,y) vector; Then the coordinates of the center of the burning flame are calculated using the following formula: Among them, x is the center x coordinate, O y is the center y coordinate; Finally, the coordinates of the center of the combustion flame are compared with the geometric center of the boiler furnace to obtain the combustion flame offset distance.

5. The boiler operation combustion diagnosis method based on multi-source heterogeneous information mining according to claim 1 is characterized in that: The specific process of step S4 is as follows: First, the furnace numerical physical field data and the flame image features obtained in step S3 are synchronized and isomorphized to form numerical temperature fields and flame image frame data of a unified working condition step and a unified regular grid; then, the flame image features are calculated by feature extraction to obtain a flame image feature matrix; the flame image feature matrix is ​​subjected to matrix operation to correct and calculate each grid parameter of the numerical temperature field, and finally the furnace numerical physical field distribution parameters are obtained; then, the furnace numerical physical field distribution parameters are subjected to feature extraction to finally obtain numerical physical field features; Since the numerical physical field is mainly used to reflect the temperature of each heating surface inside the boiler, the numerical physical field feature uses the proportion of the over-temperature area of ​​the heating surface in the boiler furnace as a characteristic indicator. The process of obtaining the proportion of the over-temperature area of ​​the heating surface in the boiler furnace is as follows: Firstly, the temperature of each grid position in the furnace is obtained through the numerical physical field distribution parameters of the furnace, and the grids corresponding to the water-cooled wall, the full-screen superheater wall, and the rear-screen superheater wall are screened out through the grid number, and the near-wall temperature parameters of the corresponding grids are obtained. By comparing with the metal heating threshold, the positions of each over-temperature grid are obtained, and finally the proportion of over-temperature areas of the heating surface in the boiler furnace is obtained by counting the over-temperature grids.

6. The boiler operation combustion diagnosis method based on multi-source heterogeneous information mining according to claim 1 is characterized in that: The combustion state diagnosis comprehensive evaluation index in step S5 is expressed as follows: Q=α1S+α2C+α3P+α4η+α5O Among them, Q represents the comprehensive evaluation index of combustion state diagnosis, S represents combustion stability, C represents combustion center bias, P represents nitrogen oxide generation, η represents boiler efficiency, and O represents the quantitative evaluation of local overtemperature conditions; α1, α2, α3, α4 and α5 represent the corresponding weights of the parameters respectively.

7. The boiler operation combustion diagnosis method based on multi-source heterogeneous information mining according to claim 1 is characterized in that: The specific process of step S6 is as follows: Since the performance indicators in the multi-source heterogeneous characteristic parameter set corresponding to the current operating combustion state diagnosis category of the boiler obtained in step S5 are greatly different, the multi-source heterogeneous characteristic parameter set is normalized, which is specifically expressed as follows: Among them, z is the sample feature, ΔZ is the corresponding feature value range, is the normalized feature result; Then, the distance between each performance index in the normalized multi-source heterogeneous characteristic parameter set and each performance index in the preset combustion classification sample set is calculated. The preset combustion classification sample set is constructed by obtaining historical thermal parameter characteristics, flame parameter characteristics and numerical physical field characteristics. The distance is specifically expressed as follows: Among them, X i and Y i Represents two data points in N-dimensional space, N represents the number of features contained in each data point, and i represents the i-th feature; Finally, the corresponding k nearest neighbor results are obtained by calculating the N-dimensional Euclidean space distance, and the k nearest neighbor results are based on the preset combustion classification sample set and classified by a weighted majority voting scheme, so as to obtain the corresponding KNN combustion state classification result; When the KNN combustion state classification result is consistent with the current operating combustion state diagnosis category of the boiler, it is considered that the current operating combustion state diagnosis category of the boiler is accurate and reliable, and the current operating combustion state diagnosis category of the boiler is used as the boiler operating combustion diagnosis result.

8. A boiler operation combustion diagnosis system based on multi-source heterogeneous information mining, used to implement the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining according to any one of claims 1 to 7, characterized in that: The system comprises: The boiler operation data acquisition module is used to obtain the current thermal parameter data of the boiler and the furnace combustion flame image data; A furnace numerical physical field module is used to calculate the current thermal parameter data of the boiler and the furnace combustion flame image data obtained by the boiler operation data acquisition module using a CFD numerical simulation method, and finally obtain furnace numerical physical field data; A flame image preprocessing module preprocesses the furnace combustion flame image data to obtain preprocessed furnace combustion flame image data; A flame image feature extraction module is used to extract features from the preprocessed furnace combustion flame image to obtain flame image features corresponding to the furnace combustion flame image data; A furnace numerical physical field correction module, used to correct the furnace numerical physical field data according to the flame image characteristics to obtain furnace numerical physical field distribution parameters; A thermal parameter characteristic calculation module, used to calculate the current thermal parameter characteristics of the boiler according to the current thermal parameter data of the boiler, wherein the thermal parameter characteristics include nitrogen oxide generation and boiler efficiency; A flame parameter feature calculation module, used to calculate the current flame parameter features of the boiler according to the flame image features, wherein the flame parameter features include the average value of the combustion flame offset distance and the inter-frame feature variance; A numerical physical field characteristic calculation module, used to calculate the numerical physical field characteristics according to the numerical physical field distribution parameters of the furnace, wherein the numerical physical field characteristics are the proportion of the over-temperature area of ​​the heating surface in the furnace; A multi-source feature data fusion module is used to fuse the thermal parameter features, flame parameter features and numerical physical field features respectively obtained by the thermal parameter feature calculation module, the flame parameter feature calculation module and the numerical physical field feature calculation module to construct a multi-source heterogeneous feature parameter set; A combustion state diagnosis level evaluation quantitative strategy building module is used to set corresponding weights for each indicator in the multi-source heterogeneous characteristic parameter set according to the three perspectives of boiler safety, environmental protection and economy combined with boiler historical operation indicators to form a combustion state diagnosis level evaluation quantitative strategy; A combustion state diagnosis level evaluation quantification strategy module is used to perform quantitative evaluation and corresponding value assignment on each data in the multi-source heterogeneous characteristic parameter set through the combustion state diagnosis level evaluation quantification strategy, so as to obtain the multi-source heterogeneous characteristic parameter set after quantitative evaluation and value assignment; A boiler operation combustion state diagnosis module is used to calculate a combustion state diagnosis comprehensive evaluation index according to the multi-source heterogeneous characteristic parameter set after the quantitative evaluation assignment, and calculate the current operation combustion state category of the boiler through the combustion state diagnosis comprehensive evaluation index; The combustion state verification module is used to verify the current operation combustion state category of the boiler by using the KNN fast classification algorithm. When the verification is successful, the boiler operation combustion diagnosis result output is obtained; The combustion diagnosis result display module is used to display the boiler operation combustion diagnosis result output by the combustion state verification module.

9. A computer device comprising a processor and a memory for storing a program executable by the processor, characterized in that: When the processor executes the program stored in the memory, it implements the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining as described in any one of claims 1 to 7.

10. A storage medium storing a program, characterized in that: When the program is executed by a processor, the boiler operation combustion diagnosis method based on multi-source heterogeneous information mining as described in any one of claims 1 to 7 is implemented.

Images