CN110648066B - Method for preferential generation quota of reservoir power station - Google Patents
Method for preferential generation quota of reservoir power station Download PDFInfo
- Publication number
- CN110648066B CN110648066B CN201910899923.8A CN201910899923A CN110648066B CN 110648066 B CN110648066 B CN 110648066B CN 201910899923 A CN201910899923 A CN 201910899923A CN 110648066 B CN110648066 B CN 110648066B
- Authority
- CN
- China
- Prior art keywords
- power station
- reservoir
- power generation
- priority
- power
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000010248 power generation Methods 0.000 claims abstract description 125
- 238000007637 random forest analysis Methods 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000012360 testing method Methods 0.000 claims abstract description 19
- 230000008901 benefit Effects 0.000 claims abstract description 13
- 238000003066 decision tree Methods 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 39
- 238000005070 sampling Methods 0.000 claims description 19
- 238000012163 sequencing technique Methods 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 11
- 230000000694 effects Effects 0.000 claims description 11
- 230000015572 biosynthetic process Effects 0.000 claims description 10
- 238000003973 irrigation Methods 0.000 claims description 9
- 230000002262 irrigation Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000011161 development Methods 0.000 claims description 8
- 238000011160 research Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 7
- 238000007621 cluster analysis Methods 0.000 claims description 5
- 230000005611 electricity Effects 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000009826 distribution Methods 0.000 abstract description 6
- 230000006872 improvement Effects 0.000 description 7
- 238000003860 storage Methods 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 5
- HEMHJVSKTPXQMS-UHFFFAOYSA-M Sodium hydroxide Chemical compound [OH-].[Na+] HEMHJVSKTPXQMS-UHFFFAOYSA-M 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000010437 gem Substances 0.000 description 1
- 229910001751 gemstone Inorganic materials 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000007417 hierarchical cluster analysis Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000003621 irrigation water Substances 0.000 description 1
- 230000003211 malignant effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000001483 mobilizing effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 230000002269 spontaneous effect Effects 0.000 description 1
- 239000000126 substance Substances 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
-
- 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
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Biodiversity & Conservation Biology (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a system and a method for a priority power generation quota of a reservoir power station, which comprises the following steps: constructing a reservoir power station sample data set; reservoir power station clustering analysis; selecting a sorting standard; extracting a training set; forming a test set; constructing a decision tree model; forming a random forest and voting and classifying; calculating a priority power generation index; allocating a power generation plan preferentially; determining the total amount of a priority power generation plan of a reservoir power station; determining the weight of the priority power generation plan distribution; determining a benchmark utilization hour number; and distributing the priority power generation of each power station. The invention has the advantages that: the public functional value of the reservoir power station is comprehensively embodied, and the public transparency and rational basis of the priority power generation quota are realized; the deviation caused by minority factor classification is reduced, and the integral positioning of the reservoir power station is preliminarily ensured; the influence of human factors is avoided. A common technician in the industry can operate the system to complete the prior power generation quota of the reservoir power station, so that the labor cost is reduced, and the quota efficiency is improved.
Description
Technical Field
The invention relates to the technical field of power generation quota of reservoir power stations, in particular to a method for generating power preferentially by a reservoir power station.
Background
The reservoir power station (reservoir power station for short) with the water storage regulation capacity of seasons and above has the characteristics of cleanness and environmental protection as well as new energy sources such as radial flow type hydroelectric power, wind power and photovoltaic power generation, but the reservoir power station also has various public welfare functions such as irrigation, water supply, flood control, peak regulation and frequency modulation, and the like, contributes greatly to the society and power grids, and particularly has irreplaceable effects on power grids in Sichuan and Yunnan which take hydroelectric power as main energy sources in the southwest region. With the formation of an electric power market, wind power and photovoltaic power generation implement a 'reserve price-keeping' full-amount purchasing policy, and a reservoir power station and a radial-flow power station implement a 'plan + market' double-track system, namely, the electric quantity is divided into plan electric quantity (a priority power generation plan) and market electric quantity according to whether to participate in the market, wherein the plan electric quantity implements government pricing, and the market electric quantity and the electric price are formed by the market. The cost of the reservoir power station is obviously higher than that of the common radial-flow power station in terms of immigration arrangement, environmental ecological management and restoration and the like due to the reasons of wide submergence range, long construction period and the like. The undifferentiated bidding rules lead to that the high-cost reservoir power station does not rival the low-cost radial power station, which is at a disadvantage position in the market, and the advantage of high power quality is not well reflected. The fact proves that the cost-benefit of the power generation production of the water reservoir power station under the spontaneous action of the current market mechanism is asymmetric, the external benefit cannot be fully exerted, the market cannot find the real value of the power generation production, and the configuration efficiency of the energy is low; the malignant competition causes the enterprises to be overwhelmed, and the 'excessive marketization' not only influences the normal production and operation of the existing reservoir power station, but also seriously hinders the enthusiasm of the enterprises for investing in newly-built reservoir power stations, and can bring serious influence on the health coordination and sustainable development of the power industry for a long time. At present, no related method for the prior power generation quota of the reservoir power station exists.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a method for the prior generation quota of a reservoir power station, which solves the defects in the prior art.
In order to realize the purpose, the technical scheme adopted by the invention is as follows:
a method for generating power quota by priority of a reservoir power station comprises the following steps:
step 11, constructing a reservoir power station sample data set;
according to the development tasks of the reservoir power station, analyzing the functional values of the reservoir power station in the aspects of power generation, flood control, irrigation, water supply and shipping, and collecting related quantitative indexes; and collecting benefit indexes including power grid peak regulation and frequency modulation, wind-light-water complementation, emission reduction and water abandon alleviation of the reservoir power station by combining with actual operation conditions. And constructing a sample data set by taking the indexes as the characteristics of the research object.
Step 12, reservoir power station clustering analysis;
according to the features of the data set, comprising: selecting a proper clustering analysis method according to the data quantity, the classification variable, the vacancy value and the specific classification number, clustering the reservoir power station according to the clustering processes of data preprocessing, cluster function definition, clustering or grouping and estimation output, and classifying the reservoir power station into a classes.
step 21, selecting a sorting standard;
and (3) carrying out scoring evaluation on the importance of the variables influencing the classification of the reservoir power station by adopting a random forest algorithm, and measuring the contribution of the variables by adopting a Gini index of the average change quantity of the node splitting purities. In order to reduce the randomness of the algorithm, the average score of 1000 random forests is used as the final score according to the law of large numbers. And selecting variables as the standard of the reservoir power station sequencing and the priority power generation quota according to the scoring result.
The random forest algorithm is adopted for carrying out classification problem research, and the method mainly comprises the following steps: and 4 steps of training set extraction, test set formation, decision tree model construction and forest voting classification formation.
Step 211, extracting a training set;
and (3) adopting a self-service sampling method to perform replaced random sampling, and extracting N samples from the sample capacity N to form a training set for generating a classification tree. Repeating the Ntree times to obtain Ntree training sets;
step 212, test set formation;
according to the sampling principle of the self-service sampling method, because the sampling is performed with the place-back extraction, each round needs to be performed for N times, and the probability p that the sample i is not extracted in each random extractioniComprises the following steps:
pi＝(1-1/N)N...........(1)
while
I.e. when N is sufficiently large, piConverging to 1/e, that is, about 36.8% of the original samples are not extracted in each round of extraction, and the data which is not extracted in each round is used as the out-of-bag data (out of bag) of the decision tree for the model result test, and the Ntree is repeatedly executed for Ntree times to obtain Ntree test sets;
step 213, constructing a decision tree model;
constructing a decision tree by using a training set obtained by sampling, wherein for each node branch of the tree, m is randomly extracted from n characteristics of a characteristic vectortry(mtryN, in the random forest classification algorithm, generally taking
n is the dimension of the feature vector) as the features to be selected, and then selecting one feature with the best classification effect as a split point; each classification tree is subjected to extremely-intensive splitting and full growth until training samples in the nodes belong to the same class;
step 214, forming a random forest and voting and classifying;
and constructing Ntree classification trees into random forests by using the obtained Ntree training sets, voting each classification tree in the random forests for each test set, and taking the class with the largest number of votes as the classification result of the test sample.
Step 22, calculating a priority power generation index;
the weight coefficient of the key factor is the result of the normalization of the random forest score, as shown in formula (3), the comprehensive effect of the key factor is named as a priority power generation index and is used as the standard of the ranking, namely RcjIs the standard of various reservoir power station sequencingcijThe method is a standard for sequencing and distributing priority power generation amount of each power station in the category.
In the formula: w is akIs the weight of the kth key factor; skScoring the random forest of the kth key factor; and m is the number of the selected key factors.
In the formula: i isA power station variable; j is the plant category; n isjThe total number of j-type power stations; rcjThe average value of the priority power generation indexes of the j-type power stations is obtained; r iscijThe priority power generation index of j-type ith power station; xkijThe characteristic value of the kth key factor of the ith power station of j types is obtained; other parameters have the same meaning as above.
And 3, preferentially distributing the power generation plan, comprising the following substeps:
step 31, determining the total amount E of the priority power generation plan of the reservoir power station;
and according to the annual plan arrangement, deducting the priority power generation plans of other power supplies of wind power, photovoltaic power generation, thermal power and radial flow type hydropower to obtain the total priority power generation plan of the reservoir power station.
Step 32, determining the weight distributed by the priority power generation plan;
the normalization result of the average value of the priority power generation indexes of various power stations is used as the weight for distributing the average priority power generation hours of various reservoir power stations; the normalized result of the priority power generation index of each power station in each category is used as a weight for distributing the priority power generation hours of each reservoir power station, as shown in formula (5).
In the formula: a is the number of reservoir power station categories; w is ajThe weight of the j-th reservoir power station; w is aijThe weight of the jth reservoir power station is the jth; other parameters have the same meaning as above.
Step 33, determining the number of reference utilization hours;
the reference utilization hours are calculated according to equation (6) and the reference utilization hours for each category are calculated according to equation (7).
In the formula: e is the total planned electric quantity of the reservoir power station; n is a radical ofijInstalled capacity of jth reservoir power station; t is tcBenchmark hour of utilizationThe number is not specific for convenient calculation and establishment; other parameters have the same meaning as above.
In the formula: ejPlanning the total quantity of electric quantity for the j-th reservoir power station; t is tcjThe number of hours of reference utilization of the jth reservoir power station is set for convenient calculation and has no specific meaning; other parameters have the same meaning as above.
the average priority generating hours of each type of reservoir power station is calculated according to a formula (8), and the priority generating hours of each type of reservoir power station is calculated according to a formula (9). And (4) checking the calculated priority generation planned hours of each reservoir power station to see whether the power generation capacity is exceeded or not, and if the power generation capacity is exceeded, distributing the exceeded part to other power stations of the class according to the weight.
In the formula:
the average priority generation utilization hours of the j-th power station; other parameters have the same meaning as above.
tij＝wij×tcj...........(9)
In the formula: t is tijThe number of hours of preferential power generation utilization of the jth reservoir power station of the jth class; other parameters have the same meaning as above.
The invention also discloses a system for the priority power generation quota of the reservoir power station, which comprises the following components: the system comprises a classification module, an ordering module and a quota module.
The classification module is used for carrying out cluster analysis on the reservoir power station, and the specific implementation is shown in step 1;
the sorting module determines the priority power generation index of each classified power station by adopting a random forest method based on the classification result of the classification module, and sorts various power stations by taking the priority power generation index as a sorting standard so as to ensure the integral positioning of the reservoir power stations, and the specific implementation is shown in step 2;
and (3) determining the weight coefficients of the priority power generation of various power stations in the total and the priority power generation of various power stations in the classification by the quota module based on the classification and sorting module results, and performing the priority power generation quota of the reservoir power station, which is specifically realized in the step 3.
Compared with the prior art, the invention has the advantages that:
(1) the method for carrying out the prior power generation quota of the reservoir power station according to social contribution is provided by combining the engineering development task of the reservoir power station and the additional functional value given to the reservoir power station along with the social development change, the public and transparent and rational basis of the prior power generation quota is realized, certain compensation can be carried out on the high cost of the reservoir power station construction, and the investment construction of the reservoir power station is promoted;
(2) the reservoir power stations are classified based on the clustering results of a plurality of variables, so that the deviation caused by minority factor classification is reduced, and the overall positioning of the reservoir power stations is preliminarily ensured;
(3) the method adopts a random forest algorithm to extract key factors influencing classification, and avoids the influence of human factors by using the average score of 1000 times of random forests as the weight coefficient of the key factors according to the law of large numbers that the average result of a large number of random phenomena has stability;
(4) a 'classification-sequencing-quota' reservoir power station priority power generation quota system is formed, a common technician in the industry can operate the system to complete the reservoir power station priority power generation quota, the labor cost is reduced, and the quota efficiency is improved.
Drawings
FIG. 1 is a block diagram of a framework of an embodiment of the invention;
FIG. 2 is a flow chart of random forest scoring according to an embodiment of the present invention;
FIG. 3 is a comparison graph of average projected hours for various classification plants in accordance with an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings by way of examples.
A reservoir power plant priority generation quota system, comprising: the system comprises a classification module, an ordering module and a quota module.
The classification module is used for carrying out cluster analysis on the reservoir power station, is the basis of the sequencing module and the quota module, and is specifically realized in the step 1;
the sorting module determines the priority power generation index of each classified power station by adopting a random forest method based on the classification result of the classification module, and sorts various power stations by taking the priority power generation index as a sorting standard so as to ensure the integral positioning of the reservoir power stations, and the specific implementation is shown in step 2;
and (3) determining the weight coefficients of the priority power generation of various power stations in the total and the priority power generation of various power stations in the classification by the quota module based on the classification and sorting module results, and performing the priority power generation quota of the reservoir power station, which is specifically realized in the step 3.
As shown in fig. 1, a method for allocating power generation priority to a reservoir power station includes the following steps:
step 11, constructing a reservoir power station sample data set;
the sample data set for the classification of the reservoir power station mainly represents the index of the public welfare functional value of the reservoir power station, and under the current social environment and policy background, the public welfare functional value of the reservoir power station not only represents the comprehensive utilization of the power station, but also comprises additional values which are increased along with the emergence of new industries. According to the development task of the reservoir power station, analyzing the functional values of the reservoir power station in the aspects of power generation, flood control, irrigation, water supply, shipping and the like, and collecting related quantitative indexes; and collecting the benefit indexes of the power grid peak regulation and frequency modulation, wind-light-water complementation, emission reduction, water abandon relief and the like of the reservoir power station by combining with the actual operation condition. And constructing a sample data set by taking the indexes capable of reflecting the contribution of the reservoir power station in all aspects as the characteristics of a research object.
Step 12, reservoir power station clustering analysis;
according to the characteristics of the data set, such as the size of data volume, whether classification variables exist or not, whether vacancy values exist or not and whether specific class numbers are known in advance or not, a proper clustering analysis method is selected, the reservoir power stations are clustered according to the clustering processes of data preprocessing, clustering function definition, clustering or grouping and output evaluation, and finally the reservoir power stations are divided into a classes.
step 21, selecting a sorting standard;
the importance of variables influencing the classification of the reservoir power station is graded and evaluated by adopting a random forest algorithm, namely the importance of each independent variable is graded by calculating the average contribution of each independent variable on each classification tree in a random forest, and the Gini index of the average change amount of the node splitting purity is adopted to measure the contribution of each variable. In order to reduce the randomness of the algorithm, the average score of 1000 random forests is used as the final score according to the law of large numbers. And selecting the most important variables as the standard of the reservoir power station sequencing and the priority power generation quota according to the scoring result.
The random forest algorithm is adopted to research the classification problem, and the method mainly comprises 4 steps of training set extraction, test set formation, decision tree model construction and forest voting classification formation.
Step 211, extracting a training set;
and (3) extracting N samples from the sample capacity N to form a training set by adopting a bootstrap sampling method (bootstrap sampling) with replaced random sampling, and generating a classification tree. Repeating the Ntree times to obtain Ntree training sets;
step 212, test set formation;
according to the sampling principle of the self-service sampling method, because the sampling is performed with the place-back extraction, each round needs to be performed for N times, and the probability p that the sample i is not extracted in each random extractioniComprises the following steps:
pi＝(1-1/N)N...........(1)
while
I.e. when N is sufficiently large, piConverging to 1/e, that is, about 36.8% of the original samples are not extracted in each round of extraction, and the data which is not extracted in each round is used as the out-of-bag data (out of bag) of the decision tree for the model result test, and the Ntree is repeatedly executed for Ntree times to obtain Ntree test sets;
step 213, constructing a decision tree model;
constructing a decision tree by using a training set obtained by sampling, wherein for each node branch of the tree, m is randomly extracted from n characteristics of a characteristic vectortry(mtryN, in the random forest classification algorithm, generally taking
n is the dimension of the feature vector) as the features to be selected, and then selecting one feature with the best classification effect as a split point; each classification tree is subjected to extremely-intensive splitting and full growth until training samples in the nodes belong to the same class;
step 214, forming a random forest and voting and classifying;
and constructing Ntree classification trees into random forests by using the obtained Ntree training sets, voting each classification tree in the random forests for each test set, and taking the class with the largest number of votes as the classification result of the test sample.
The flow of random forest scoring is shown in fig. 2.
Step 22, calculating a priority power generation index;
the weight coefficient of the key factor is the result of the normalization of the random forest score (formula (3)), the comprehensive effect of the key factor is named as a priority power generation index and is used as the standard of the ranking, namely RcjIs the standard of various reservoir power station sequencingcijThe power generation capacity of each power station is sorted and distributed with priority in the categoryThe standard of (2).
In the formula: w is akIs the weight of the kth key factor; skScoring the random forest of the kth key factor; and m is the number of the selected key factors.
In the formula: i is a power station variable; j is the plant category; n isjThe total number of j-type power stations; rcjThe average value of the priority power generation indexes of the j-type power stations is obtained; r iscijThe priority power generation index of j-type ith power station; xkijThe characteristic value of the kth key factor of the ith power station of j types is obtained; other parameters have the same meaning as above.
And 3, preferentially distributing the power generation plan, comprising the following substeps:
step 31, determining the total amount E of the priority power generation plan of the reservoir power station;
and according to the annual plan arrangement, deducting the priority power generation plans of other power supplies such as wind power, photovoltaic power generation, thermal power, radial flow type hydropower and the like to obtain the total priority power generation plan of the reservoir power station.
Step 32, determining the weight distributed by the priority power generation plan;
the normalization result of the average value of the priority power generation indexes of various power stations is used as the weight for distributing the average priority power generation hours of various reservoir power stations; the normalized result of the priority power generation index of each power station in each category is used as a weight for distributing the priority power generation hours of each reservoir power station, as shown in formula (5).
In the formula: w is ajThe weight of the j-th reservoir power station; w is aijThe weight of the jth reservoir power station is the jth; it is composed ofHis parameters have the same meanings as above.
Step 33, determining the number of reference utilization hours;
the number of reference utilization hours is introduced for convenience of calculation and understanding, the number of reference utilization hours is a reference standard, the product of the number of reference utilization hours and the weight coefficient of each class of power station is the average priority generation hours of the class of reservoir power station, the number of reference utilization hours is calculated according to equation (6), and the number of reference utilization hours of each class is calculated according to equation (7).
In the formula: a is the category number of reservoir power stations; e is the total planned electric quantity of the reservoir power station; n is a radical ofijInstalled capacity of jth reservoir power station; t is tcThe reference utilization hours are not particularly significant for convenient calculation and establishment; other parameters have the same meaning as above.
In the formula: ejPlanning the total quantity of electric quantity for the j-th reservoir power station; t is tcjThe number of hours of reference utilization of the jth reservoir power station is set for convenient calculation and has no specific meaning; other parameters have the same meaning as above.
the average priority generating hours of each type of reservoir power station is calculated according to a formula (8), and the priority generating hours of each type of reservoir power station is calculated according to a formula (9). And (4) checking the calculated priority generation planned hours of each reservoir power station to see whether the power generation capacity is exceeded or not, and if the power generation capacity is exceeded, distributing the exceeded part to other power stations of the class according to the weight.
In the formula:
the average priority generation utilization hours of the j-th power station; other parameters have the same meaning as above.
tij＝wij×tcj...........(9)
In the formula: t is tijThe number of hours of preferential power generation utilization of the jth reservoir power station of the jth class; other parameters have the same meaning as above.
The invention takes Sichuan province as an example to research the preferential generation quota of the reservoir power station. At present, the number of the water reservoir power stations of Sichuan province is 29, wherein the chemical water power station belongs to a new water supply reservoir power station, and relevant data are not collected, so the method carries out cluster analysis on 28 water reservoir power stations except for the school. The invention discloses a method for clustering reservoir power stations, which is characterized in that the reservoir power stations play an important role in the aspects of comprehensive utilization of water resources, flood prevention and flood control in drainage basins, improvement of energy and power structures, guarantee of power supply safety, improvement of power supply quality, improvement of ecological environment and the like, but the importance degree of each power station is different, so that different contribution rates of the reservoir power stations to a power grid and the society are reflected, the method is based on the principles of fairness, justice and disclosure, and based on the contribution to the power grid and the society, the cost difference of new and old power stations is considered at the same time, and the reservoir power stations are clustered based on 16 indexes (shown in table 1) comprising installed capacity, power generation output, power generation amount, storage capacity coefficient, construction cost and the like.
TABLE 1 Cluster analysis index for reservoir power station
Table 2 shows the results of the hierarchical cluster analysis of the 28-water-base power station based on the 16 index variables. The waterfall ditch hydropower station has large volume, installed capacity (360 ten thousand kW) and regulated storage capacity (38.82 hundred million m)3) The method is the first in provincial dispatching reservoir power stations, mainly undertakes the peak-regulating and frequency-regulating tasks of the Sichuan power grid, has high contribution rate to the power grid and the society, and is separately classified into a proper category. The Baozhu temple, the pavilion seam, the purple terrace and the pilcharm power station are divided into a second class, and the four power stations belong to reservoir power stations which are high in social public welfare and large in installation; the water conservancy hub at the pavilion opening is used as the only controllable project in the main flow cascade of the Jialing river, and the main development tasks of the water conservancy hub are power generation, flood control, irrigation and water supply and have a shipping function; the temple hydropower station mainly generates electricity, has flood control and irrigation benefits, and has an irrigation area of 15.5 kilohm2(ii) a The purple terrace water conservancy project has incomplete year adjusting capability, and the main task of engineering development is to ensure irrigation water in irrigation areas of river dams and urban water supply of metropolis, and simultaneously has comprehensive benefits of flood control, power generation, tourism and the like; the Maergai hydropower station develops a task of generating electricity and has the function of supplying water to the irrigation areas of Chengdu and Dujiang weirs together with the water conservancy pivot of the purple terrace. The tillering, tile roof mountain, Buxi, buffalo, bridge and tillite power stations are classified into the third category, which is 6 power stations with the first library capacity coefficient row in 28 power stations, all of which are above 25%, and all have annual (many years) adjustment capability, so that the tillering power stations classified into the cluster are not too thick. The remaining 17 water reservoir plants are classified as a fourth category. In conclusion, the hierarchical clustering result is reasonable, and can be used for the subsequent distribution research of the prior power generation amount. It should be noted that the first category, the second category, the third category and the fourth category merely indicate that the difference between the categories is large, and the power stations within the categories have high similarity, but do not represent priorities.
TABLE 2 results of the classification
And for the classification result, a random forest algorithm is adopted to score the contribution degrees of the 16 indexes to the classification result. According to the core idea of the law of large numbers that the average result of a large number of random phenomena has stability, the average score of 1000 random forests is adopted as the final result. As shown in table 3, the importance scores of 4 indexes, including the storage capacity coefficient, the water utilization rate, the dead-time electricity quantity ratio and the construction cost, are 2.4876, 2.2910, 1.9165 and 1.0272, which are all greater than 1, and are relatively important influencing factors. Therefore, the comprehensive effect of the four indexes is used as the standard of the sequencing and preferential power generation quota of various power stations.
TABLE 3 random forest index importance rating Table (1000 results)
According to actual conditions, the reservoir power station has different regulation periods and regulation capacities due to different reservoir capacity coefficients, so that the comprehensive benefits of the power station per se in the aspects of comprehensive utilization of water resources, flood prevention and flood control in a drainage basin, improvement of an energy power structure, guarantee of power supply safety, improvement of power supply quality, improvement of ecological environment and the like are different. On the other hand, for the Sichuan taking hydropower as a main energy source, the electric quantity in the dry period is relatively poor in the rich water period due to the influence of the seasonal regularity of rich water and poor dry period, so that the electric quantity in the dry period is balanced in the dry period and plays an important role in guaranteeing the safety of a power grid. Furthermore, under the situation that the power supply is larger than the demand and a large amount of water is abandoned in the water power in Sichuan, the reservoir power station is the main force for enriching and compensating the depletion and relieving the situation of the abandoned water, and the contribution of the reservoir power station to the power grid in the aspect of reducing the abandoned water is expressed by adopting a water utilization index. The construction cost index mainly considers the cost difference of the new and old power stations. Therefore, the comprehensive effect of the four indexes, namely the storage capacity coefficient, the construction cost, the dead-period electricity quantity proportion and the water utilization rate, is used as the standard of the sequencing and the priority power generation quota of various power stations, and is named as a priority power generation index. The weights of the indexes are normalized by the random forest score, and the weight coefficient of each index is shown in table 4.
TABLE 4 weight coefficient of each index of preferential power generation index
Table 5 shows the priority power generation index and the ranking results of each type of power station. As shown in the table, the preferential power generation indexes of various reservoir power stations are 33.87%, 44.38%, 71.35% and 29.22% respectively, wherein the highest preferential power generation index is a third power station which comprises four years of adjusting power stations of Metallurgical, Wakuyashan, Buxi and Buffalo, and two years of adjusting power stations of big bridge and sodium hydroxide; the temple of the jewel, the pavilion mouth, the plaza, the pilgrimage cover and the like are arranged at the second place; the waterfall ditch level column is third. And according to the sequencing result, respectively naming the first-type reservoir power station, the second-type reservoir power station, the third-type reservoir power station and the fourth-type reservoir power station as a third-priority power station, a second-priority power station, a first-priority power station and a fourth-priority power station. The priority ranking is in accordance with the reality, and the comprehensive effect of selecting four indexes of the storage capacity coefficient, the construction cost, the dead-period electric quantity proportion and the water utilization rate is verified from the side face to be reasonable as the standard for distributing the priority generating capacity.
TABLE 5 priority generation index and ranking results for various power stations
According to the file of the Chuan Jing letter electric power content (2019) 68, the total amount of the priority power generation plans of 30 general-dispatching water distribution reservoir power stations in 2019 of Sichuan province is 252.13 hundred million kWh, two beaches are deducted, the power generation plans of the other 28 water reservoir power stations incorporated in the invention are 206.46 hundred million kWh. Based on the data and the premise, the method adopts a method based on the priority power generation index to re-quota the priority power generation of the 28 water reservoir power stations in 2019, and for the quota priority power generation amount and the basic utilization hours, the power grid buys according to the wholesale power price and the purchase cost is calculated. The quota and purchase charge calculation results are shown in table 6, and the average planned hours for each classification utility is shown in fig. 3.
TABLE 6 priority power generation quota results based on priority power generation index
Unit: expense (Yi Yuan)
Note: the "hour bias" in the table is negative, indicating that the quota results are lower than in the original plan.
As shown in table 6 and fig. 3, compared with the priority power generation quota of the reservoir power station in the economic and informatization hall, the priority power generation quota method based on the priority power generation index of the invention has high or low distribution results. Wherein the average projected hours of utilization for the first priority and second priority power stations are reduced by 262 and 163 hours, respectively, and the third priority and fourth priority power stations are increased by 75 and 68 hours, respectively. The distribution method and results have several advantages: the quota method based on the priority power generation index is open and transparent, and has reasonable data and more convincing power and public credibility; the original priority generation plan quota method of the economic and informatization hall only has a basic principle, a specific distribution method is not clear and vague, and is contrary to the basic principle of fairness, openness and justice, so that the initiative of the reservoir power station participating in the market is not favorably mobilized. Secondly, the method is beneficial to mobilizing the enthusiasm of the reservoir power station and encourages the reservoir power station to participate in the market. The planned utilization hours of the pavilion and temple power stations in the first priority power station and the second priority power station are reduced more, the generated energy of the power stations is purchased by the power grid according to the total amount of the batch-repeat power price (standard pole power price), the power stations basically do not participate in the market, and the benefit is good, so that the planned utilization hours are properly reduced, the power stations properly participate in the market, the income of the power stations is not influenced, the market order is maintained, the production enthusiasm of other power stations is stimulated, and the fluctuation range of the planned utilization hours of other power stations in the first priority power station and the second priority power station is small; the third and fourth priority power stations have common loss due to large market participation intensity, and the planned generating hours are improved based on quota results of the priority generating indexes, so that normal production and operation of the third and fourth priority power stations are guaranteed, the market participation enthusiasm of the third and fourth priority power stations is mobilized, and especially the improvement of the planned utilizing hours of the waterfall ditch power stations of the third priority power station is compensation for main peak-and-frequency-regulation tasks of a power grid and sufficient determination of the effect of the third priority power station on the power grid. And thirdly, from the perspective of a power grid, the planned electric quantity of the reservoir power station is purchased according to the wholesale price (the price of the marker post), and the purchasing cost of the quota method is 0.73 million yuan lower than that of the original plan, so that the power purchasing cost is saved.
It will be appreciated by those of ordinary skill in the art that the examples described herein are intended to assist the reader in understanding the manner in which the invention is practiced, and it is to be understood that the scope of the invention is not limited to such specifically recited statements and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.
Claims (1)
1. A quota method of a reservoir power station priority power generation quota system is characterized in that: the quota method is realized on the basis of a reservoir power station priority power generation quota system, and the reservoir power station priority power generation quota system comprises the following steps: the system comprises a classification module, a sorting module and a quota module;
the classification module is used for carrying out cluster analysis on the reservoir power station;
the sorting module determines the priority power generation index of each classified power station by adopting a random forest method based on the classification result of the classification module, and sorts various power stations by taking the priority power generation index as a sorting standard so as to ensure the integral positioning of the reservoir power stations;
the quota module determines the weight coefficients of the priority power generation of various power stations in the total and the priority power generation of various power stations in the classification based on the classification and sorting module results, and performs the priority power generation quota of the reservoir power station;
the quota method comprises the following steps:
step 1, reservoir power station classification comprises the following substeps:
step 11, constructing a reservoir power station sample data set;
according to the development tasks of the reservoir power station, analyzing the functional values of the reservoir power station in the aspects of power generation, flood control, irrigation, water supply and shipping, and collecting related quantitative indexes; collecting benefit indexes including power grid peak regulation and frequency modulation, wind-light-water complementation, emission reduction and water abandon alleviation of a reservoir power station by combining with actual operation conditions; taking the indexes as the characteristics of a research object, and constructing a sample data set;
step 12, reservoir power station clustering analysis;
according to the features of the data set, comprising: selecting a proper clustering analysis method according to the size of data volume, whether classification variables exist, whether vacancy values exist and whether specific class numbers are known in advance, clustering the reservoir power station according to a data preprocessing, clustering function defining, clustering or grouping and evaluation output clustering process, and dividing the reservoir power station into a classes;
step 2, reservoir power station sequencing comprises the following substeps:
step 21, selecting a sorting standard;
the importance of variables influencing the classification of the reservoir power station is graded and evaluated by adopting a random forest algorithm, and the contribution of the variables is measured by adopting a Gini index of the average change quantity of node splitting purities; in order to reduce the randomness of the algorithm, the average score of the random forest for 1000 times is used as a final score according to a law of large numbers; selecting variables as the standard of reservoir power station sequencing and preferential power generation quota according to the scoring result;
the random forest algorithm is adopted for carrying out classification problem research, and the method mainly comprises the following steps: 4 steps of training set extraction, test set formation, decision tree model construction and forest voting classification formation;
step 211, extracting a training set;
random sampling with replacement is carried out by adopting a self-service sampling method, and N samples are extracted from the sample capacity N to form a training set for generating a classification tree; repeating the Ntree times to obtain Ntree training sets;
step 212, test set formation;
according to the sampling principle of the self-service sampling method, because the sampling is performed with the place-back extraction, each round needs to be performed for N times, and the probability p that the sample i is not extracted in each random extractioniComprises the following steps:
pi＝(1-1/N)N...........(1)
while
I.e. when N is sufficiently large, piConverging to 1/e, namely, in each round of extraction, 36.8% of samples in the original samples cannot be extracted, and data which is not extracted in each round is used as out-of-bag data (out of bag) of the decision tree for model result testing, and the Ntree is repeatedly executed for Ntree times to obtain Ntree test sets;
step 213, constructing a decision tree model;
constructing a decision tree by using a training set obtained by sampling, wherein for each node branch of the tree, m is randomly extracted from n characteristics of a characteristic vectortrySelecting one characteristic with the best classification effect as a split point from the characteristics serving as the characteristics to be selected; each classification tree is subjected to extremely-intensive splitting and full growth until training samples in the nodes belong to the same class;
step 214, forming a random forest and voting and classifying;
constructing Ntree classification trees into random forests by using the obtained Ntree training sets, voting each classification tree in the random forests for each test set, and taking the category with the largest number of votes as the classification result of the test sample;
step 22, calculating a priority power generation index;
the weight coefficient of the key factor is the result of the normalization of the random forest score, as shown in formula (3), the comprehensive effect of the key factor is named as a priority power generation index and is used as the standard of the ranking, namely RcjIs the standard of various reservoir power station sequencingcijThe individual power stations are sorted and distributed within the categoryA criterion of priority power generation;
in the formula: w is akIs the weight of the kth key factor; skScoring the random forest of the kth key factor; m is the number of the selected key factors;
in the formula: i is a power station variable; j is the plant category; n isjThe total number of j-type power stations; rcjThe average value of the priority power generation indexes of the j-type power stations is obtained; r iscijThe priority power generation index of j-type ith power station; xkijThe characteristic value of the kth key factor of the ith power station of j types is obtained;
and 3, preferentially distributing the power generation plan, comprising the following substeps:
step 31, determining the total amount E of the priority power generation plan of the reservoir power station;
according to the annual plan arrangement, deducting the priority power generation plans of other power supplies of wind power, photovoltaic power generation, thermal power and radial flow type hydropower to obtain the total priority power generation plan of the reservoir power station;
step 32, determining the weight distributed by the priority power generation plan;
the normalization result of the average value of the priority power generation indexes of various power stations is used as the weight for distributing the average priority power generation hours of various reservoir power stations; taking the normalization result of the priority power generation index of each power station in each category as the weight for distributing the priority power generation hours of each reservoir power station, as shown in a formula (5);
in the formula: a is the number of reservoir power station categories; w is ajThe weight of the j-th reservoir power station; w is aijThe weight of the jth reservoir power station is the jth;
step 33, determining the number of reference utilization hours;
the reference utilization hours are calculated according to equation (6), and the reference utilization hours of each category are calculated according to equation (7);
in the formula: e is the total planned electric quantity of the reservoir power station; n is a radical ofijInstalled capacity of jth reservoir power station; t is tcThe reference utilization hours are not particularly significant for convenient calculation and establishment; other parameters have the same meanings as above;
in the formula: ejPlanning the total quantity of electric quantity for the j-th reservoir power station; t is tcjThe number of hours of reference utilization of the jth reservoir power station is set for convenient calculation and has no specific meaning; other parameters have the same meanings as above;
step 4, distributing the priority power generation of each power station;
firstly, calculating the average preferential generation hours of each type of reservoir power station according to a formula (8), and calculating the preferential generation hours of each type of reservoir power station according to a formula (9); checking the calculated priority generation planned hours of each reservoir power station to see whether the power generation capacity of each reservoir power station is exceeded or not, and if the power generation capacity is exceeded, distributing the exceeded part to other power stations of the class according to the weight;
in the formula:
as a class j power stationAverage priority number of electricity generation utilization hours; other parameters have the same meanings as above;
tij＝wij×tcj…………(9)
in the formula: t is tijThe number of hours of preferential power generation utilization of the jth reservoir power station of the jth class; other parameters have the same meaning as above.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910899923.8A CN110648066B (en) | 2019-09-23 | 2019-09-23 | Method for preferential generation quota of reservoir power station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910899923.8A CN110648066B (en) | 2019-09-23 | 2019-09-23 | Method for preferential generation quota of reservoir power station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110648066A CN110648066A (en) | 2020-01-03 |
| CN110648066B true CN110648066B (en) | 2021-09-14 |
Family
ID=69010954
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910899923.8A Expired - Fee Related CN110648066B (en) | 2019-09-23 | 2019-09-23 | Method for preferential generation quota of reservoir power station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110648066B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111340295B (en) * | 2020-02-27 | 2024-03-26 | 中国能源建设集团云南省电力设计院有限公司 | Green power supply development route optimization and assessment method based on envelope curve model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260801A (en) * | 2015-10-29 | 2016-01-20 | 大连理工大学 | Long-term electric power and energy balance analysis method for large-scale power station group of hydropower-rich power grid |
| CN106022960A (en) * | 2016-05-19 | 2016-10-12 | 武汉大学 | Data mining-based optimal selection method for reservoir multi-target dispatching rule parameters |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080033786A1 (en) * | 2006-08-04 | 2008-02-07 | General Electric Company | Power generation mix forecasting modeling method |
-
2019
- 2019-09-23 CN CN201910899923.8A patent/CN110648066B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260801A (en) * | 2015-10-29 | 2016-01-20 | 大连理工大学 | Long-term electric power and energy balance analysis method for large-scale power station group of hydropower-rich power grid |
| CN106022960A (en) * | 2016-05-19 | 2016-10-12 | 武汉大学 | Data mining-based optimal selection method for reservoir multi-target dispatching rule parameters |
Non-Patent Citations (1)
| Title |
|---|
| 水库群系统发电调度的合作博弈研究;徐雨妮 等;《人民长江》;20190628;第211-218页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110648066A (en) | 2020-01-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104573879B (en) | Photovoltaic plant based on optimal similar day collection goes out force prediction method | |
| CN112184070B (en) | Multi-objective optimization scheduling method and system for cascade hydropower station with cooperative ecological flow demand | |
| WO2024051524A1 (en) | Joint prediction method and apparatus for hydraulic, wind and photovoltaic generation power | |
| CN108062560A (en) | A kind of power consumer feature recognition sorting technique based on random forest | |
| CN107609697B (en) | A kind of Wind power forecasting method | |
| CN105117810A (en) | Residential electricity consumption mid-term load prediction method under multistep electricity price mechanism | |
| CN108376262A (en) | A kind of analysis model construction method of wind power output typical characteristics | |
| CN110334870B (en) | Photovoltaic power station short-term power prediction method based on gated cyclic unit network | |
| CN102147839A (en) | Method for forecasting photovoltaic power generation quantity | |
| CN110766134A (en) | Photovoltaic power station short-term power prediction method based on cyclic neural network | |
| CN115829105A (en) | Photovoltaic power prediction method based on historical data feature search | |
| CN106022538A (en) | Photovoltaic power generating predicting method based on K-mean clustering improved generalized weather | |
| CN115775045A (en) | Photovoltaic balance prediction method based on historical similar days and real-time multi-dimensional study and judgment | |
| CN113837426A (en) | Weather-typing-based photovoltaic power prediction method | |
| CN104915727B (en) | Various dimensions allomer BP neural network optical power ultra-short term prediction method | |
| CN110648066B (en) | Method for preferential generation quota of reservoir power station | |
| CN108256923A (en) | A kind of ETC customer segmentation methods based on vehicle pass-through feature | |
| CN112561252B (en) | Reactive power combination evaluation method for power grid in new energy-containing region | |
| CN109657846A (en) | Power grid alternative subsidy scale impact factor screening technique | |
| CN105956915A (en) | Credit grade optimal dividing method based on credit similarity maximization | |
| CN102800016B (en) | Based on the credit rating method of adjustment that credit grade is mated with loss given default | |
| CN111709586B (en) | Water level typing design and control method for flood diversion in reservoir basin | |
| CN108694475A (en) | Short-term time scale photovoltaic cell capable of generating power amount prediction technique based on mixed model | |
| CN108898322A (en) | A kind of electric grid investment strategy benefit integrated evaluating method based on FCM algorithm | |
| CN116596279B (en) | Intelligent park energy consumption scheduling system |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210914 |