CN109858125B - Thermal power unit power supply coal consumption calculation method based on radial basis function neural network - Google Patents
Thermal power unit power supply coal consumption calculation method based on radial basis function neural network Download PDFInfo
- Publication number
- CN109858125B CN109858125B CN201910061924.5A CN201910061924A CN109858125B CN 109858125 B CN109858125 B CN 109858125B CN 201910061924 A CN201910061924 A CN 201910061924A CN 109858125 B CN109858125 B CN 109858125B
- Authority
- CN
- China
- Prior art keywords
- data
- neural network
- operation data
- radial basis
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Feedback Control In General (AREA)
- Control Of Turbines (AREA)
Abstract
The invention relates to a thermal power unit power supply coal consumption calculation method based on a radial basis function neural network, which comprises the following steps: historical operation data of each unit is obtained from a distributed control system of a power plant, and historical power supply coal consumption data is obtained from a plant-level real-time monitoring information system of the power plant; the transient operation data of the unit is identified from the historical operation data, the transient operation data is removed, and the steady operation data of the unit is obtained; constructing a radial basis neural network based on steady-state operation data and historical power supply coal consumption data, and testing the radial basis neural network until the output of the radial basis neural network meets a preset error value; real-time operation data of each unit of the power plant are obtained in real time from a distributed control system of the power plant, the real-time operation data are used as input data of a radial basis function neural network input layer, and the unit power supply coal consumption data are calculated. The method and the device realize quick and accurate calculation of the coal consumption of the thermal power plant.
Description
Technical Field
The invention belongs to the technical field of thermal power generation, and particularly relates to a thermal power unit power supply coal consumption calculation method based on a radial basis function neural network.
Background
With the gradual improvement of the national energy saving and consumption reduction requirements of the thermal power plant and the high-speed development of the intelligent power plant, the calculation of the power supply coal consumption of the thermal power plant is more and more focused. The power supply coal consumption refers to the standard coal amount which is consumed by 1kWh of electric energy per time when the thermal power plant provides electricity, and is one of main economic indexes of production and operation of the power plant. The real-time calculation of the unit power supply coal consumption is realized, which is not only the necessary trend in the process of the automation technology evolution of the thermal power unit, but also the fundamental requirement of the energy-saving refined management of the thermal power unit. The calculation of the power supply coal consumption of the thermal power plant mainly adopts the traditional efficiency calculation method, such as a forward balance method and a reverse balance method. The power supply coal consumption is a typical nonlinear and time-varying parameter, the power supply coal consumption of the unit can continuously change along with the change of the working condition of the unit, the measuring process is complex, the measuring result error is large, the measuring cost is high, and the measured result cannot accurately reflect the power supply coal consumption characteristic of the unit in a period of time. Therefore, there is a need for a power supply coal consumption calculation method that can accurately reflect the power supply coal consumption.
Disclosure of Invention
The invention aims to provide a thermal power generating unit power supply coal consumption calculation method based on a radial basis function neural network, wherein the radial basis function neural network is applied to thermal power generating unit coal consumption calculation, and after reasonable and effective unit operation data are collected and parameters of the neural network are modified and optimized, the rapid and accurate calculation of a thermal power plant coal consumption curve is realized.
The invention provides a thermal power unit power supply coal consumption calculation method based on a radial basis function neural network, which comprises the following steps:
step 1, acquiring historical operation data of each unit from a distributed control system of a power plant, and acquiring historical power supply coal consumption data from a plant-level real-time monitoring information system of the power plant;
step 2, identifying transient operation data of the unit from the historical operation data and removing the transient operation data, so as to obtain steady operation data of the unit;
step 3, constructing a radial basis neural network based on steady-state operation data and historical power supply coal consumption data, and testing the radial basis neural network until the output of the radial basis neural network meets a preset error value;
and 4, acquiring real-time operation data of each unit of the power plant from the distributed control system of the power plant in real time, taking the real-time operation data as input data of the radial basis function neural network input layer, and calculating to obtain the unit power supply coal consumption data.
Further, the historical operation data in the step 1 comprises a plurality of steam turbine heat consumption rate, plant power consumption rate of a power plant, boiler unit load, hearth outlet oxygen amount, opening degree of each tuyere baffle, coal feeding amount of each coal mill and hearth temperature in a period of time.
Further, the method for obtaining steady-state operation data of the unit in the step 2 includes:
if the sum of the actual load of the unit and the mean square error of the main steam pressure parameter and the set value thereof is smaller than the set threshold value within 15min, judging that the unit is in a steady-state working condition, wherein the operation data under the steady-state working condition is steady-state operation data; otherwise, recursion is carried out for 5min in sequence, whether the unit is in a steady-state working condition is judged again, and the operation data meet the steady-state working condition.
Further, step 3 includes:
taking steady-state operation data and historical power supply coal consumption data as input data of a radial basis function neural network input layer, and dividing training data and test data; and randomly selecting 75% of data from the whole data as training data for the radial basis function neural network training sample, and using the rest 25% of data for the radial basis function neural network test sample.
Further, step 3 further includes:
and in the process of testing the radial basis function neural network, adjusting the number of neurons of an hidden layer of the radial basis function neural network to obtain the change condition of the output value of the radial basis function neural network.
Further, step 3 further includes:
the historical power supply coal consumption data is subjected to continuous differential differentiation so as to correspond to steady-state operation data.
Further, the power supply coal consumption data in the step 4 comprise a plurality of kinds of steam turbine heat consumption rate, plant power consumption rate, boiler unit load, hearth outlet oxygen amount, opening degree of each tuyere baffle, coal feeding amount of each coal mill and hearth temperature.
By means of the scheme, the radial basis function neural network with a certain predictability and a prospective property for the coal consumption characteristic is trained by using a large amount of power plant operation data through the thermal power unit power supply coal consumption calculation method based on the radial basis function neural network, so that the accurate and rapid calculation of the power plant coal consumption characteristic can be realized, and the accurate and effective coal consumption characteristic data has important value for guiding the safe and economic operation of the power plant under the background that the production cost is reduced when the 'blending and the burning' of coal is carried out in each large power plant at present; and communicating real-time operation data of the thermal power plant to the trained radial basis function neural network, and further calculating real-time data of coal consumption of the thermal power plant. The method not only can effectively improve the real-time performance of the coal consumption characteristic detection of the thermal power plant, but also greatly reduces the higher labor cost and the consumable cost required by the detection by using the traditional method.
The foregoing description is only an overview of the present invention, and is intended to provide a better understanding of the present invention, as it is embodied in the following description, with reference to the preferred embodiments of the present invention and the accompanying drawings.
Drawings
Fig. 1 is a flow chart of a thermal power generating unit power supply coal consumption calculation method based on a radial basis function neural network.
FIG. 2 is a diagram of a radial basis function neural network in accordance with the present invention;
FIG. 3 is a schematic flow chart of the calculation of the coal consumption of the radial basis function neural network according to an embodiment of the present invention.
Detailed Description
The following describes in further detail the embodiments of the present invention with reference to the drawings and examples. The following examples are illustrative of the invention and are not intended to limit the scope of the invention.
The embodiment provides a thermal power generating unit power supply coal consumption calculation method based on a radial basis function neural network, which comprises the following steps:
step S1, historical operation data of each unit is obtained from a distributed control system of a power plant, and historical power supply coal consumption data is obtained from a plant-level real-time monitoring information system of the power plant;
step S2, transient operation data of the unit are identified from the historical operation data and removed, so that steady operation data of the unit are obtained;
s3, constructing a radial basis neural network based on steady-state operation data and historical power supply coal consumption data, and testing the radial basis neural network until the output of the radial basis neural network meets a preset error value;
and S4, acquiring real-time operation data of each unit of the power plant from the distributed control system of the power plant in real time, taking the real-time operation data as input data of a radial basis function neural network input layer, and calculating to obtain the unit power supply coal consumption data.
By the thermal power generating unit power supply coal consumption calculation method based on the radial basis function neural network, reasonable and effective unit operation data are collected, and after parameters of the radial basis function neural network are modified and optimized, the rapid and accurate calculation of a thermal power plant coal consumption curve can be realized.
In this embodiment, the historical operating data in step S1 includes a plurality of steam turbine heat rate, plant power consumption rate, boiler unit load, furnace outlet oxygen, opening of each tuyere baffle, coal feeding amount of each coal mill, and furnace temperature for a period of time.
In this embodiment, the method for obtaining steady-state operation data of the unit in step S2 includes:
if the sum of the actual load of the unit and the mean square error of the main steam pressure parameter and the set value thereof is smaller than the set threshold value within 15min, judging that the unit is in a steady-state working condition, wherein the operation data under the steady-state working condition is steady-state operation data; otherwise, recursion is carried out for 5min in sequence, whether the unit is in a steady-state working condition is judged again, and the operation data meet the steady-state working condition.
In this embodiment, step S3 includes:
taking steady-state operation data and historical power supply coal consumption data as input data of a radial basis function neural network input layer, and dividing training data and test data; and randomly selecting 75% of data from the whole data as training data for the radial basis function neural network training sample, and using the rest 25% of data for the radial basis function neural network test sample.
In this embodiment, step 3 further includes:
and in the process of testing the radial basis function neural network, adjusting the number of neurons of an hidden layer of the radial basis function neural network to obtain the change condition of the output value of the radial basis function neural network.
In this embodiment, step S3 further includes:
the historical power supply coal consumption data is subjected to continuous differential differentiation so as to correspond to steady-state operation data.
In this embodiment, the power supply coal consumption data in step S4 includes a plurality of steam turbine heat consumption rate, plant power consumption rate, boiler unit load, furnace outlet oxygen amount, opening of each tuyere baffle, coal feeding amount of each coal mill, and furnace temperature.
The present invention will be described in further detail below.
Radial Basis Function (RBF) neural networks are neural networks that employ radial basis functions based on the principle that biological neurons have local response characteristics. The radial basis function neural network is a forward network and is divided into three layers, wherein the first layer is an input layer, and the number of nodes is equal to the input dimension; the second layer is an implicit layer, and the number of the nodes depends on different complexity of the problem; the third layer is an output layer, and the number of nodes is equal to the dimension of output data. The radial basis function neural network has the characteristics of simple structure, high convergence rate, capability of approaching any nonlinear function and the like.
Referring to fig. 2 and 3, w in fig. 2 represents a weight vector between network layers, and b represents a threshold value of each layer of the network.
According to the method for calculating the coal consumption of the power plant units based on the radial basis function neural network, real-time operation data of each unit in a Distributed Control System (DCS) of the power plant are collected through a data acquisition system and are used as input of the radial basis function neural network, and then the neural network with set parameters can directly give out the coal consumption data of the power plant units. The method specifically comprises the following steps:
(1) Collecting operation parameters which are easy to collect in a power plant such as a steam turbine heat rate, a power plant station power consumption rate, a boiler unit load, a hearth outlet oxygen amount, opening of each air port baffle, coal feeding amount of each coal mill, hearth temperature and the like in a power plant DCS historic station in a period of time (such as one year); the acquisition time interval of each set of operation data is 30 minutes, the data of 1 year are continuously acquired, transient data are removed, 15000 operation data are acquired, and the whole data scale is 15000×7=105000.
(2) Transmitting the operation parameters acquired in the step (1) to a terminal required by executing an algorithm through an OPC communication protocol, and automatically converting the operation parameters into a data format which can be identified by the algorithm at the terminal.
(3) And identifying transient operation data of the unit in the data, removing the transient operation data, and finishing steady operation data of the unit. The conditions for judging the system parameters to enter the steady state are as follows: within 15min, the sum of the mean square error of the actual load, the main steam pressure and other parameters of the unit and the respective set values is smaller than a specified threshold value, namely the unit is considered to be in a steady-state working condition, and the parameter data is stored into a steady-state operation working condition database; otherwise, recursing for 5min at the same time, and re-judging the stable state until the data meets the stable state requirement.
(4) Collecting power supply coal consumption data of a power plant recorded in a plant-level real-time monitoring information system (SIS) of the power plant, and taking the power supply coal consumption data and the operation parameters of each unit obtained by the arrangement in the step (3) as input data of a radial basis function neural network input layer, and dividing training data and test data of the network. It should be noted that, since the frequency of power supply coal consumption recorded in the SIS of a general power plant is once a day, in order to correspond to the operation data, continuous differentiation of the power supply coal consumption parameters is required. The learning strategy of the network uses a supervised selection center, and randomly selects 75% of the whole data as training samples for training the neural network, and the rest 25% of the samples are used for network test samples.
(5) Attempts are continuously made to increase the number of neurons in the middle layer and to check the output value change of the network.
(6) The algorithm terminates provided that the output of the neural network meets a predetermined error value.
(7) And outputting the calculation result through a neural network output layer.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, and it should be noted that it is possible for those skilled in the art to make several improvements and modifications without departing from the technical principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.
Claims (1)
1. A thermal power generating unit power supply coal consumption calculation method based on a radial basis function neural network is characterized by comprising the following steps:
step 1, acquiring historical operation data of each unit from a distributed control system of a power plant, and acquiring historical power supply coal consumption data from a plant-level real-time monitoring information system of the power plant;
step 2, identifying transient operation data of the unit from the historical operation data and removing the transient operation data, so as to obtain steady operation data of the unit;
step 3, constructing a radial basis neural network based on the steady-state operation data and the historical power supply coal consumption data, and testing the radial basis neural network until the output of the radial basis neural network meets a preset error value;
step 4, real-time operation data of each unit of the power plant are obtained in real time from a distributed control system of the power plant, the real-time operation data are used as input data of an input layer of the radial basis function neural network, and the unit power supply coal consumption data are calculated;
the historical operation data in the step 1 comprises a plurality of steam turbine heat consumption rate, plant power consumption rate of a power plant, boiler unit load, hearth outlet oxygen amount, opening degree of each air port baffle, coal feeding amount of each coal mill and hearth temperature in a period of time;
the method for obtaining the steady-state operation data of the unit in the step 2 comprises the following steps:
if the sum of the actual load of the unit and the mean square error of the main steam pressure parameter and the set value thereof is smaller than the set threshold value within 15min, judging that the unit is in a steady-state working condition, wherein the operation data under the steady-state working condition is steady-state operation data; otherwise, recursing for 5min in sequence, and judging whether the unit is in a steady-state working condition again until the operation data meets the steady-state working condition;
the step 3 comprises the following steps:
taking the steady-state operation data and the historical power supply coal consumption data as input data of the radial basis function neural network input layer, and dividing training data and test data; wherein 75% of the data is randomly selected from the whole data to be used as training data for radial basis function neural network training samples, and the rest 25% of the data is used for radial basis function neural network test samples;
the step 3 further includes:
in the process of testing the radial basis function neural network, the number of neurons of an hidden layer of the radial basis function neural network is adjusted to obtain the change condition of the output value of the radial basis function neural network;
the step 3 further includes:
performing continuous differentiation on the historical power supply coal consumption data to correspond to the steady-state operation data;
and the power supply coal consumption data in the step 4 comprise a plurality of types of steam turbine heat consumption rate, plant power consumption rate of a power plant, boiler unit load, hearth outlet oxygen amount, opening degree of each air port baffle, coal feeding amount of each coal mill and hearth temperature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910061924.5A CN109858125B (en) | 2019-01-22 | 2019-01-22 | Thermal power unit power supply coal consumption calculation method based on radial basis function neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910061924.5A CN109858125B (en) | 2019-01-22 | 2019-01-22 | Thermal power unit power supply coal consumption calculation method based on radial basis function neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109858125A CN109858125A (en) | 2019-06-07 |
| CN109858125B true CN109858125B (en) | 2023-10-27 |
Family
ID=66895625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910061924.5A Active CN109858125B (en) | 2019-01-22 | 2019-01-22 | Thermal power unit power supply coal consumption calculation method based on radial basis function neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109858125B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111852596B (en) * | 2020-06-29 | 2023-04-18 | 东北电力大学 | Method for predicting operation parameters and relative power generation coal consumption rate of boiler steam turbine generator unit |
| CN112926269B (en) * | 2021-03-15 | 2022-12-09 | 上海交通大学 | Method and system for grouping and cleaning power plant edge node data |
| CN113344192B (en) * | 2021-05-31 | 2022-01-11 | 中国标准化研究院 | Enterprise-level motor system energy-saving optimization automatic control method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101799848A (en) * | 2010-03-09 | 2010-08-11 | 江西省电力科学研究院 | Method for obtaining energy loss analysis parameter answer value of furnace of thermal power set |
| CN104598664A (en) * | 2014-10-13 | 2015-05-06 | 芜湖扬宇机电技术开发有限公司 | Performance monitoring method of thermal power generating unit |
| WO2017050207A1 (en) * | 2015-09-21 | 2017-03-30 | 华北电力科学研究院有限责任公司 | Method for analyzing energy efficiency of generator set |
| CN107726358A (en) * | 2017-10-12 | 2018-02-23 | 东南大学 | Boiler Combustion Optimization System and method based on CFD numerical simulations and intelligent modeling |
-
2019
- 2019-01-22 CN CN201910061924.5A patent/CN109858125B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101799848A (en) * | 2010-03-09 | 2010-08-11 | 江西省电力科学研究院 | Method for obtaining energy loss analysis parameter answer value of furnace of thermal power set |
| CN104598664A (en) * | 2014-10-13 | 2015-05-06 | 芜湖扬宇机电技术开发有限公司 | Performance monitoring method of thermal power generating unit |
| WO2017050207A1 (en) * | 2015-09-21 | 2017-03-30 | 华北电力科学研究院有限责任公司 | Method for analyzing energy efficiency of generator set |
| CN107726358A (en) * | 2017-10-12 | 2018-02-23 | 东南大学 | Boiler Combustion Optimization System and method based on CFD numerical simulations and intelligent modeling |
Non-Patent Citations (2)
| Title |
|---|
| 基于BP和RBF神经网络的机组能耗特性研究;王惠杰等;《汽轮机技术》;20171025(第05期);全文 * |
| 径向基神经网络在水轮发电机组耗流量计算中的应用;赵麦换等;《水电自动化与大坝监测》;20030830(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109858125A (en) | 2019-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108562854B (en) | A kind of motor abnormal condition on-line early warning method | |
| CN109858125B (en) | Thermal power unit power supply coal consumption calculation method based on radial basis function neural network | |
| CN109886471A (en) | Fired power generating unit load distribution method based on neural network and intelligent optimization algorithm | |
| CN109185917B (en) | Boiler combustion state online diagnosis method and system based on flame intensity signal | |
| CN106773666A (en) | A kind of model parameter acquisition methods for presurized water reactor primary Ioops system | |
| CN109767080B (en) | Demand side response evaluation method of community comprehensive energy system | |
| CN102930343A (en) | Method for energy optimization of distributed power generation and energy supply system | |
| CN110363334B (en) | Grid line loss prediction method of photovoltaic grid connection based on gray neural network model | |
| CN102637020A (en) | Integrated-type control system of roll steel heating furnace | |
| Ma et al. | Economic operation evaluation of active distribution network based on fuzzy Borda method | |
| CN108734419B (en) | Blast furnace gas scheduling system modeling method based on knowledge migration | |
| CN113158547B (en) | Regional comprehensive energy system optimal configuration method considering economy and reliability | |
| CN101598927A (en) | A kind of soda carbonization technique control system and control method thereof based on neural network | |
| CN105354761B (en) | Safety and efficiency evaluation method and system for accessing wind power into power grid | |
| CN111290267A (en) | Thermal power model identification device and identification method based on LabVIEW | |
| CN112653185B (en) | Efficiency evaluation method and system of distributed renewable energy power generation system | |
| CN113868836B (en) | Intelligent thermodynamic system on-line expert analysis platform based on big data | |
| CN115102164A (en) | Multi-unit collaborative optimization load distribution method, system, equipment and storage medium | |
| Sun et al. | Identifying supply-demand mismatches in district heating system based on association rule mining | |
| CN109268091A (en) | A kind of steam turbine valve discharge characteristic optimization system | |
| CN111552911B (en) | Quantitative analysis method for technical line loss influence factors based on multi-scene generation | |
| CN114110716A (en) | Operation energy efficiency monitoring control system and method for thermoelectric unit, computer equipment and storage medium | |
| CN112348696A (en) | Heat supply unit peak regulation upper limit evaluation method and system based on BP neural network | |
| Li et al. | Power grid load state information perception forecasting technology for battery energy storage system based on Elman neural network | |
| CN112668135A (en) | Metal wall temperature trend prediction model of thermal power plant boiler and establishment method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |