CN109858125A - A kind of fired power generating unit net coal consumption rate calculation method based on radial base neural net - Google Patents
A kind of fired power generating unit net coal consumption rate calculation method based on radial base neural net Download PDFInfo
- Publication number
- CN109858125A CN109858125A CN201910061924.5A CN201910061924A CN109858125A CN 109858125 A CN109858125 A CN 109858125A CN 201910061924 A CN201910061924 A CN 201910061924A CN 109858125 A CN109858125 A CN 109858125A
- Authority
- CN
- China
- Prior art keywords
- data
- net
- base neural
- radial base
- coal consumption
- 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.)
- Granted
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 fired power generating unit net coal consumption rate calculation method based on radial base neural net that the present invention relates to a kind of, it include: the history data of each unit to be obtained from power plant dcs, and history net coal consumption rate data are obtained from power plant level of factory Sis Based On Pi Database；The transient state operation data that unit is identified from history data is rejected, and obtains the steady-state operation data of unit；Radial base neural net is constructed based on steady-state operation data and history net coal consumption rate data, and radial base neural net is tested, until the output of radial base neural net meets preset error amount；It obtains the real-time running data of each unit in power plant in real time from power plant dcs, and using real-time running data as the input data of radial base neural net input layer, unit net coal consumption rate data is calculated.The present invention realizes the quick and precisely calculating to firepower electrical plant coal consumption.
Description
Technical field
The invention belongs to thermal power generating technology field more particularly to a kind of fired power generating unit confessions based on radial base neural net
Electric coal consumption calculating method.
Background technique
As country is to stepping up of requiring of energy saving for power plants consumption reduction and the high speed development of intelligent power plant, thermal power plant's power supply
The calculating of coal consumption more and more attention has been paid to.Net coal consumption rate refers to that thermal power plant is often provided out the mark that 1kWh electric energy averagely consumes
Quasi- coal amount is one of main economic index of power plant production and operation.It realizes that unit net coal consumption rate calculates in real time, is both thermoelectricity
The fundamental requirement of inexorable trend and Energy-saving for Thermal Power Units fine-grained management during unit automation technological evolvement.Thermoelectricity
Factory's net coal consumption rate, which calculates, mainly uses traditional efficiency calculation method, such as positive balance method and counter balancing method.Net coal consumption rate is typical
Non-linear, time-varying parameter, the net coal consumption rate of unit can over time and the variation of unit operating condition and constantly change, no
Only measurement process is complicated, measuring result error is big, and measurement cost is higher, and the result measured can not accurately reflect unit
Net coal consumption rate characteristic in a period of time.Therefore, it is necessary to the net coal consumption rate calculation methods that one kind can accurately reflect net coal consumption rate.
Summary of the invention
The fired power generating unit net coal consumption rate calculation method based on radial base neural net that the object of the present invention is to provide a kind of, will
Radial base neural net is applied among thermal power generation unit coal consumption calculating, by acquiring rationally effective data unit operation,
And after carrying out the modification optimization of parameter to neural network, the quick and precisely calculating to firepower electrical plant coal consumption curve is realized.
The fired power generating unit net coal consumption rate calculation method based on radial base neural net that the present invention provides a kind of, comprising:
Step 1, the history data of each unit is obtained from power plant dcs, and from power plant
History net coal consumption rate data are obtained in grade Sis Based On Pi Database；
Step 2, the transient state operation data of unit is identified from history data and is rejected, to obtain unit
Steady-state operation data；
Step 3, radial base neural net is constructed based on steady-state operation data and history net coal consumption rate data, and to radial base
Neural network is tested, until the output of radial base neural net meets preset error amount；
Step 4, the real-time running data of each unit in power plant is obtained in real time from power plant dcs, and will be real
When input data of the operation data as radial base neural net input layer, unit net coal consumption rate data are calculated.
Further, history data includes thermal loss of steam turbine rate, Auxiliary System in Power Plant in a period of time in step 1
It is rate, boiler controller system load, oxygen at furnace exit, each air port baffle opening, each coal pulverizer coal-supplying amount, a variety of in fire box temperature.
Further, the method for the steady-state operation data of acquisition unit includes: in step 2
If in 15min, respectively the sum of mean square deviation of setting value is small with its for the actual load and main steam pressure force parameter of unit
In the threshold value of setting, then determine that unit is in steady state condition, the operation data under the steady state condition is steady-state operation data；
Otherwise, up time recursion 5min determines whether unit is in steady state condition again, is until operation data meets steady state condition condition
Only.
Further, step 3 includes:
Using steady-state operation data and history net coal consumption rate data as the input data of radial base neural net input layer, and
Mark off training data and test data；Wherein, 75% data are randomly selected from overall data as training data, are used for
Remaining 25% data is used for radial base neural net test sample by radial base neural net training sample.
Further, step 3 further include:
It is carried out in test process to radial base neural net, adjusts radial base neural net hidden layer neuron number,
To obtain the output valve situation of change of radial base neural net.
Further, step 3 further include:
Continuous differencing is carried out to history net coal consumption rate data, to correspond to steady-state operation data.
Further, net coal consumption rate data include thermal loss of steam turbine rate, Auxiliary System in Power Plant rate, boiler controller system in step 4
It is load, oxygen at furnace exit, each air port baffle opening, each coal pulverizer coal-supplying amount, a variety of in fire box temperature.
According to the above aspect of the present invention, by the fired power generating unit net coal consumption rate calculation method based on radial base neural net, by making
Being trained with a large amount of power plant's operation datas has certain predictive and perspective radial base neural net to coal consumption characteristic, can be with
Realization accurately quickly calculates power plant's coal consumption characteristic, all reduces production cost at progress " coal mixing combustion " in each high-power station now
Background under, accurately and effectively coal consumption performance data for instruct power plant safety economical operation have important value；Firepower is sent out
Power plant's real-time running data is communicated into trained radial base neural net, and then calculates the real-time number of thermal power plant's coal consumption
According to.This method not only can effectively improve the real-time of thermal power plant's coal consumption Characteristics Detection, also significantly reduces and uses tradition side
Method carries out detecting required higher cost of labor and consumables cost.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And can be implemented in accordance with the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention and the accompanying drawings.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow charts of the fired power generating unit net coal consumption rate calculation method of radial base neural net.
Fig. 2 is radial base neural net structure chart in the present invention；
Fig. 3 is that radial base neural net coal consumption calculates outline flowchart in one embodiment of the invention.
Specific embodiment
With reference to the accompanying drawings and examples, specific embodiments of the present invention will be described in further detail.Implement below
Example is not intended to limit the scope of the invention for illustrating the present invention.
Present embodiments provide a kind of fired power generating unit net coal consumption rate calculation method based on radial base neural net, comprising:
Step S1, obtains the history data of each unit from power plant dcs, and from power plant
History net coal consumption rate data are obtained in level of factory Sis Based On Pi Database；
Step S2 identifies the transient state operation data of unit and is rejected from history data, to obtain machine
The steady-state operation data of group；
Step S3 constructs radial base neural net based on steady-state operation data and history net coal consumption rate data, and to radial direction
Base neural net is tested, until the output of radial base neural net meets preset error amount；
Step S4 obtains the real-time running data of each unit in power plant from power plant dcs in real time, and will
Unit net coal consumption rate data are calculated in input data of the real-time running data as radial base neural net input layer.
By the fired power generating unit net coal consumption rate calculation method based on radial base neural net, rationally effective unit is acquired
Operation data, and to radial base neural net carry out parameter modification optimization after, can be realized to firepower electrical plant coal consumption curve
Quick and precisely calculate.
In the present embodiment, history data includes thermal loss of steam turbine rate, power plant in a period of time in step S1
It is power consumption rate, boiler controller system load, oxygen at furnace exit, each air port baffle opening, each coal pulverizer coal-supplying amount, more in fire box temperature
Kind.
In the present embodiment, the method for the steady-state operation data of acquisition unit includes: in step S2
If in 15min, respectively the sum of mean square deviation of setting value is small with its for the actual load and main steam pressure force parameter of unit
In the threshold value of setting, then determine that unit is in steady state condition, the operation data under the steady state condition is steady-state operation data；
Otherwise, up time recursion 5min determines whether unit is in steady state condition again, is until operation data meets steady state condition condition
Only.
In the present embodiment, step S3 includes:
Using steady-state operation data and history net coal consumption rate data as the input data of radial base neural net input layer, and
Mark off training data and test data；Wherein, 75% data are randomly selected from overall data as training data, are used for
Remaining 25% data is used for radial base neural net test sample by radial base neural net training sample.
In the present embodiment, step 3 further include:
It is carried out in test process to radial base neural net, adjusts radial base neural net hidden layer neuron number,
To obtain the output valve situation of change of radial base neural net.
In the present embodiment, step S3 further include:
Continuous differencing is carried out to history net coal consumption rate data, to correspond to steady-state operation data.
In the present embodiment, net coal consumption rate data include thermal loss of steam turbine rate, Auxiliary System in Power Plant rate, boiler in step S4
It is unit load, oxygen at furnace exit, each air port baffle opening, each coal pulverizer coal-supplying amount, a variety of in fire box temperature.
Invention is further described in detail below.
Radial base (RBF) neural network is the principle for having local acknowledgement's characteristic according to biological neuron, applies radial direction
A kind of neural network of basic function.Radial base neural net is a kind of feedforward network, is divided into three layers, and first layer is input layer, section
Point number is equal to the dimension of input；The second layer is hidden layer, and node number is depending on the differing complexity of problem；Third layer is
Output layer, node number are equal to the dimension of output data.Radial base neural net have structure simple, fast convergence rate, can
The features such as Approximation of Arbitrary Nonlinear Function.
Join shown in Fig. 2, Fig. 3, w indicates that the weight vector between network layer, b indicate the threshold value of each layer of network in Fig. 2.
The power plant unit coal consumption calculating method based on radial base neural net, collects power plant by data collection system
The real-time running data of each unit in dcs (DCS), and using these data as radial base neural net
Input, then, the neural network for setting parameter can directly give the unit coal consumption data in power plant.This method specifically includes
Following steps:
(1) it collects in power plant DCS history station, (such as 1 year) thermal loss of steam turbine rate, Auxiliary System in Power Plant in a period of time
The power plants such as rate, boiler controller system load, oxygen at furnace exit, each air port baffle opening, each coal pulverizer coal-supplying amount, fire box temperature are easy
In the operating parameter of acquisition；It is divided between the acquisition time of every group of operation data 30 minutes, the continuous acquisition 1 year item data removes
Temporal Data is removed, each operation data 15000, overall data scale is 15000*7=105000.
(2) operating parameter for collecting step (1) is transferred to required for executing algorithm by OPC communications protocol
Terminal, and the data format that algorithm can identify is automatically converted in terminal.
(3) it identifies the transient state operation data of unit in data and is rejected, sort out the steady-state operation data of unit.
Decision-making system parameter enters the condition of stable state are as follows: in 15min, the parameters such as actual load and main steam pressure of unit are each with it
The sum of mean square deviation from setting value is less than defined threshold value, that is, is considered at steady state condition, supplemental characteristic is stored in stable operation
In floor data library；Otherwise, up time recursion 5min, determines stable state again, until data meet steady state requirement.
(4) collect power plant's level of factory Sis Based On Pi Database (SIS) in record power plant's net coal consumption rate data, and by its with
Step (3) arranges the operating parameter of each obtained unit, together the input data as radial base neural net input layer, and
Mark off the training data and test data of network.It should be noted that due to the net coal consumption rate frequency recorded in general power plant SIS
It is secondary be once a day, it is corresponding in order to be carried out with operation data, continuous differencing need to be carried out to net coal consumption rate parameter.The study of network
Strategy use has supervision Selection Center, and 75% randomly choosed in overall data is used as training sample, for neural network
Training, remaining 25% sample are used for network test sample.
(5) it continuously attempts to increase middle layer neuron and number, checks the output valve variation of network.
(6) algorithm terminates, and condition is that the output of neural network meets preset error amount.
(7) calculated result is exported by neural network output layer.
The above is only a preferred embodiment of the present invention, it is not intended to restrict the invention, it is noted that for this skill
For the those of ordinary skill in art field, without departing from the technical principles of the invention, can also make it is several improvement and
Modification, these improvements and modifications also should be regarded as protection scope of the present invention.
Claims (7)
1. a kind of fired power generating unit net coal consumption rate calculation method based on radial base neural net characterized by comprising
Step 1, the history data of each unit is obtained from power plant dcs, and real from power plant level of factory
When supervisory information system in obtain history net coal consumption rate data；
Step 2, the transient state operation data of unit is identified from the history data and is rejected, to obtain unit
Steady-state operation data；
Step 3, radial base neural net is constructed based on the steady-state operation data and history net coal consumption rate data, and to the diameter
It is tested to base neural net, until the output of the radial base neural net meets preset error amount；
Step 4, the real-time running data of each unit in power plant is obtained in real time from power plant dcs, and by the reality
When input data of the operation data as the radial base neural net input layer, unit net coal consumption rate data are calculated.
2. the fired power generating unit net coal consumption rate calculation method according to claim 1 based on radial base neural net, feature
It is, history data described in step 1 includes thermal loss of steam turbine rate in a period of time, Auxiliary System in Power Plant rate, boiler machine
It is group load, oxygen at furnace exit, each air port baffle opening, each coal pulverizer coal-supplying amount, a variety of in fire box temperature.
3. the fired power generating unit net coal consumption rate calculation method according to claim 2 based on radial base neural net, feature
It is, the method that the steady-state operation data of unit are obtained described in step 2 includes:
If in 15min, the actual load and main steam pressure force parameter of unit with its respectively the sum of mean square deviation of setting value be less than set
Fixed threshold value then determines that unit is in steady state condition, and the operation data under the steady state condition is steady-state operation data；It is no
Then, up time recursion 5min determines whether unit is in steady state condition again, is until operation data meets steady state condition condition
Only.
4. the fired power generating unit net coal consumption rate calculation method according to claim 3 based on radial base neural net, feature
It is, the step 3 includes:
Using the steady-state operation data and history net coal consumption rate data as the input number of the radial base neural net input layer
According to, and mark off training data and test data；Wherein, 75% data are randomly selected from overall data as the training
Data are used for radial base neural net training sample, by remaining 25% data, are used for radial base neural net test sample.
5. the fired power generating unit net coal consumption rate calculation method according to claim 4 based on radial base neural net, feature
It is, the step 3 further include:
The radial base neural net is being carried out to adjust the radial base neural net hidden layer neuron in test process
Number, to obtain the output valve situation of change of radial base neural net.
6. the fired power generating unit net coal consumption rate calculation method according to claim 5 based on radial base neural net, feature
It is, the step 3 further include:
Continuous differencing is carried out to the history net coal consumption rate data, to correspond to the steady-state operation data.
7. the fired power generating unit net coal consumption rate calculation method according to claim 6 based on radial base neural net, feature
It is, net coal consumption rate data described in step 4 includes thermal loss of steam turbine rate, Auxiliary System in Power Plant rate, boiler controller system load, burner hearth
Export oxygen amount, each air port baffle opening, each coal pulverizer coal-supplying amount, a variety of in fire box 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 true CN109858125A (en) | 2019-06-07 |
| CN109858125B 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) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111852596A (en) * | 2020-06-29 | 2020-10-30 | 东北电力大学 | Method for predicting operation parameters and relative power generation coal consumption rate of boiler steam turbine generator unit |
| CN112926269A (en) * | 2021-03-15 | 2021-06-08 | 上海交通大学 | Method and system for grouping and cleaning data of edge nodes of power plant |
| CN113344192A (en) * | 2021-05-31 | 2021-09-03 | 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 |
|---|
| 王惠杰等: "基于BP和RBF神经网络的机组能耗特性研究", 《汽轮机技术》 * |
| 赵麦换等: "径向基神经网络在水轮发电机组耗流量计算中的应用", 《水电自动化与大坝监测》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111852596A (en) * | 2020-06-29 | 2020-10-30 | 东北电力大学 | Method for predicting operation parameters and relative power generation coal consumption rate of boiler steam turbine generator unit |
| 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 |
| CN112926269A (en) * | 2021-03-15 | 2021-06-08 | 上海交通大学 | Method and system for grouping and cleaning data of edge nodes of power plant |
| CN113344192A (en) * | 2021-05-31 | 2021-09-03 | 中国标准化研究院 | Enterprise-level motor system energy-saving optimization automatic control method and system |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109858125B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109886471A (en) | Fired power generating unit load distribution method based on neural network and intelligent optimization algorithm | |
| CN109858125A (en) | A kind of fired power generating unit net coal consumption rate calculation method based on radial base neural net | |
| CN103742901B (en) | A kind of defining method of Properties of CFB power consumption analysis optimization target values system | |
| CN103326348B (en) | A kind of system improving the analysis of area power grid power supply capacity and overall process on-line monitoring | |
| CN110263395A (en) | The power plant's denitration running optimizatin method and system analyzed based on numerical simulation and data | |
| CN109583585A (en) | A kind of station boiler wall temperature prediction neural network model | |
| CN104534507A (en) | Optimal control method for combustion of boiler | |
| CN102360181B (en) | Low-temperature heat real-time optimization system based on general algorithm sequential quadratic programming (GA-SQP) mixed optimization strategy | |
| CN109767080B (en) | Demand side response evaluation method of community comprehensive energy system | |
| CN108490790A (en) | A kind of overheating steam temperature active disturbance rejection cascade control method based on multiple-objection optimization | |
| CN111080065A (en) | Network source coordination source side performance edge computing and analyzing platform based on multi-metadata fusion | |
| CN110935567A (en) | Thermal power generating unit dry-type electric precipitator optimization control method and system | |
| CN110707755A (en) | Comprehensive energy system ultra-short-term scheduling method based on energy hub under consideration of non-ideal communication condition | |
| CN110061496A (en) | The confidence level optimization method of source net lotus Multiobjective programming models | |
| CN111027258A (en) | Intelligent prediction method for generating load and heating load of supercritical unit | |
| CN110619409A (en) | Universal energy station scheduling method and device for self-adaptive perturbation quantum particle swarm | |
| CN108734419B (en) | Blast furnace gas scheduling system modeling method based on knowledge migration | |
| CN106610589A (en) | Online hardware closed-loop network source coordination linear active-disturbance-rejection control method | |
| CN110994621B (en) | Power grid active power flow adjusting method and device based on improved direct current power flow algorithm | |
| CN114110716A (en) | Operation energy efficiency monitoring control system and method for thermoelectric unit, computer equipment and storage medium | |
| CN112653185B (en) | Efficiency evaluation method and system of distributed renewable energy power generation system | |
| CN114881207A (en) | Steam pressure prediction method and device based on LSTM deep circulation neural network | |
| CN111552911B (en) | Quantitative analysis method for technical line loss influence factors based on multi-scene generation | |
| CN108090636B (en) | Line loss rate trend prediction method based on partial pressure line loss model | |
| Li et al. | Power grid load state information perception forecasting technology for battery energy storage system based on Elman neural network |
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 |