CN111523730A - Method for predicting coal consumption of cogeneration unit - Google Patents
Method for predicting coal consumption of cogeneration unit Download PDFInfo
- Publication number
- CN111523730A CN111523730A CN202010329040.6A CN202010329040A CN111523730A CN 111523730 A CN111523730 A CN 111523730A CN 202010329040 A CN202010329040 A CN 202010329040A CN 111523730 A CN111523730 A CN 111523730A
- Authority
- CN
- China
- Prior art keywords
- data
- coal consumption
- model
- influence
- parameters
- 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.)
- Pending
Links
- 239000003245 coal Substances 0.000 title claims abstract description 90
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000012360 testing method Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 26
- 230000015654 memory Effects 0.000 claims abstract description 9
- 230000000694 effects Effects 0.000 claims abstract description 5
- 230000008859 change Effects 0.000 claims description 13
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 3
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 3
- 229910052799 carbon Inorganic materials 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 229910052739 hydrogen Inorganic materials 0.000 claims description 3
- 239000001257 hydrogen Substances 0.000 claims description 3
- 125000004435 hydrogen atom Chemical class [H]* 0.000 claims description 3
- 230000007787 long-term memory Effects 0.000 claims description 3
- 229910052757 nitrogen Inorganic materials 0.000 claims description 3
- 229910052760 oxygen Inorganic materials 0.000 claims description 3
- 239000001301 oxygen Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 230000006403 short-term memory Effects 0.000 claims description 3
- 229910052717 sulfur Inorganic materials 0.000 claims description 3
- 239000011593 sulfur Substances 0.000 claims description 3
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims 1
- 238000010248 power generation Methods 0.000 description 9
- 230000007613 environmental effect Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000005485 electric heating Methods 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005494 condensation Effects 0.000 description 1
- 238000009833 condensation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Human Resources & Organizations (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
The invention discloses a method for predicting coal consumption of a cogeneration unit, which comprises the following steps: 1) taking out required data from the power plant SIS system, wherein the required data comprises stable working condition judgment basis and power supply coal consumption influence parameter data; 2) finding out data in stable working conditions in the obtained data, and dividing the data into training data and test data; 3) establishing a long-time memory network model and training the model by using training data; 4) adjusting model parameters according to the training result, and testing the prediction effect of the model by using the test data; 5) and selecting one of the influence parameters as a variable, fixing other influence parameters, and obtaining the influence relation between the influence parameters and the coal consumption of the unit. The invention considers more influencing factors, can more accurately reflect the coal consumption characteristic curve of the unit under the current environment and provides the coal consumption characteristic curves under various conditions, such as: coal consumption curves under typical temperatures in four seasons and coal consumption curves under different thermoelectric ratios.
Description
Technical Field
The invention belongs to the technical field of thermal power generation, and particularly relates to a method for predicting coal consumption of a cogeneration unit.
Background
The coal consumption characteristic curve of the cogeneration unit represents the functional relationship between the coal consumption of the unit and the power generation load and the heat supply load (or the heat-power ratio), and the economy and the energy consumption level of the unit are visually reflected. This provides important reference for unit economic operation and load optimization distribution. The coal consumption characteristic curve is generally obtained from performance parameters provided by a manufacturer or through thermodynamic experiment data, and the originally obtained coal consumption curve is inconsistent with the current actual operation condition after the unit runs for a long time or the coal quality is changed. The factors influencing the power supply coal consumption of the cogeneration unit comprise: at present, most documents only consider the influence of the unit load and the thermoelectric ratio on the power supply coal consumption, and common cogeneration units only have a coal consumption characteristic curve under a pure condensation working condition, but do not consider the power supply coal consumption with the heat supply load.
The existing methods are as follows:
[1] New science, Torrela, Juyuan, etc. the fitting algorithm of the coal consumption characteristic curve of the thermal power plant unit researches [ J ] protection and control of the electric power system, 2014, (10) 85-89.
The research object of the method is a pure generator set, the power generation load is used as an influence factor, and the relation between the power generation load and the coal consumption of the generator set is fitted by utilizing a genetic algorithm.
[2] Lvkai, KingHongyu, Zhoujia, and the like, research on the characteristics of electric heating coal of a cogeneration unit [ J ] thermal power generation, 2018, v.47; no.378(5) 48-54.
The method establishes an electric heating coal relation evaluation model for a cogeneration unit based on EBSILON software, thereby calculating and analyzing a relation characteristic curve of electric load-heat load-standard coal consumption under a design working condition.
The existing method has the following defects:
1. numerous parameters influencing the coal consumption characteristic curve are not fully analyzed, only the electric load is used as the influence parameter of the coal consumption, the influence of coal quality change, the environmental temperature and the environmental pressure is not considered, and the accuracy of the coal consumption characteristic curve is reduced to a certain extent. But also the influence of the heat supply amount (or the heat-power ratio) needs to be taken into consideration for the cogeneration unit.
2. Other commercial software is used in part of the method, and the cost for calculating the coal consumption characteristic curve is increased.
Disclosure of Invention
The invention aims to provide a method for predicting coal consumption of a cogeneration unit, which considers more influence factors, can more accurately reflect a coal consumption characteristic curve of the unit under the current environment and provides the coal consumption characteristic curves under various conditions, such as: coal consumption curves under typical temperatures in four seasons and coal consumption curves under different thermoelectric ratios.
The invention is realized by adopting the following technical scheme:
a method for predicting coal consumption of a cogeneration unit comprises the following steps:
1) taking out required data from the power plant SIS system, wherein the required data comprises stable working condition judgment basis and power supply coal consumption influence parameter data;
2) finding out data in stable working conditions in the obtained data, and dividing the data into training data and test data;
3) establishing a long-time memory network model and training the model by using training data;
4) adjusting model parameters according to the training result, and testing the prediction effect of the model by using the test data;
5) one of the influence parameters is selected as a variable, and other influence parameters are fixed to obtain the influence relation between the influence parameter and the coal consumption of the unit, so that the coal consumption under different conditions can be predicted.
The further improvement of the invention is that the specific implementation method of the step 1) is as follows:
101) the stable working condition judgment basis is as follows: the unit generated power, main steam temperature, main steam pressure, main feedwater flow and heat supply load, the power supply coal consumption influence parameter has: the method comprises the following steps of supplying power load, ambient temperature, ambient pressure, thermoelectric ratio, operating personnel and coal quality, wherein the coal quality comprises a fire coal heat value, received base ash, received base moisture, received base carbon, received base hydrogen, received base oxygen, received base nitrogen and received base sulfur; when taking data, the data interval is 5 s;
102) judging whether the cover parameters can be used as stable working conditions within a certain period of time according to the change amplitude of each parameter within 5min according to the stable working condition judgment basis taken out in the step 101); when all the conditions are met, judging that the time range belongs to a stable working condition; then integrating the data of the stable working condition together, and dividing the data into a training set and a testing set according to the ratio of 6: 1;
103) establishing a long-term and short-term memory network model, and selecting proper initial hyper-parameters comprises the following steps: dropout, learning rate, activation function, the forward propagation process calculated by the model is as follows:
a) the memory element receives the output h of the previous timet-1And the external information x at this momenttAs model inputs, combine them into a long vector, transform it into ftNamely:
ft=σ(Wf·[ht-1,xt]+bf(1)
b) respectively carrying out sigma transformation and tanh change on the long vectors combined in the step a), and calling it andentering an input gate; wherein the value of it is used to decide whether or not to accept the input information, i.e.
it=σ(Wi·[ht-1,xt]+bi) (2)
c) According to ftThe value of (A) determines the hidden layer output C for the last momentt-1Degree of retention, input after it scalingAdd up to Ct-1In (b) to obtain Ct:
d) Finally in the output gate, from OtC determining how to subject to tanh variationtAs output and is called a part of the input at the next moment;
ot=σ(Wo·[ht-1,xt]+bo) (5)
ht=ot*tanh(Ct) (6)
f) finally, updating the weights of the output unit, the hiding unit and the input unit by adopting a standard reverse error propagation algorithm until the error is converged;
104) evaluating the training result by using the average percentage error index, gradually correcting the model hyper-parameter to reduce the error of the test set, and taking the current model as the optimal model when the error of the test set cannot be improved by correcting the model hyper-parameter; wherein the calculation formula of MAPE is as follows:
105) selecting a variable parameter, and fixing the other parameters to obtain the influence relation between the parameter and the power supply coal consumption; when the ambient temperature, the ambient pressure and the thermoelectric ratio are fixed, the relation between the power supply coal consumption and the power of the generator under the situation can be predicted.
The invention has at least the following beneficial technical effects:
1. the invention aims at a more accurate relation curve of the electricity load-heat load-coal consumption obtained by the cogeneration unit, and can provide an important reference basis for economic operation and optimal load distribution of the unit, thereby achieving the purpose of reducing the power generation cost of a power plant.
2. According to the coal consumption characteristic curve of the cogeneration unit, the parameter with high correlation with the coal consumption is selected, other commercial software is not needed, and the calculation cost is reduced.
3. To reduce the cost of power generation, plants must find ways to operate more economically. The invention provides a coal consumption curve of the power supply coal consumption changing along with the power generation load when the current external natural environment, the thermoelectric ratio (or heat supply load), the current operation operators and the coal quality of the coal are operated, and provides a reference for the operation optimization distribution of a power plant, thereby achieving the purpose of reducing the power generation cost.
4. The invention has the technical key points that different refined relation curves of coal consumption and power supply load are obtained by utilizing a long-time memory network (LSTM) and utilizing relevant uncontrollable influence factors (coal quality information, environmental temperature, environmental pressure, current operation operators, thermoelectric ratio or thermal load).
Drawings
FIG. 1 is a flow chart of a method of predicting coal consumption of a cogeneration unit in accordance with the present invention;
fig. 2 is a graph of the relationship between the predicted power coal consumption and the actual power coal consumption and the power of the generator.
Detailed Description
The invention is further described below with reference to the following figures and examples.
As shown in fig. 1, the method for predicting coal consumption of a cogeneration unit provided by the present invention comprises the following steps:
firstly, 5 s-level data (see table 1) for judging stable working condition conditions and operation data (see table 2) related to power supply coal consumption are taken out from an SIS system, stable working condition data are screened out, an LSTM (long-short time memory network) is utilized to fit the influence relation between coal consumption influence factors and the coal consumption, then the relation curve of the electric load and the power supply coal consumption is obtained by manually setting the heat value of coal, the content of each element of the coal, the typical environment temperature, the typical environment pressure and the typical thermoelectric ratio (or the heat load), and by analogy, the relation curve of the environment temperature, the environment pressure, the heat load and the power supply coal consumption can be obtained respectively.
The invention specifically comprises the following implementation steps:
1) taking out required data from the power plant SIS system, wherein the required data comprises stable working condition judgment basis and power supply coal consumption influence parameter data, and the details are shown in tables 1 and 2; the method specifically comprises the following steps:
101) the stable working condition judgment basis is as follows: the unit generated power, main steam temperature, main steam pressure, main feedwater flow and heat supply load, the power supply coal consumption influence parameter has: the method comprises the following steps of supplying power load, ambient temperature, ambient pressure, thermoelectric ratio, operating personnel and coal quality, wherein the coal quality comprises a fire coal heat value, received base ash, received base moisture, received base carbon, received base hydrogen, received base oxygen, received base nitrogen and received base sulfur; when taking data, the data interval is 5 s;
102) and judging whether the cover parameter can be used as a stable working condition within a certain period of time according to the change amplitude of each parameter within 5min according to the stable working condition judgment basis taken out in the step 101), wherein the relative change amplitude of the generating power (MW) of the unit within 5min is not more than 1.5%, the absolute change amplitude of the temperature (DEG C) of main steam is not more than 5%, the absolute change amplitude of the pressure (MPa) of the main steam is not more than 0.6%, the absolute change amplitude of the temperature of reheated steam is not more than 5%, the relative change amplitude of the flow (t/h) of the main water supply is not more than 1.5%, and the relative change amplitude of the heat supply load (GJ/h) is not more than 5. And when all the conditions are met, judging that the time range belongs to the stable working condition. And then integrating the data of the stable working conditions together according to the following ratio of 6: the ratio of 1 is divided into a training set and a test set.
103) Establishing a long-term and short-term memory network model, and selecting appropriate initial hyper-parameters such as: dropout, learning rate, activation function, the forward propagation process calculated by the model is as follows:
a) the memory cell receives the output (h) from the previous timet-1) And the information (x) outside at that momentt) As model inputs, combine them into a long vector, transform it into ftNamely:
ft=σ(Wf·[ht-1,xt]+bf(1)
b) after combining in step a)Respectively perform sigma transformation and tanh transformation, called it andand enters the input gate. Where the value of it is used to decide whether to accept the input information (i.e. to accept the input information)):
it=σ(Wi·[ht-1,xt]+bi) (2)
c) According to ftDetermines the hidden layer output (C) for the last momentt-1) In the passage of itInput after scalingAdd up to Ct-1In (b) to obtain Ct:
d) Finally in the output gate, from OtC determining how to subject to tanh variationtAs an output and is referred to as part of the input at the next time.
ot=σ(Wo·[ht-1,xt]+bo) (5)
ht=ot*tanh(Ct) (6)
f) And finally, updating the weights of the output unit, the hiding unit and the input unit by adopting a standard reverse error propagation algorithm until the error is converged.
104) And evaluating the training result by using an MAPE (mean percentage error) index, gradually correcting the hyper-parameters of the model to reduce the error of the test set, and taking the current model as the optimal model when the error of the test set cannot be improved by correcting the hyper-parameters of the model. Wherein the calculation formula of MAPE is as follows:
105) And selecting a variable parameter, and fixing the other parameters to obtain the influence relation between the parameter and the power supply coal consumption. Such as fixed ambient temperature, ambient pressure and thermoelectric ratio, the relationship between the power supply coal consumption and the power of the generator under the situation can be predicted.
2) Finding out data in stable working conditions in the obtained data, and dividing the data into training data and test data;
3) establishing a long-time memory network (LSTM) model and training the model by using training data;
4) adjusting model parameters according to the training result, and testing the prediction effect of the model by using the test data;
5) and selecting one of the influence parameters as a variable, fixing other influence parameters, and obtaining the influence relation between the influence parameters and the coal consumption of the unit.
TABLE 1 Stable Condition determination parameters
Serial number | Parameter name | Unit of | Variation mode | Amplitude of change/5 min |
1 | Generating power of unit | MW | Relative to each other | 1.5% |
2 | Temperature of main steam | ℃ | Absolute | 5 |
3 | Main steam pressure | MPa | Absolute | 0.6 |
4 | Reheat steam temperature | ℃ | Absolute | 5 |
5 | Main water supply flow | t/h | Relative to each other | 1.5% |
6 | Heating load | GJ/h | Relative to each other | 1.5% |
TABLE 2 influence parameters of power supply coal consumption (non-adjustable parameters)
Representing unregistered parameters in some plant SIS systems
Examples
Collecting data
Time of day | Main steam pressure | Temperature of main steam | Main water supply flow | Reheat steam temperature | Heating load | Power of generator | Ambient temperature | Atmospheric pressure | Thermoelectric ratio | Coal consumption of power supply |
2019/1/10 0:00 | 18.1885 | 567.6161 | 825.975 | 562.517 | 1061.32 | 228.05 | 12.787 | 99.89801 | 4.658269 | 222.239 |
2019/1/10 0:00 | 18.1885 | 567.6161 | 825.975 | 562.517 | 1061.26 | 228.05 | 12.8137 | 99.898 | 4.65852 | 222.358 |
2019/1/10 0:00 | 18.1728 | 567.5021 | 827.034 | 562.49 | 1061.26 | 227.505 | 12.8137 | 99.8953 | 4.65884 | 222.358 |
2019/1/10 0:00 | 18.1728 | 567.5021 | 827.034 | 562.49 | 1061.26 | 227.505 | 12.827 | 99.8953 | 4.65884 | 222.466 |
2019/1/10 0:00 | 18.1992 | 567.418 | 827.0241 | 562.4611 | 1061.23 | 226.905 | 12.827 | 99.9035 | 4.65956 | 222.513 |
2019/1/10 0:00 | 18.1847 | 567.2751 | 826.2881 | 562.4041 | 1061.26 | 226.905 | 12.827 | 99.9035 | 4.66017 | 222.513 |
And (4) taking out relevant data in the SIS system of the 300MW unit of a certain power plant. According to table 1, stable working condition judgment conditions, stable working condition judgment programs are compiled, relevant data are screened, 5 s-level working condition data within 1 day are included in examples, 547 pieces of data corresponding to the stable working conditions are obtained, coal burning information is not recorded in a factory, the data are ignored here, and obtained influence factors include: generator power, ambient temperature, atmospheric pressure, thermoelectric ratio and power supply coal consumption. And (4) carrying out normalization processing on the data, and then dividing the normalized data into a training set and a test set, wherein the training set comprises 440 pieces, and the rest are the test sets. Influencing factors of power supply coal consumption: the method comprises the steps of taking generator power, environment temperature, atmospheric pressure and a thermoelectric ratio as input, taking power supply coal consumption as output, training an LSTM model, testing and predicting accuracy by using a test set after training is finished, testing the MAPE (average percentage error) value predicted by using an LSTM algorithm to be 0.015, assuming that the environment temperature is 13.586 ℃, the environment pressure is 99.994kPa and the thermoelectric ratio is 5.213, and obtaining a relation curve of the predicted power supply coal consumption, actual power supply coal consumption and the generator power as shown in figure 2.
Claims (2)
1. A method for predicting coal consumption of a cogeneration unit is characterized by comprising the following steps:
1) taking out required data from the power plant SIS system, wherein the required data comprises stable working condition judgment basis and power supply coal consumption influence parameter data;
2) finding out data in stable working conditions in the obtained data, and dividing the data into training data and test data;
3) establishing a long-time memory network model and training the model by using training data;
4) adjusting model parameters according to the training result, and testing the prediction effect of the model by using the test data;
5) and selecting one of the influence parameters as a variable, fixing other influence parameters, and obtaining the influence relation between the influence parameters and the coal consumption of the unit.
2. The method for predicting the coal consumption of the cogeneration unit according to claim 1, wherein the specific implementation method of the step 1) is as follows:
101) the stable working condition judgment basis is as follows: the unit generated power, main steam temperature, main steam pressure, main feedwater flow and heat supply load, the power supply coal consumption influence parameter has: the method comprises the following steps of supplying power load, ambient temperature, ambient pressure, thermoelectric ratio, operating personnel and coal quality, wherein the coal quality comprises a fire coal heat value, received base ash, received base moisture, received base carbon, received base hydrogen, received base oxygen, received base nitrogen and received base sulfur; when taking data, the data interval is 5 s;
102) judging whether the cover parameters can be used as stable working conditions within a certain period of time according to the change amplitude of each parameter within 5min according to the stable working condition judgment basis taken out in the step 101); when all the conditions are met, judging that the time range belongs to a stable working condition; and then integrating the data of the stable working conditions together according to the following ratio of 6: 1, dividing the ratio into a training set and a testing set;
103) establishing a long-term and short-term memory network model, and selecting proper initial hyper-parameters comprises the following steps: dropout, learning rate, activation function, the forward propagation process calculated by the model is as follows:
a) the memory element receives the output h of the previous timet-1And the external information x at this momenttAs model inputs, combine them into a long vector, transform it into ftNamely:
ft=σ(Wf·[ht-1,xt]+bf(1)
b) respectively carrying out sigma transformation and tanh change on the long vectors combined in the step a), and calling the long vectors as itAndentering an input gate; wherein itIs used to decide whether to accept the input information, i.e.
it=σ(Wi·[ht-1,xt]+bi) (2)
c) According to ftThe value of (A) determines the hidden layer output C for the last momentt-1In the passage of itInput after scalingAdd up to Ct-1In (b) to obtain Ct:
d) Finally in the output gate, from OtC determining how to subject to tanh variationtAs output and is called a part of the input at the next moment;
ot=σ(Wo·[ht-1,xt]+bo) (5)
ht=ot*tanh(Ct) (6)
f) finally, updating the weights of the output unit, the hiding unit and the input unit by adopting a standard reverse error propagation algorithm until the error is converged;
104) evaluating the training result by using the average percentage error index, gradually correcting the model hyper-parameter to reduce the error of the test set, and taking the current model as the optimal model when the error of the test set cannot be improved by correcting the model hyper-parameter; wherein the calculation formula of MAPE is as follows:
105) selecting a variable parameter, and fixing the other parameters to obtain the influence relation between the parameter and the power supply coal consumption; when the ambient temperature, the ambient pressure and the thermoelectric ratio are fixed, the relation between the power supply coal consumption and the power of the generator under the situation can be predicted.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010329040.6A CN111523730A (en) | 2020-04-23 | 2020-04-23 | Method for predicting coal consumption of cogeneration unit |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010329040.6A CN111523730A (en) | 2020-04-23 | 2020-04-23 | Method for predicting coal consumption of cogeneration unit |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111523730A true CN111523730A (en) | 2020-08-11 |
Family
ID=71903762
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010329040.6A Pending CN111523730A (en) | 2020-04-23 | 2020-04-23 | Method for predicting coal consumption of cogeneration unit |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111523730A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112053083A (en) * | 2020-09-15 | 2020-12-08 | 华润电力投资有限公司北方分公司 | Marginal cost measuring and calculating method, device, terminal and readable storage medium |
CN112749205A (en) * | 2020-12-09 | 2021-05-04 | 华能陕西发电有限公司 | System and method for acquiring relation curve between power of coal-fired generator set and power supply coal consumption |
CN112862213A (en) * | 2021-03-09 | 2021-05-28 | 中国华能集团清洁能源技术研究院有限公司 | Heat supply demand estimation method, system and equipment based on periodic feedback LSTM |
CN113112315A (en) * | 2021-06-15 | 2021-07-13 | 国能信控互联技术有限公司 | Electric power frequency modulation transaction auxiliary decision-making method and system |
CN113361748A (en) * | 2021-05-10 | 2021-09-07 | 东南大学 | Unit power supply coal consumption prediction method suitable for online load distribution |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140121848A1 (en) * | 2011-10-23 | 2014-05-01 | Chongqing Electric Power Research Institute | Cogeneration unit and wind power joint heating system and scheduling method therefor |
CN105184087A (en) * | 2015-09-21 | 2015-12-23 | 华北电力科学研究院有限责任公司 | Method and device for calculating influences of changes of environment temperature on coal consumption of coal-fired power generation unit |
CN205175694U (en) * | 2015-11-12 | 2016-04-20 | 华电电力科学研究院 | Combined heat and power units economic benefits's on -line monitoring device |
CN107330543A (en) * | 2017-06-01 | 2017-11-07 | 华北电力大学 | A kind of coal consumption method for optimization analysis based on load characteristics clustering and Controlling UEP |
CN107525684A (en) * | 2017-07-03 | 2017-12-29 | 国网山东省电力公司电力科学研究院 | A kind of ratification method and system of cogeneration units heat supply period net coal consumption rate |
CN109657783A (en) * | 2019-01-08 | 2019-04-19 | 浙江大学 | The coal cutter memorized cutting system with long temporary memory of strong robust |
CN109785187A (en) * | 2019-03-14 | 2019-05-21 | 国网山东省电力公司电力科学研究院 | A kind of electric set electric supply coal consumption detection data bearing calibration |
CN110285403A (en) * | 2019-06-10 | 2019-09-27 | 华北电力大学 | Main Steam Temperature Control method based on controlled parameter prediction |
CN110322096A (en) * | 2019-03-11 | 2019-10-11 | 华电电力科学研究院有限公司 | A kind of method of determining cogeneration units heat supply coal consumption |
-
2020
- 2020-04-23 CN CN202010329040.6A patent/CN111523730A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140121848A1 (en) * | 2011-10-23 | 2014-05-01 | Chongqing Electric Power Research Institute | Cogeneration unit and wind power joint heating system and scheduling method therefor |
CN105184087A (en) * | 2015-09-21 | 2015-12-23 | 华北电力科学研究院有限责任公司 | Method and device for calculating influences of changes of environment temperature on coal consumption of coal-fired power generation unit |
CN205175694U (en) * | 2015-11-12 | 2016-04-20 | 华电电力科学研究院 | Combined heat and power units economic benefits's on -line monitoring device |
CN107330543A (en) * | 2017-06-01 | 2017-11-07 | 华北电力大学 | A kind of coal consumption method for optimization analysis based on load characteristics clustering and Controlling UEP |
CN107525684A (en) * | 2017-07-03 | 2017-12-29 | 国网山东省电力公司电力科学研究院 | A kind of ratification method and system of cogeneration units heat supply period net coal consumption rate |
CN109657783A (en) * | 2019-01-08 | 2019-04-19 | 浙江大学 | The coal cutter memorized cutting system with long temporary memory of strong robust |
CN110322096A (en) * | 2019-03-11 | 2019-10-11 | 华电电力科学研究院有限公司 | A kind of method of determining cogeneration units heat supply coal consumption |
CN109785187A (en) * | 2019-03-14 | 2019-05-21 | 国网山东省电力公司电力科学研究院 | A kind of electric set electric supply coal consumption detection data bearing calibration |
CN110285403A (en) * | 2019-06-10 | 2019-09-27 | 华北电力大学 | Main Steam Temperature Control method based on controlled parameter prediction |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112053083A (en) * | 2020-09-15 | 2020-12-08 | 华润电力投资有限公司北方分公司 | Marginal cost measuring and calculating method, device, terminal and readable storage medium |
CN112749205A (en) * | 2020-12-09 | 2021-05-04 | 华能陕西发电有限公司 | System and method for acquiring relation curve between power of coal-fired generator set and power supply coal consumption |
CN112749205B (en) * | 2020-12-09 | 2023-03-03 | 华能陕西发电有限公司 | System and method for acquiring relation curve between power of coal-fired power generating unit and power supply coal consumption |
CN112862213A (en) * | 2021-03-09 | 2021-05-28 | 中国华能集团清洁能源技术研究院有限公司 | Heat supply demand estimation method, system and equipment based on periodic feedback LSTM |
CN112862213B (en) * | 2021-03-09 | 2024-02-20 | 中国华能集团清洁能源技术研究院有限公司 | Heat supply demand prediction method, system and equipment based on periodic feedback LSTM |
CN113361748A (en) * | 2021-05-10 | 2021-09-07 | 东南大学 | Unit power supply coal consumption prediction method suitable for online load distribution |
CN113361748B (en) * | 2021-05-10 | 2022-11-04 | 东南大学 | Unit power supply coal consumption prediction method suitable for online load distribution |
CN113112315A (en) * | 2021-06-15 | 2021-07-13 | 国能信控互联技术有限公司 | Electric power frequency modulation transaction auxiliary decision-making method and system |
CN113112315B (en) * | 2021-06-15 | 2021-10-15 | 国能信控互联技术有限公司 | Electric power frequency modulation transaction auxiliary decision-making method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111523730A (en) | Method for predicting coal consumption of cogeneration unit | |
Song et al. | Hourly heat load prediction model based on temporal convolutional neural network | |
Zhao et al. | A MILP model concerning the optimisation of penalty factors for the short-term distribution of byproduct gases produced in the iron and steel making process | |
CN112232575B (en) | Comprehensive energy system regulation and control method and device based on multi-element load prediction | |
CN104915518A (en) | Establishing method and application of two-dimensional prediction model of silicon content in hot metal in blast furnace | |
CN113822481A (en) | Comprehensive energy load prediction method based on multi-task learning strategy and deep learning | |
CN112836429A (en) | Multi-objective optimization coal blending method based on coal quality prediction | |
Tan et al. | Optimization of PEMFC system operating conditions based on neural network and PSO to achieve the best system performance | |
CN111952965B (en) | CCHP system optimized operation method based on predictive control and interval planning | |
CN112052570A (en) | Economy backpressure optimization method of wet cooling unit of thermal power plant based on wolf algorithm | |
Webberley et al. | Study of artificial neural network based short term load forecasting | |
CN109388863B (en) | ARIMA model-based distributed photovoltaic output power prediction method | |
Yang et al. | Dynamic flexibility optimization of integrated energy system based on two-timescale model predictive control | |
Jaisumroum et al. | Forecasting uncertainty of Thailand's electricity consumption compare with using artificial neural network and multiple linear regression methods | |
Liu et al. | Forecasting of wind velocity: An improved SVM algorithm combined with simulated annealing | |
CN109270917B (en) | Intelligent power plant steam turbine bearing-oriented closed-loop control system fault degradation state prediction method | |
CN110684547A (en) | Optimized control method for biomass pyrolysis carbonization kiln | |
CN113836819B (en) | Bed temperature prediction method based on time sequence attention | |
CN115759458A (en) | Load prediction method based on comprehensive energy data processing and multi-task deep learning | |
CN112580897B (en) | Method for optimal scheduling of multi-energy power system based on parrot algorithm | |
CN115544856A (en) | Day-ahead optimized scheduling method for electric heating integrated energy system | |
Chen et al. | Variation-cognizant probabilistic power flow analysis via multi-task learning | |
Li et al. | Multi-operation conditions prediction based on least square support vector machine for blast furnace gas system in steel industry | |
Ito et al. | Multipoint-measurement multipoint-heating greenhouse temperature control with wooden pellet fuel using an adaptive model predictive control approach with a genetic algorithm | |
CN111520740B (en) | Method for coordinately optimizing operation of multiple porous medium combustors |
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 |