Thermal power generating unit coordination system modeling method based on C-type mean value clustering analysis
Through the C-class mean clustering analysis method, the thermal power unit coordination system was modeled, which solved the problem of complex model calculation and low accuracy, and realized more efficient calculation and more accurate model. It was suitable for the nonlinear characteristics of thermal power unit coordination system. .
Patent Information
- Application Number
- CN202410871532.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-01
- Publication Date
- 2024-11-15
AI Technical Summary
The model calculation of thermal power unit coordination system is complex and has poor accuracy. The existing methods have poor optimization results and are prone to falling into local optimization.
Using a method based on C-class mean clustering analysis, multiple sub-models are obtained by determining historical running data, constructing sample sets, initializing clustering centers and iterative clustering, and weighting is performed through membership function to obtain a coordinated system model.
It effectively reduces the calculation amount and improves the model accuracy, which can better demonstrate the globality and partial characteristics of the coordinated system, and solves the nonlinear problem.
Smart Images

Figure CN118963258A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data modeling, and in particular to a modeling method of a thermal power unit coordination system based on C-type mean clustering analysis. Background Art
[0002] The coordinated control system of thermal power units is a multivariable, nonlinear coupling system that controls the boiler, steam turbine and water supply system as a whole. Due to the characteristics of the system itself, it has high requirements on the control algorithm. Establishing a dynamic mathematical model that can accurately describe the coordinated system is the basis for adopting advanced control algorithms. Therefore, it is necessary to establish an accurate modeling method for the coordinated system.
[0003] The coordinated controlled object of the supercritical unit is a three-input and three-output multivariable thermal system. There is a complex coupling relationship between the variables. The system has large inertia and strong nonlinear characteristics. The conventional modeling scheme is to perform numerical calculations at each operating point. The accuracy is very high, but the amount of calculation is large, which is not conducive to engineering applications. Therefore, the nonlinear system is linearized at multiple operating points to obtain a local linear sub-model, and then weighted calculations are performed through weighted methods to obtain a modeling method that takes into account a certain accuracy and amount of calculation and is convenient for engineering practice.
[0004] However, the methods used to divide the model of the coordinated system of thermal power units include gap measurement, genetic algorithm, etc. Due to the relatively complex model structure of the coordinated system and the large number of identification parameters, these methods have poor optimization effects and require a lot of historical data. Secondly, the algorithm is prone to fall into local optimality and has poor accuracy. Summary of the invention
[0005] The present invention provides a modeling method for a thermal power unit coordination system based on C-type mean cluster analysis, which is used to solve the defects of the thermal power unit coordination system in the prior art that the model calculation is complex and the accuracy is poor.
[0006] On the one hand, the present invention provides a method for modeling a coordinated system of a thermal power unit based on C-type mean cluster analysis, comprising:
[0007] Determine historical operation data of the thermal power unit, and filter the historical operation data to obtain modeling sample data;
[0008] Determine the maximum number of sub-models, the number of initial cluster centers, the weights of each dimension data, and the maximum number of iterations for C-means cluster analysis;
[0009] Based on the input and output data of the thermal power unit, constructing a sample set in the modeling sample data;
[0010] Initializing the cluster center of the C-means cluster analysis, determining the search range of the C-means cluster analysis and the model structure of the thermal power unit coordination system;
[0011] Based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means cluster analysis, and the structure of the thermal power unit coordination system model, the sample set is clustered using the C-means cluster analysis, and when it iterates to the maximum number of iterations, a plurality of thermal power unit coordination sub-models are obtained;
[0012] By using the membership function, all the coordination sub-models of the thermal power units are weighted to obtain a coordination system model of the thermal power units.
[0013] According to a method for modeling a coordinated system of a thermal power unit based on C-mean cluster analysis provided by the present invention, clustering the sample set using C-mean cluster analysis includes:
[0014] Determine the Euclidean distance between each sample and the cluster center;
[0015] When the Euclidean distance is the smallest, the objective function of the C-means cluster analysis is determined as follows:
[0016]
[0017] Among them, u ki represents the membership value of the i-th sample belonging to the k-th cluster center, m represents the fuzzy index, d ki It represents the Euclidean distance between the i-th sample and the k-th cluster center, N represents the number of samples, c represents the number of cluster centers, and minJ represents the performance index.
[0018] According to a method for modeling a thermal power unit coordination system based on C-type mean cluster analysis provided by the present invention, determining the Euclidean distance between each sample and the cluster center includes:
[0019] Determine the dimensional difference weights of data of different dimensions;
[0020] Based on a preset formula and the dimension difference weights, determining the Euclidean distance between each sample and the cluster center;
[0021] The default formula is:
[0022]
[0023] Among them, d ki represents the Euclidean distance, L s represents the dimension difference weight of the s-th dimension data, x i (s) represents the historical data of the i-th sample, v k(s) represents the data of the kth cluster center, and l represents the number of samples.
[0024] According to a method for modeling a coordinated system of a thermal power unit based on C-means cluster analysis provided by the present invention, the method of determining the maximum number of sub-models, the number of initial cluster centers, and the maximum number of iterations of the C-means cluster analysis includes:
[0025] Determine the model accuracy and calculation speed based on preset requirements;
[0026] Based on the model accuracy and the calculation speed, the maximum number of sub-models is determined to be a first preset value, the number of initial cluster centers is determined to be a second preset value, and the maximum number of iterations is determined to be a first preset range.
[0027] According to a method for modeling a thermal power unit coordination system based on C-means cluster analysis provided by the present invention, the initialization of the cluster center of the C-means cluster analysis includes:
[0028] Determine a first operating point, a second operating point, and a third operating point;
[0029] The load, main steam pressure and separator temperature of the first operating point, the second operating point and the third operating point are determined respectively to obtain the cluster center of the C-means cluster analysis.
[0030] According to a method for modeling a thermal power unit coordination system based on C-type mean cluster analysis provided by the present invention, the weighted value when weighted processing is performed on all the thermal power unit coordination sub-models is obtained by the following formula:
[0031]
[0032] Among them, ψ i represents the weight of the objective function assigned to the i-th sub-model, d i Indicates the weighted distance between the current operating condition and the i-th sub-model operating point in the sub-model set, M s represents the total number of sub-models, and k represents the cluster center.
[0033] According to a method for modeling a thermal power unit coordination system based on C-type mean cluster analysis provided by the present invention, the determining of historical operation data of the thermal power unit includes:
[0034] Determine the load of the front unit of the thermal power unit, the load setting value, the total coal quantity, the main steam pressure setting value, the main steam pressure actual value, the total feed water quantity and the intermediate point temperature setting value and actual value as the historical operation data;
[0035] The historical operation data is selected within a range of 30% to 100% of the rated load, with a time interval of less than 2s and including steady-state conditions of no less than 3 load points.
[0036] According to a method for modeling a thermal power unit coordination system based on C-mean cluster analysis provided by the present invention, the input in the input and output data includes: a comprehensive valve position instruction of a steam turbine, a feedwater flow rate and a fuel amount;
[0037] The outputs in the input and output data include: actual power generated by the unit, steam temperature at the separator outlet and main steam pressure.
[0038] The present invention also provides a thermal power unit coordination system modeling device based on C-type mean cluster analysis, comprising:
[0039] A determination module is used to determine the historical operation data of the thermal power unit and screen the historical operation data to obtain modeling sample data; determine the maximum number of sub-models, the number of initial cluster centers, the weight of each dimension data and the maximum number of iterations of the C-means cluster analysis;
[0040] A construction module, used to construct a sample set in the modeling sample data based on the input and output data of the thermal power unit;
[0041] An initialization module, used to initialize the cluster center of the C-means cluster analysis, determine the search range of the C-means cluster analysis and the model structure of the thermal power unit coordination system;
[0042] A clustering module, configured to cluster the sample set using C-means clustering analysis based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means clustering analysis, and the structure of the thermal power unit coordination system model, and obtain a plurality of thermal power unit coordination sub-models when iterating to the maximum number of iterations;
[0043] The weighting module is used to obtain a thermal power unit coordination system model after weighted processing of all the thermal power unit coordination sub-models using a membership function.
[0044] On the other hand, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements a method for modeling a coordinated system of thermal power units based on C-type mean clustering analysis as described in any one of the above-mentioned methods.
[0045] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for modeling a coordinated system of thermal power units based on C-type mean clustering analysis as described in any one of the above.
[0046] On the other hand, the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned methods for modeling a coordinated system of thermal power units based on C-type mean clustering analysis.
[0047] The method for modeling a coordinated system of a thermal power plant unit based on C-mean cluster analysis provided by the present invention comprises the following steps: determining historical operation data of the thermal power plant unit, and screening the historical operation data to obtain modeling sample data; determining the maximum number of sub-models, the initial number of cluster centers, the weight of each dimensional data, and the maximum number of iterations of the C-mean cluster analysis; constructing a sample set in the modeling sample data based on the input and output data of the thermal power plant unit; initializing the cluster centers of the C-mean cluster analysis, determining the search range of the C-mean cluster analysis and the model structure of the coordinated system of the thermal power plant unit; and determining the maximum number of sub-models, the initial number of cluster centers, the weight of each dimensional data, and the maximum number of iterations of the C-mean cluster analysis. The value, the initialized cluster center, the search range of the C-means cluster analysis and the structure of the coordinated system model of the thermal power units are selected. The sample set is clustered using the C-means cluster analysis. When it iterates to the maximum number of iterations, multiple coordinated sub-models of the thermal power units are obtained. After weighted processing of all the coordinated sub-models of the thermal power units using the membership function, the coordinated system model of the thermal power units is obtained. The sub-models divided by the C-means cluster analysis can better show the globality and partial characteristics of the coordinated system model, and the nonlinear problem can be effectively solved by controlling the weighted multiple sub-models. While reducing the amount of calculation, the model accuracy is also effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0049] Figure 1 It is a flow chart of a method for modeling a thermal power unit coordination system based on C-type mean cluster analysis provided by an embodiment of the present invention;
[0050] Figure 2 It is a schematic diagram of changes in various controlled quantities corresponding to changes in coal quantity under 30% load provided by an embodiment of the present invention;
[0051] Figure 3 It is a schematic diagram of changes in water supply corresponding to each controlled quantity under 30% load provided by an embodiment of the present invention;
[0052] Figure 4 It is a schematic diagram of the change of each controlled variable corresponding to the change of the comprehensive valve position instruction under 30% load provided by an embodiment of the present invention;
[0053] Figure 5 It is a structural schematic diagram of a thermal power unit coordination system modeling device based on C-type mean cluster analysis provided by an embodiment of the present invention;
[0054] Figure 6 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0055] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0056] Combine the following Figure 1-Figure 6 A schematic flow chart describing the method for modeling a thermal power unit coordination system based on C-type mean cluster analysis of the present invention.
[0057] Figure 1 It is a flow chart of a method for modeling a coordinated system of thermal power units based on C-type mean cluster analysis provided in an embodiment of the present invention.
[0058] like Figure 1 As shown, the thermal power unit coordination system modeling method based on C-type mean cluster analysis provided by the embodiment of the present invention can be executed by a server, and the method mainly includes the following steps:
[0059] 101. Determine the historical operating data of the thermal power units, and filter the historical operating data to obtain modeling sample data.
[0060] In a specific implementation process, the historical operation data of the thermal power unit is first determined. The historical operation data mainly refers to the relevant operation parameters, including the load of the front unit of the thermal power unit, the load setting value, the total coal volume, the main steam pressure setting value, the main steam pressure actual value, the total feed water volume and the intermediate point temperature setting value and actual value, as the historical operation data. Among them, the selection range of the historical operation data is 30% to 100% of the rated load range, the time interval is within 2s, and it should include the steady-state conditions of no less than 3 load points within the rated load range of 30% to 100%.
[0061] Determining the historical operation data of thermal power units mainly depends on the information systems of power plants, such as plant-level monitoring information systems and distributed control systems. These systems collect and store various data in real time during the operation of the units, including analog data, switch data, and dynamic data. Specifically, analog data mainly comes from the production process, such as continuously changing physical quantities such as temperature, pressure, and flow. These data are crucial for equipment fault diagnosis, unit performance calculation, operation optimization, and information management. Switch data mainly reflects the switch status of equipment or systems, such as the opening or closing of valves, the start or stop of motors, etc. Dynamic data describes the dynamic changes of the unit during operation, such as load changes, temperature fluctuations, etc. That is, analog data, switch data, and dynamic data include the load of the front unit of the thermal power unit, load setting value, total coal volume, main steam pressure setting value, main steam pressure actual value, total feed water volume, and intermediate point temperature setting value and actual value.
[0062] Then, based on the modeling requirements, qualified data is screened from the historical operating data and obviously abnormal data is removed, so that the modeling sample data can be closer to the actual working conditions.
[0063] 102. Determine the maximum number of sub-models, the number of initial cluster centers, the weights of each dimension data, and the maximum number of iterations for C-means cluster analysis.
[0064] First, based on the preset requirements, determine the model accuracy and calculation speed. The preset requirements refer to the requirements for the model, the accuracy data of the model output, etc. Then, based on the model accuracy and calculation speed, determine the maximum number of sub-models Cmax as the first preset value, such as 6, the number of initial cluster centers C0 as the second preset value, such as 3, and the maximum number of iterations as the first preset range, such as t max ∈[200,400].
[0065] 103. Based on the input and output data of the thermal power unit, a sample set is constructed in the modeling sample data.
[0066] The sample set is constructed by constructing the thermal power unit coordination system as a three-input and three-output system. The input is the turbine comprehensive valve position command, feedwater flow and fuel quantity, and the output is the actual power of the unit, the separator outlet steam temperature (intermediate point temperature) and the main steam pressure. Therefore, after coupling, there are a total of 9 transfer functions composed of a 3*3 system, as shown in formula (1):
[0067]
[0068] Total k 11 ,k 12 …k 32 ,k 33 、T 11 ,T 12 …T32 ,T 33 , τ 11 ,τ 12 …τ 32 ,τ 33 、T 111 ,T 112 …T 331 ,T 332 45 parameters need to be identified. Among them, k 11 ,k 12 …k 32 ,k 33 represents the gain of the transfer function, T 11 ,T 12 …T 32 ,T 33 represents the time constant of the transfer function, τ 11 ,τ 12 …τ 32 ,τ 33 represents the delay time of the transfer function, T 111 ,T 112 …T 331 ,T 332 represents the inertia time of the transfer function.
[0069] 104. Initialize the cluster center of the C-means cluster analysis, determine the search range of the C-means cluster analysis and the model structure of the thermal power unit coordination system.
[0070] Initializing the cluster center may be to determine the first operating point #1, the second operating point #2, and the third operating point #3; respectively determine the load, main steam pressure, and separator temperature of the first operating point, the second operating point, and the third operating point, and obtain the cluster center of the C-means cluster analysis, as shown in Table 1:
[0071]
[0072] Table 1
[0073] 105. Based on the maximum number of sub-models, the number of initial cluster centers, the weights of data in each dimension, the initialized cluster centers, the search range of C-means cluster analysis and the model structure of the thermal power unit coordination system, the sample set is clustered using C-means cluster analysis. When it iterates to the maximum number of iterations, multiple thermal power unit coordination sub-models are obtained.
[0074] After obtaining the relevant parameters, the sample set can be clustered using C-means clustering analysis. When the maximum number of iterations is reached, the current cluster centers v1, ...vn, i.e., the coordination sub-models of each thermal power unit, are obtained.
[0075] 106. After weighting all the coordination sub-models of thermal power units using the membership function, the coordination system model of thermal power units is obtained.
[0076] Then, the membership function is used to add the different sub-model weights in a weighted manner to obtain the final thermal power unit coordination system model.
[0077] Since C-means cluster analysis uses Euclidean distance to divide sub-models, it can better demonstrate the globality and partial characteristics of the coordinated system model. Control through multi-model weighting can effectively solve nonlinear problems, which not only effectively reduces the algorithm calculation workload, but also effectively improves the accuracy of model detection.
[0078] Further, based on the above embodiment, the sample set is clustered using C-means cluster analysis in this embodiment, including: determining the Euclidean distance between each sample and the cluster center; when the Euclidean distance is the smallest, determining the objective function of the C-means cluster analysis, such as formula (2):
[0079]
[0080] Among them, u ki Indicates the membership value of the i-th sample belonging to the k-th cluster center, m represents the fuzzy index, which is 2, and d ki It represents the Euclidean distance between the i-th sample and the k-th cluster center, N represents the number of samples, c represents the number of cluster centers, minJ represents the performance index, and st is a mathematical symbol used to indicate that certain variables or expressions need to meet specific conditions or constraints.
[0081] And u ki The calculation method of is as follows:
[0082]
[0083] Among them, d ji Represents the Euclidean distance between the i-th sample and the j-th cluster center.
[0084] Determining the Euclidean distance between each sample and the cluster center includes: determining the dimensional difference weight L of data of different dimensions s ; Based on the preset formula and dimension difference weight, determine the Euclidean distance between each sample and the cluster center; the preset formula is as follows:
[0085]
[0086] Among them, d ki represents the Euclidean distance, L s represents the dimension difference weight of the s-th dimension data, x i(s) represents the historical data of the i-th sample, v k (s) represents the data of the kth cluster center, and l represents the number of samples.
[0087] Further, based on the above embodiment, the weighted value of all the thermal power unit coordination sub-models in this embodiment when weighted processing is performed is obtained by the following formula (5):
[0088]
[0089] Among them, ψ i represents the weight of the objective function assigned to the i-th sub-model, d i Indicates the weighted distance between the current operating condition and the i-th sub-model operating point in the sub-model set, M s represents the total number of sub-models, and k represents the cluster center.
[0090] Then the final model can be obtained through formula (6).
[0091] V=Ψ1v1+Ψ2v2+Ψ3v3 (6)
[0092] Among them, V represents the coordinated system model of thermal power units, v1v2 and v3 represent the coordinated sub-models of thermal power units, and Ψ1, Ψ2 and Ψ3 represent the weights of the corresponding sub-models.
[0093] In order to verify the effect of the method of the present invention, a 660MW power plant is taken as an example, and the model at 30% load is shown in Table 2 below:
[0094]
[0095]
[0096] Table 2
[0097] Figure 2 It is a schematic diagram of changes in various controlled quantities corresponding to changes in coal quantity under 30% load provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of changes in water supply corresponding to each controlled quantity under 30% load provided by an embodiment of the present invention; Figure 4 : is a schematic diagram of the change of each controlled quantity corresponding to the change of the comprehensive valve position command under 30% load provided by the embodiment of the present invention; wherein, the coal feed rate, water feed rate and steam turbine comprehensive valve position command step change, the main steam pressure, load and separator temperature change trend are as follows Figure 2 , Figure 3 and Figure 4 ,From the figure, it can be seen that the modeling of the ,coordinated system using C-means analysis clustering method is highly consistent with the ,actual system, thus while reducing the amount of calculation, it also ,effectively ensures the accuracy of the model calculation.
[0098] Based on the same general inventive concept, the present invention also protects a thermal power unit coordination system modeling device based on C-type mean clustering analysis. The thermal power unit coordination system modeling device based on C-type mean clustering analysis provided by the present invention is described below. The thermal power unit coordination system modeling device based on C-type mean clustering analysis described below and the thermal power unit coordination system modeling method based on C-type mean clustering analysis described above can be referenced to each other.
[0099] Figure 5 It is a structural schematic diagram of a thermal power unit coordination system modeling device based on C-type mean cluster analysis provided by an embodiment of the present invention.
[0100] like Figure 5 As shown, the thermal power unit coordination system modeling device based on C-type mean cluster analysis provided in this embodiment includes:
[0101] Determination module 501 is used to determine the historical operation data of the thermal power unit, and screen the historical operation data to obtain modeling sample data; determine the maximum number of sub-models, the number of initial cluster centers, the weight of each dimension data and the maximum number of iterations of C-means cluster analysis;
[0102] A construction module 502 is used to construct a sample set in the modeling sample data based on the input and output data of the thermal power unit;
[0103] Initialization module 503, used to initialize the cluster center of C-means cluster analysis, determine the search range of C-means cluster analysis and the model structure of the thermal power unit coordination system;
[0104] The clustering module 504 is used to cluster the sample set using C-means clustering analysis based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means clustering analysis, and the structure of the thermal power unit coordination system model, and obtain multiple thermal power unit coordination sub-models when iterating to the maximum number of iterations;
[0105] The weighting module 505 is used to obtain a thermal power unit coordination system model by weighting all the thermal power unit coordination sub-models using a membership function.
[0106] Furthermore, the clustering module 504 in this embodiment is specifically used for:
[0107] Determine the Euclidean distance between each sample and the cluster center;
[0108] When the Euclidean distance is the smallest, the objective function of the C-means cluster analysis is determined as follows:
[0109]
[0110] Among them, u ki represents the membership value of the i-th sample belonging to the k-th cluster center, m represents the fuzzy index, d ki It represents the Euclidean distance between the i-th sample and the k-th cluster center, N represents the number of samples, c represents the number of cluster centers, and minJ represents the performance index.
[0111] Furthermore, the clustering module 504 in this embodiment is also specifically used for:
[0112] Determine the dimensional difference weights of data of different dimensions;
[0113] Based on a preset formula and the dimension difference weights, determining the Euclidean distance between each sample and the cluster center;
[0114] The default formula is:
[0115]
[0116] Among them, d ki represents the Euclidean distance, L s represents the dimension difference weight of the s-th dimension data, x i (s) represents the historical data of the i-th sample, v k (s) represents the data of the kth cluster center, and l represents the number of samples.
[0117] Furthermore, the determination module 501 in this embodiment is specifically used for:
[0118] Determine the model accuracy and calculation speed based on preset requirements;
[0119] Based on the model accuracy and the calculation speed, the maximum number of sub-models is determined to be a first preset value, the number of initial cluster centers is determined to be a second preset value, and the maximum number of iterations is determined to be a first preset range.
[0120] Furthermore, the initialization module 503 in this embodiment is specifically used for:
[0121] Determine a first operating point, a second operating point, and a third operating point;
[0122] The load, main steam pressure and separator temperature of the first operating point, the second operating point and the third operating point are determined respectively to obtain the cluster center of the C-means cluster analysis.
[0123] Furthermore, the weighting module 505 in this embodiment is specifically used for:
[0124]
[0125] Among them, ψ irepresents the weight of the objective function assigned to the i-th sub-model, d i Indicates the weighted distance between the current operating condition and the i-th sub-model operating point in the sub-model set, M s represents the total number of sub-models, and k represents the cluster center.
[0126] Furthermore, the determination module 501 in this embodiment is specifically used for:
[0127] Determine the load of the front unit of the thermal power unit, the load setting value, the total coal quantity, the main steam pressure setting value, the main steam pressure actual value, the total feed water quantity and the intermediate point temperature setting value and actual value as the historical operation data;
[0128] The historical operation data is selected within a range of 30% to 100% of the rated load, with a time interval of less than 2s and including steady-state conditions of no less than 3 load points.
[0129] Furthermore, the input of the input and output data in this embodiment includes: turbine integrated valve position instruction, feed water flow and fuel quantity;
[0130] The outputs in the input and output data include: actual power generated by the unit, steam temperature at the separator outlet and main steam pressure.
[0131] Figure 6 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.
[0132] like Figure 6As shown, the electronic device may include: a processor (processor) 610, a communication interface (Communications Interface) 620, a memory (memory) 630 and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call the logic instructions in the memory 630 to execute a thermal power unit coordination system modeling method based on C-mean clustering analysis, the method comprising: determining the historical operation data of the thermal power unit, and screening the modeling sample data from the historical operation data; determining the maximum number of sub-models, the number of initial cluster centers, the weight of each dimension data and the maximum number of iterations of the C-means clustering analysis; constructing a sample set in the modeling sample data based on the input and output data of the thermal power unit; initializing the cluster centers of the C-means clustering analysis, and determining the C-means clustering analysis. The search range of the C-means clustering analysis and the structure of the coordinated system model of the thermal power units are determined; based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means clustering analysis and the structure of the coordinated system model of the thermal power units, the sample set is clustered using the C-means clustering analysis, and when it iterates to the maximum number of iterations, a plurality of coordinated sub-models of the thermal power units are obtained; and after weighted processing of all the coordinated sub-models of the thermal power units is performed using a membership function, a coordinated system model of the thermal power units is obtained.
[0133] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0134] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the thermal power unit coordination system modeling method based on C-means clustering analysis provided by the above methods, the method comprising: determining the historical operation data of the thermal power unit, and screening the modeling sample data from the historical operation data; determining the maximum number of sub-models, the number of initial cluster centers, the weight of each dimensional data and the maximum number of iterations of the C-means clustering analysis; based on the input and output data of the thermal power unit, in the modeling sample data A sample set is constructed from the data; the cluster center of the C-means cluster analysis is initialized, and the search range of the C-means cluster analysis and the coordination system model structure of the thermal power units are determined; based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means cluster analysis and the coordination system model structure of the thermal power units, the sample set is clustered using the C-means cluster analysis, and when it is iterated to the maximum number of iterations, a plurality of coordination sub-models of the thermal power units are obtained; and after weighted processing of all the coordination sub-models of the thermal power units is performed using a membership function, a coordination system model of the thermal power units is obtained.
[0135] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the thermal power unit coordination system modeling method based on C-means cluster analysis provided by the above methods, the method comprising: determining the historical operation data of the thermal power unit, and screening the modeling sample data from the historical operation data; determining the maximum number of sub-models, the initial number of cluster centers, the weight of each dimensional data, and the maximum number of iterations of the C-means cluster analysis; constructing a sample set from the modeling sample data based on the input and output data of the thermal power unit; initializing the The cluster center of the C-means cluster analysis is used to determine the search range of the C-means cluster analysis and the structure of the coordinated system model of the thermal power units; based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means cluster analysis and the structure of the coordinated system model of the thermal power units, the sample set is clustered using the C-means cluster analysis, and when it iterates to the maximum number of iterations, a plurality of coordinated sub-models of the thermal power units are obtained; and after weighted processing of all the coordinated sub-models of the thermal power units is performed using a membership function, a coordinated system model of the thermal power units is obtained.
[0136] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0137] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for modeling a coordinated system of thermal power units based on C-mean cluster analysis, characterized in that: include: Determine historical operation data of the thermal power unit, and filter the historical operation data to obtain modeling sample data; Determine the maximum number of sub-models, the number of initial cluster centers, the weights of each dimension data, and the maximum number of iterations for C-means cluster analysis; Based on the input and output data of the thermal power unit, constructing a sample set in the modeling sample data; Initializing the cluster center of the C-means cluster analysis, determining the search range of the C-means cluster analysis and the model structure of the thermal power unit coordination system; Based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means cluster analysis, and the structure of the thermal power unit coordination system model, the sample set is clustered using the C-means cluster analysis, and when it iterates to the maximum number of iterations, a plurality of thermal power unit coordination sub-models are obtained; By using the membership function, all the coordination sub-models of the thermal power units are weighted to obtain a coordination system model of the thermal power units.
2. The method for modeling a thermal power unit coordination system based on C-mean cluster analysis according to claim 1 is characterized in that: The clustering of the sample set by using C-means cluster analysis includes: Determine the Euclidean distance between each sample and the cluster center; When the Euclidean distance is the smallest, the objective function of the C-means cluster analysis is determined as follows: Among them, u ki represents the membership value of the i-th sample belonging to the k-th cluster center, m represents the fuzzy index, d ki It represents the Euclidean distance between the i-th sample and the k-th cluster center, N represents the number of samples, c represents the number of cluster centers, and minJ represents the performance index.
3. The method for modeling a thermal power unit coordination system based on C-mean cluster analysis according to claim 2 is characterized in that: Determining the Euclidean distance between each sample and the cluster center includes: Determine the dimensional difference weights of data of different dimensions; Based on a preset formula and the dimension difference weights, determining the Euclidean distance between each sample and the cluster center; The default formula is: Among them, d ki represents the Euclidean distance, L s represents the dimension difference weight of the s-th dimension data, x i (s) represents the historical data of the i-th sample, v k (s) represents the data of the kth cluster center, and l represents the number of samples.
4. The method for modeling a thermal power unit coordination system based on C-mean cluster analysis according to claim 1, characterized in that: The determining of the maximum number of sub-models, the number of initial cluster centers, and the maximum number of iterations of the C-means cluster analysis includes: Determine the model accuracy and calculation speed based on preset requirements; Based on the model accuracy and the calculation speed, the maximum number of sub-models is determined to be a first preset value, the number of initial cluster centers is determined to be a second preset value, and the maximum number of iterations is determined to be a first preset range.
5. The method for modeling a thermal power unit coordination system based on C-mean cluster analysis according to claim 4 is characterized in that: Initializing the cluster center of the C-means cluster analysis includes: Determine a first operating point, a second operating point, and a third operating point; The load, main steam pressure and separator temperature of the first operating point, the second operating point and the third operating point are determined respectively to obtain the cluster center of the C-means cluster analysis.
6. The method for modeling a thermal power unit coordination system based on C-mean cluster analysis according to claim 5 is characterized in that: The weighted value when weighted processing is performed on all the thermal power unit coordination sub-models is obtained by the following formula: Among them, ψ i represents the weight of the objective function assigned to the i-th sub-model, d i Indicates the weighted distance between the current operating condition and the i-th sub-model operating point in the sub-model set, M s represents the total number of sub-models, and k represents the cluster center.
7. The method for modeling a coordinated system of thermal power units based on C-mean cluster analysis according to any one of claims 1 to 6, characterized in that: The determining of the historical operation data of the thermal power unit includes: Determine the load of the front unit of the thermal power unit, the load setting value, the total coal quantity, the main steam pressure setting value, the main steam pressure actual value, the total feed water quantity and the intermediate point temperature setting value and actual value as the historical operation data; The historical operation data is selected within a range of 30% to 100% of the rated load, with a time interval of less than 2s and including steady-state conditions of no less than 3 load points.
8. The method for modeling a coordinated system of thermal power units based on C-mean cluster analysis according to any one of claims 1 to 6, characterized in that: The input of the input and output data includes: comprehensive valve position instruction of steam turbine, feed water flow and fuel quantity; The outputs in the input and output data include: actual power generated by the unit, steam temperature at the separator outlet and main steam pressure.
9. A thermal power unit coordination system modeling device based on C-type mean cluster analysis, characterized in that: include: A determination module is used to determine the historical operation data of the thermal power unit and to filter out the modeling sample data from the historical operation data; Determine the maximum number of sub-models, the number of initial cluster centers, the weights of each dimension data, and the maximum number of iterations for C-means cluster analysis; A construction module, used to construct a sample set in the modeling sample data based on the input and output data of the thermal power unit; An initialization module, used to initialize the cluster center of the C-means cluster analysis, determine the search range of the C-means cluster analysis and the model structure of the thermal power unit coordination system; A clustering module, configured to cluster the sample set using C-means clustering analysis based on the maximum number of sub-models, the number of initial cluster centers, the weights of the data in each dimension, the initialized cluster centers, the search range of the C-means clustering analysis, and the structure of the thermal power unit coordination system model, and obtain a plurality of thermal power unit coordination sub-models when iterating to the maximum number of iterations; The weighting module is used to obtain a thermal power unit coordination system model after weighted processing of all the thermal power unit coordination sub-models using a membership function.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for modeling a coordinated system of thermal power units based on C-type mean clustering analysis as described in any one of claims 1 to 8 is implemented.
