Coal consumption prediction system for cogeneration units in load-carrying power centers based on big data

Abstract

The present invention discloses a coal consumption prediction system for a cogeneration unit with a load power center based on big data, and the prediction system comprises the following steps: S1: carrying out unit performance tests, collecting original test data of coal consumption for power generation and coal consumption for power supply under different load rates and when the unit is running with a load power center and when the load power center is carrying different loads; S2: collecting original data of coal consumption for power generation under different operation modes, different heat loads in heating season and non-heating season; S3: collecting test data for preprocessing operations as a data set for model building; S4: establishing a data prediction model, estimating the coal consumption for power generation and coal consumption for power supply of each unit under different variable conditions; S5: statistically analyzing the prediction results and obtaining an analysis report; S6: optimizing the operation mode of the unit according to external environmental conditions. The present invention can scientifically predict coal consumption under different complex working conditions, finely manage, formulate optimized operation plans, and improve the economic operation level of the unit.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy conservation and consumption reduction, and specifically relates to a coal consumption prediction system for a load power center cogeneration unit based on big data. Background Art

[0002] With the continuous development of the economy and society, people's environmental awareness is constantly improving. In the context of low-carbon economy, energy conservation and emission reduction in the power industry is imminent. In addition, with the current high coal prices and the difficult operation of thermal power units, energy conservation and consumption reduction work is even more urgent. Due to the complex and changeable operating environment and operating conditions of thermal power units, it is no longer possible to meet the requirements of refined management by only relying on past operation inspections to guide economic adjustments.

[0003] The present invention provides a coal consumption prediction system for a load power center cogeneration unit based on big data to solve the deficiencies in the prior art, which is very necessary. Summary of the invention

[0004] The purpose of the present invention is to solve the problem of economical operation of cogeneration units in load power centers under different operating conditions in view of the complex and changeable operating environment and operating conditions of thermal power units, and to provide a coal consumption prediction system for cogeneration units in load power centers based on big data.

[0005] Invention concept: In view of the fact that our company has two units, each equipped with a load power center, and the particularity of the heating market that all uses steam heating, it is more complicated than conventional cogeneration units, and there are more variables affecting the energy consumption changes of the units, and the analysis process is more complicated. In order to clarify the actual energy consumption of units under different operating conditions, better scientific and timely prediction, pre-control, and refined management, provide core support for the company's operations, realize the integration of existing test data information, establish a power generation coal consumption prediction model system, and improve the economic operation level of the units.

[0006] To this end, the technical solution of the present invention is a coal consumption prediction system for a load power center cogeneration unit based on big data, which is characterized by comprising the following steps:

[0007] S1: Carry out unit performance test, collect original test data of power generation coal consumption and power supply coal consumption at different load rates, when the unit is running with load power center, and when the load power center is carrying different loads;

[0008] S2: Collect the original data of coal consumption for power generation under different operation modes and different heat loads in heating season and non-heating season, and store the acquired data in the database to provide a data basis for data preprocessing;

[0009] S3: Collect raw data for preprocessing, fit, expand and refresh the test data, and establish a database for the model;

[0010] S4: Establish a data prediction model to estimate the power generation coal consumption and power supply coal consumption of each unit under different variables;

[0011] S5: Statistical analysis of prediction results and drawing up an analysis report;

[0012] S6: Optimize the unit operation mode by combining external environmental conditions and internal factors that affect the unit operation.

[0013] Furthermore, the different operating modes of the unit in step S2 include: a single-unit operation mode in the non-heating season, a dual-unit operation mode in the non-heating season, and a dual-unit operation mode in the heating season. The coal consumption for power generation corresponding to different electrical loads and thermal loads under these three operating modes is collected, and the data obtained is stored in a database to provide a data basis for data preprocessing.

[0014] Furthermore, establishing the prediction model in step S4 includes the following steps:

[0015] S41: Select the unit test data set, use the intelligent data modeling software DTEMPOWER, and establish the prediction model of power generation coal consumption and power supply coal consumption for units 1 and 2 under the load power center operation mode;

[0016] S42: Select the training set of power generation coal consumption data corresponding to different unit loads and heat loads in the non-heating season dual-unit operation mode, and use the intelligent data modeling software DTEMPOWER to establish the non-heating season unit heating power generation coal consumption prediction model for Unit 1 and Unit 2 respectively;

[0017] S43: Select the training set of power generation coal consumption data corresponding to different unit loads and heat loads under the dual-machine operation mode in the heating season, and use the intelligent data modeling software DTEMPOWER to establish a prediction model for the coal consumption of heating power generation with units in the dual-machine operation mode in the heating season;

[0018] S44: Output the training results of S41, S42, and S43 models, keep the training results, select the training set as the data basis, and output the coal consumption data sets of power generation under three variables, namely, power generation, heating, and power generation of the load power center, for Unit 1 and Unit 2 in the non-heating season, and under the dual-machine operation in the non-heating season and the dual-machine operation in the heating season. Use the intelligent data modeling software DTEMPOWER to establish the coal consumption prediction model for power generation with load power center and heating in the non-heating season for Unit 1, the coal consumption prediction model for power generation with load power center and heating in the non-heating season for Unit 2, the coal consumption prediction model for power generation with load power center and heating in the non-heating season for dual-machine operation, and the coal consumption prediction model for power generation with dual-machine operation in the heating season;

[0019] S45: Output S44 different working condition model training sets, verify and summarize with historical data, regularly select training sets, and optimize the prediction model.

[0020] Furthermore, in steps S41, S42, S43, and S44, a coal consumption prediction model for power center with heating and power generation with load for Unit 1 in the non-heating season, a coal consumption prediction model for power center with heating and power generation with load for Unit 2 in the non-heating season, a coal consumption prediction model for power center with heating and power generation with load for dual-unit operation in the non-heating season, and a coal consumption prediction model for power generation with load for dual-unit operation in the heating season are established respectively under three variables: unit load rate, power center load with different loads, and heating load. At this time, all the prediction models are completed.

[0021] Furthermore, in step S45, the training results of the model are output, and the training results obtained by inputting the training set are verified and saved. The saved training results are directly used for predicting the coal consumption of the unit for power generation under actual operating conditions and guiding the preparation of optimized operation plans.

[0022] Furthermore, in step S5, the prediction results are statistically analyzed to obtain an analysis report, and guidance and suggestions are made on the unit operation mode and load distribution.

[0023] Furthermore, step S6 combines the external environmental conditions and internal factors that affect the operation of the unit, the external environmental conditions include the renewable energy power generation in the region, seasonal hydropower conditions, and changes in industrial electricity consumption, and the internal factors include unit maintenance schedules, equipment defects, fuel procurement, etc.; combining the external environmental conditions and internal factors, determine the basic power generation load and heating capacity of the unit, use the established model to predict the coal consumption results of power generation under different operating conditions, and according to the prediction result report, formulate optimization of the unit operation mode, thermal power load distribution, and load power center operation mode and load distribution, etc., to obtain the optimal economic operation plan.

[0024] Beneficial effects: The coal consumption prediction system for the cogeneration unit of the load power center based on big data of the present invention improves the power generation prediction capability of the unit, provides core support for the company's indicator prediction and business prediction, and accurately predicts, intervenes in advance, and controls in advance. According to the data prediction results and external factors, the most economical operation plan is formulated. In addition, the data of the present invention is derived from performance tests and verified with historical actual data. The data is reliable, and through model construction, it is simple to use and effectively solves the coal consumption prediction problem of various complex working conditions of the cogeneration unit of the load power center. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of the present invention.

[0026] Figure 2It is a relationship diagram between the three variables of the average load of the unit, the average heating supply, and the average load of the load power center and the coal consumption for power generation under different operating conditions of the present invention, wherein:

[0027] Figure 2 a(1) is a 3D diagram of the relationship between unit load, heating and power generation coal consumption output by the prediction model of the coal consumption of the power center with heating and power generation of unit 1 in the non-heating season of the present invention;

[0028] Figure 2 a(2) is a 3D diagram showing the relationship between the unit load, the load of the power center and the power generation coal consumption output by the prediction model of the coal consumption of the power center with heating power generation in the non-heating season of the present invention;

[0029] Figure 2 a (3) is a 3D diagram of the relationship between the heating unit, the load power center load and the power generation coal consumption output by the prediction model of the coal consumption of heating power generation of unit 1 with load power center in the non-heating season of the present invention;

[0030] Figure 2 b (1) is a 3D diagram of the relationship between unit load, heating and power generation coal consumption output by the prediction model of the coal consumption of the power center with heating and power generation of unit 2 in the non-heating season of the present invention;

[0031] Figure 2 b(2) is a 3D diagram showing the relationship between the unit load, the load of the power center and the power generation coal consumption output by the prediction model of the coal consumption of the power center with heating in the non-heating season of the present invention;

[0032] Figure 2 b (3) is a 3D diagram of the relationship between the heating unit, the load power center load and the power generation coal consumption output by the prediction model of the coal consumption of heating power generation of unit 2 with load power center in the non-heating season of the present invention;

[0033] Figure 2 c (1) is a 3D diagram of the relationship between unit load, heating and power generation coal consumption output by the coal consumption prediction model for heating and power generation of a dual-unit operation-loaded power center in the non-heating season of the present invention;

[0034] Figure 2 c (2) is a 3D diagram of the relationship between unit load, load power center load and power generation coal consumption output by the prediction model of the coal consumption of heating power generation with two units in non-heating season according to the present invention;

[0035] Figure 2 c (3) is a 3D diagram of the relationship between the unit heating, load power center load and power generation coal consumption output by the prediction model of the coal consumption of heating and power generation in the non-heating season of the present invention with two units in operation with load power center;

[0036] Figure 2d(1) and 2d(2) are 3D graphs of the relationship between unit load, heating and power generation coal consumption in the dual-machine operation mode in the heating season output by the dual-machine operation power generation coal consumption prediction model of the present invention.

[0037] Figure 3 This is a chart of experimental data of Unit 1 of the present invention.

[0038] Figure 4 It is a graph of experimental data of Unit 2 of the present invention.

[0039] Figure 5 It is a curve diagram of the difference in coal consumption for power generation before and after the load power center of the unit of the present invention is put into operation.

[0040] Figure 6 It is a diagram of the model usage interface of the present invention. DETAILED DESCRIPTION

[0041] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings, but this embodiment should not be understood as limiting the present invention.

[0042] The present invention refers to Figures 1 to 6 As shown:

[0043] A coal consumption prediction system for a load-carrying power center cogeneration unit based on big data, the implementation steps are as follows:

[0044] Step S1, by carrying out unit performance tests, collecting original test data of power generation coal consumption and power supply coal consumption when the unit is operating at different load rates and when the load power center is operating at different loads;

[0045] Unit 1 data

[0046]

[0047] The experimental data of Unit 1 is as follows: Figure 3 As shown in the figure, the data set of power generation coal consumption and power supply coal consumption corresponding to different power generation and different load power center loads of Unit 1 is established through the relationship in the figure;

[0048] Unit 2 data

[0049]

[0050] The experimental data of Unit 2 is as follows: Figure 4 As shown, the relationship in the figure is used to establish a data set of power generation coal consumption and power supply coal consumption corresponding to different power generation amounts and different loads of the power center of Unit 2.

[0051] Step S2, collect the impact of different heating ratios on coal consumption under different operating modes of the unit, bring it into the data set in step S1, and produce new original data on power generation coal consumption under different heat loads in the heating season and non-heating season; establish power generation coal consumption data sets corresponding to different power generation loads and heating loads under single-machine operation, dual-machine operation in the non-heating season and dual-machine operation in the heating season.

[0052] Step S3, collect the original data for preprocessing operations as a data set for model building. First, determine the difference relationship between the power center with load and the power center without load in each load section through the test data of Unit 1 and Unit 2; obtain the impact value of the power center with different load rates on the power consumption after it is put into operation;

[0053] The difference curve of coal consumption before and after the unit load power center is put into operation is as follows: Figure 5 As shown, the influence values ​​in the figure are brought into the data set in step S2 for preprocessing, and a new data set of the power center with both heating and load for Unit 1 and Unit 2 in the non-heating season is generated, and a new data set of the power center with both heating and load for the dual-unit operation in the non-heating season is produced as the final data for building the model.

[0054] Step S4, using the final data set in step S3 and using the intelligent data modeling software DTEMPOWER, establish the following model:

[0055] S41: Select the unit test data set, use the intelligent data modeling software DTEMPOWER, and establish the prediction model of power generation coal consumption and power supply coal consumption for units 1 and 2 under the load power center operation mode;

[0056] S42: Select the training set of power generation coal consumption data corresponding to different unit loads and heat loads in the non-heating season dual-unit operation mode, and use the intelligent data modeling software DTEMPOWER to establish the non-heating season unit heating power generation coal consumption prediction model for Unit 1 and Unit 2 respectively;

[0057] S43: Select the training set of power generation coal consumption data corresponding to different unit loads and heat loads under the dual-machine operation mode in the heating season, and use the intelligent data modeling software DTEMPOWER to establish a prediction model for the coal consumption of heating power generation with units in the dual-machine operation mode in the heating season;

[0058] S44: Output the training results of S41, S42, and S43 models, keep the training results, select the training set as the data basis, and output the power generation coal consumption data sets under the three variables of power generation, heating, and load power center power generation for Unit 1 and Unit 2 in the non-heating season and in the dual-machine operation mode and the heating season. Use the intelligent data modeling software DTEMPOWER to establish the coal consumption prediction model for power generation with load power center and heating in the non-heating season for Unit 1, the coal consumption prediction model for power generation with load power center and heating in the non-heating season for Unit 2, the coal consumption prediction model for power generation with load power center and heating in the non-heating season for dual-machine operation, and the coal consumption prediction model for power generation with load power center and heating in the heating season for dual-machine operation.

[0059] like Figure 2 a(1) is a 3D diagram of the relationship between unit load, heating and power generation coal consumption output by the prediction model of the power center with heating power generation coal consumption in the non-heating season of unit 1 of the present invention; the relationship between different unit loads and different heating amounts and the corresponding power generation coal consumption can be obtained;

[0060] Figure 2 a(2) is a 3D graph of the relationship between the unit load, the load power center load and the power generation coal consumption output by the prediction model of the coal consumption of the power center with heating in the non-heating season of the present invention; the relationship between different unit loads and different load power center loads and the corresponding power generation coal consumption can be obtained;

[0061] Figure 2 a(3) is a 3D graph of the relationship between unit heating, load power center load and power generation coal consumption output by the prediction model of the coal consumption of power generation with load power center in the non-heating season of the present invention; the relationship between different unit heating and different load power center load and corresponding power generation coal consumption can be obtained;

[0062] like Figure 2 b (1) is a 3D diagram of the relationship between unit load, heating and power generation coal consumption output by the prediction model of the power center with heating power generation coal consumption in the non-heating season of unit 2 of the present invention; the relationship between different unit loads and different heating amounts and the corresponding power generation coal consumption can be obtained;

[0063] Figure 2 b (2) is a 3D graph of the relationship between the unit load, the load power center load and the power generation coal consumption output by the prediction model of the coal consumption of the power center with heating in the non-heating season of the present invention; the relationship between different unit loads and different load power center loads and the corresponding power generation coal consumption can be obtained;

[0064] Figure 2b (3) is a 3D graph of the relationship between unit heating, load power center load and power generation coal consumption output by the prediction model of the coal consumption of power generation with load power center in the non-heating season of the present invention; the relationship between different unit heating and different load power center load and corresponding power generation coal consumption can be obtained;

[0065] Figure 2 c (1) is a 3D graph of the relationship between unit load, heating and power generation coal consumption output by the prediction model of the coal consumption of power center with heating and power generation with dual-unit operation in the non-heating season of the present invention; the relationship between different unit loads and different heating amounts and the corresponding power generation coal consumption during dual-unit operation can be obtained;

[0066] Figure 2 c (2) is a 3D diagram of the relationship between unit load, load power center load and power generation coal consumption output by the prediction model of the coal consumption of heating power generation with load power center in the non-heating season of the present invention; the relationship between different unit loads and different load power center loads and corresponding power generation coal consumption during the operation of the two machines can be obtained;

[0067] Figure 2 c (3) is a 3D diagram of the relationship between unit heating, load power center load and power generation coal consumption output by the prediction model of the coal consumption of power generation with load power center in the non-heating season of the present invention; the relationship between the heating of different units and the load of different power center loads and the corresponding power generation coal consumption during the operation of the two units can be obtained;

[0068] Figure 2 d(1) and 2d(2) are 3D graphs of the relationship between unit load, heating and power generation coal consumption in the dual-unit operation mode in the heating season output by the dual-unit operation power generation coal consumption prediction model of the present invention; the relationship between different unit loads and different heating and corresponding power generation coal consumption can be obtained. In the heating season, since the load power center is affected by other conditions, the operation of the load power center does not need to be considered.

[0069] S45: Output S44 different working condition model training sets, verify and summarize with historical data, regularly select training sets, optimize prediction models, and the model usage interface is as follows Figure 6 shown.

[0070] Step S5, statistically analyzing the prediction results, obtaining an analysis report, and making guidance and suggestions on the unit operation mode and load distribution.

[0071] Step S6, optimize the unit operation mode in combination with the external environmental conditions and internal factors that affect the operation of the unit. External environmental conditions include the renewable energy power generation in the region, seasonal hydropower conditions, and changes in industrial electricity consumption, and internal factors include unit maintenance schedules, equipment defects, fuel procurement, etc.; combined with external environmental conditions and internal factors, determine the basic power generation load and heat supply of the unit, use the established model to predict the coal consumption results of power generation under different working conditions, and according to the prediction result report, formulate the optimization of the unit operation mode, thermal power load distribution, and load power center operation mode and load distribution, etc., to achieve the optimal economic operation plan.

[0072] Parts not described in detail in this specification are well-known technologies in the art.

[0073] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A coal consumption prediction system for cogeneration units in a load power center based on big data, characterized in that: The steps include: S1: Carry out unit performance test, collect original test data of power generation coal consumption and power supply coal consumption at different load rates, when the unit is running with load power center, and when the load power center is carrying different loads; S2: Collect the original data of coal consumption for power generation under different operation modes and different heat loads in heating season and non-heating season, and store the acquired data in the database to provide a data basis for data preprocessing; S3: Collect raw data for preprocessing, fit, expand and clean the test data, and establish a database for the model; S4: Establish a data prediction model to estimate the power generation coal consumption and power supply coal consumption of each unit under different variables; S5: Statistical analysis of prediction results and drawing up an analysis report; S6: Optimize the unit operation mode by combining the external environmental conditions and internal factors that affect the unit operation; in: The different operation modes of the unit in step S2 include: a single-unit operation mode in the non-heating season, a dual-unit operation mode in the non-heating season, and a dual-unit operation mode in the heating season. The power generation coal consumption corresponding to different electrical loads and thermal loads under these three operation modes is collected, and the obtained data is stored in a database to provide a data basis for data preprocessing; The step S4 of establishing the prediction model includes the following steps: S41: Select the unit test data set, use the intelligent data modeling software DTEMPOWER, and establish the prediction model of power generation coal consumption and power supply coal consumption for units 1 and 2 under the load power center operation mode; S42: Select the training set of power generation coal consumption data corresponding to different unit loads and heat loads in the non-heating season dual-unit operation mode, and use the intelligent data modeling software DTEMPOWER to establish the non-heating season unit heating power generation coal consumption prediction model for Unit 1 and Unit 2 respectively; S43: Select the training set of power generation coal consumption data corresponding to different unit loads and heat loads under the dual-machine operation mode in the heating season, and use the intelligent data modeling software DTEMPOWER to establish a prediction model for the coal consumption of heating power generation with units in the dual-machine operation mode in the heating season; S44: Output the training results of the S41, S42 and S43 models, save the training results, select the training set as the data basis, and output the power generation coal consumption data sets under the three variables of power generation, heating and load power center power generation for Unit 1 and Unit 2 in the non-heating season and the dual-machine operation in the non-heating season and the heating season. Use the intelligent data modeling software DTEMPOWER to establish the coal consumption prediction model for the power generation of Unit 1 with load power center and heating in the non-heating season, the coal consumption prediction model for the power generation of Unit 2 with load power center and heating in the non-heating season, the coal consumption prediction model for the power generation of dual-machine operation with load power center and heating in the non-heating season, and the coal consumption prediction model for the power generation of dual-machine operation in the heating season; S45: Output S44 different working condition model training sets, verify and summarize with historical data, regularly select training sets, and optimize the prediction model.

2. The coal consumption prediction system for a load power center cogeneration unit based on big data according to claim 1 is characterized by: In steps S41, S42, S43 and S44, respectively establish the coal consumption prediction model of unit 1 with load power center and heating power generation in the non-heating season, the coal consumption prediction model of unit 2 with load power center and heating power generation in the non-heating season, the coal consumption prediction model of dual-unit operation with load power center and heating power generation in the non-heating season, and the coal consumption prediction model of dual-unit operation power generation in the heating season under the three variables of unit load rate, different load power center loads and different heating loads. At this time, all the prediction models are completed.

3. The coal consumption prediction system for a load power center cogeneration unit based on big data according to claim 2 is characterized by: In step S45, the training results of the model are output, and the training results obtained by inputting the training set are verified and saved. The saved training results are directly used to predict the coal consumption of the unit under actual operating conditions and to guide the preparation of optimized operation plans.

4. The coal consumption prediction system for a load power center cogeneration unit based on big data according to claim 3 is characterized by: In step S5, the prediction results are statistically analyzed to obtain an analysis report, and guidance and suggestions are made on the unit operation mode and load distribution.

5. The coal consumption prediction system for a load power center cogeneration unit based on big data according to claim 4 is characterized by: The external environmental conditions in step S6 include the renewable energy power generation in the region, seasonal hydropower conditions and changes in industrial electricity consumption, and the internal factors include unit maintenance schedules, equipment defects and fuel procurement; the basic power generation load and heating supply of the unit are determined in combination with the external environmental conditions and internal factors, and the established model is used to predict the coal consumption results of power generation under different operating conditions. According to the prediction result report, the optimized unit operation mode, thermal power load distribution and load power center operation mode and load distribution are formulated to obtain the optimal economic operation plan.

Images