CN112365187A - Generating set market force abuse identification method based on Lasso-logit model - Google Patents
Generating set market force abuse identification method based on Lasso-logit model Download PDFInfo
- Publication number
- CN112365187A CN112365187A CN202011369681.0A CN202011369681A CN112365187A CN 112365187 A CN112365187 A CN 112365187A CN 202011369681 A CN202011369681 A CN 202011369681A CN 112365187 A CN112365187 A CN 112365187A
- Authority
- CN
- China
- Prior art keywords
- market
- abuse
- model
- lasso
- index
- 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
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000010248 power generation Methods 0.000 claims abstract description 21
- 238000012360 testing method Methods 0.000 claims abstract description 16
- 238000012216 screening Methods 0.000 claims abstract description 14
- 238000007689 inspection Methods 0.000 claims abstract description 6
- 230000033228 biological regulation Effects 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 7
- 230000001105 regulatory effect Effects 0.000 claims description 6
- 238000002790 cross-validation Methods 0.000 claims description 5
- 230000007774 longterm Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 230000001629 suppression Effects 0.000 claims description 3
- 238000010977 unit operation Methods 0.000 claims description 3
- 230000036541 health Effects 0.000 claims description 2
- 230000006399 behavior Effects 0.000 description 16
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000005611 electricity Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000001772 Wald test Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000002401 inhibitory effect Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012795 verification 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q10/067—Enterprise or organisation modelling
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
-
- 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)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (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)
Abstract
The invention discloses a generating set market force abuse identification method based on a Lasso-logit model, which comprises the following steps: collecting data information such as basic information, operation information and annual report information related to the power market; constructing a power generator set market abuse identification index system from three aspects of structural indexes, behavior indexes and influence indexes; selecting a power generation set market abuse identification index system by using a Lasso variable to carry out index screening and carrying out colinearity inspection; establishing a Lasso-logit model for market force abuse identification of the generator set according to the screening result and carrying out significance test; according to the constructed recognition model, recognizing and analyzing the market abuse of the generator set, so as to obtain a set of the market abuse generator set; two types of errors and ROC curves are used to evaluate the performance of the model. The invention can timely and accurately identify the illegal action of the abuse market force of the generator set, thereby effectively ensuring the stable operation of the electric power spot market.
Description
Technical Field
The invention relates to the technical field of power markets, in particular to a power generation set market force abuse identification method based on a Lasso-logit model.
Background
In recent years, with the advance of power reform, the Chinese power market has achieved remarkable results. The first 8 spot test points are subjected to test operation successively; part of the spot test has completed single-day and multi-day test settlement work. However, problems are also exposed in the propulsion process, among which there are no lack of market-oriented abuse behavior by market entities, compromising the competitiveness and effectiveness of the market, affecting the optimal allocation of limited resources, and consequently compromising the efficiency of the electricity market. Therefore, the illegal behavior of market force abuse of the market main body is identified, the risk of the market force can be effectively prevented, the fairness and the efficiency of the market are maintained, and the market loss is reduced.
Market abuse in the electric power market mainly refers to a risk caused by that market participation subjects continuously raise electricity prices or frequently fluctuate greatly by using market force, and the risk can exist in various electric power commodity transactions such as electric energy, reserve capacity, auxiliary service and the like, and different market participation subjects such as power generators, power selling companies, electric power users and the like. The invention mainly analyzes the power generating set which is abused by market force in the electric power spot market. Because the power generation enterprises on the power generation side in the power market in China have large market power and relatively weak law and law awareness, but the related measures for relieving the market power are not adopted in the domestic power market construction, the method has the necessity of identifying and preventing risks for power generators with strong market power. At present, the research for market abuse at home and abroad mainly comprises the construction of a market abuse index system, game theory deduction analysis of market abuse, comprehensive evaluation and the like. With the advance of the electric power spot market, mass data of the electric power market show the characteristics of unbalance, high dimension and the like, and market abuse presents the characteristics of multiple forms, difficult prevention, difficult supervision and great influence, thereby bringing great challenges to relevant market supervision organizations.
Therefore, in order to ensure fair competition and healthy development of the market, a market abuse identification method applicable to the power market is urgently needed, mass data can be effectively processed by the method, the method accords with the actual power market operation condition, and meanwhile, the problems of dimension disaster and overlarge calculation amount are avoided, so that a market supervisor can quickly and effectively identify market abuse behaviors of the generator set and take corresponding inhibition measures.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The invention is provided in view of the problems of the existing power set market abuse identification method and the establishment method based on the Lasso-logit model.
Therefore, the invention aims to provide a generating set market force abuse identification method and an establishment method based on a Lasso-logit model.
In order to solve the technical problems, the invention provides the following technical scheme: a generating set market force abuse identification method based on a Lasso-logit model comprises the following steps:
s1, collecting data information such as basic information, operation information and annual report information related to the power market;
s2, according to the data obtained in S1, a power generation unit market abuse identification index system is constructed from the three aspects of structure indexes, behavior indexes and influence indexes;
s3, selecting to screen indexes of the electric generating set market abuse identification index system by using a Lasso variable, carrying out collinearity inspection, and rejecting the indexes which are collinearity or irrelevant;
s4, establishing a Lasso-logit model for market abuse identification of the generator set according to the screening result, and carrying out significance test on model indexes;
s5, according to the constructed recognition model, recognizing and analyzing the market abuse of the generator set, so as to obtain a set of the market abuse generator set;
and S6, performing performance evaluation on the model by adopting two types of errors and an ROC curve.
As a preferred scheme of the Lasso-logit model-based electric generating set market force abuse identification method and the establishment method, the method comprises the following steps: in step S1, the basic information includes the power generation group to which the power generation enterprise belongs, installed capacity, the number of units, stand-alone capacity, market share, power transmission and distribution price, transaction rules, transaction implementation rules, and laws and regulations and relevant regulation specifications applicable to the power market;
the operation information comprises unit operation condition, market subject declaration information, medium and long term and spot electric energy market trading result information, medium and long term and operation day system operation actual information, market subject trading settlement information, market management information and operation summary information;
the annual report information comprises the financial health state of an enterprise, the signing and performing conditions of a trade contract, the conditions of complying with a scheduling discipline and a market rule, the conditions of whether administrative penalties or other regulatory measures exist by a regulatory agency, and the conditions of whether other conditions of violation of laws and regulations are processed by a government department or a judicial department.
As a preferred scheme of the Lasso-logit model-based electric generating set market force abuse identification method and the establishment method, the method comprises the following steps: in step S2, the electric generating set market force abuse identification index system includes a structure index, a behavior index and an influence index, wherein:
the structural class indicators include market share and remaining supply index;
the behavior index comprises a declaration stage and a clearing stage, wherein the declaration stage comprises weighted average quotation, high price declaration rate, quotation fluctuation rate and volume price index, and the clearing stage comprises medium-price rate, high price medium-price rate, dynamic market share, out-of-sequence capacity index, relative share index and marginal unit reaching limit rate;
the impact type index includes a market clearing price.
As a preferred scheme of the Lasso-logit model-based electric generating set market force abuse identification method and the establishment method, the method comprises the following steps: step S3 specifically includes the following steps:
s31, constructing a sample set by adopting a hierarchical sampling method, and avoiding model identification errors caused by data imbalance;
s32, selecting based on Lasso variables, obtaining mean square errors under a plurality of different penalty coefficients by adopting a cross validation method, and screening market abuse identification indexes of the generator set by selecting proper penalty coefficients;
and S33, further adopting the co-linearity test to verify the co-linearity among the indexes, and providing a basis for constructing the identification model.
As a preferred scheme of the Lasso-logit model-based electric generating set market force abuse identification method and the establishment method, the method comprises the following steps: step S4 specifically includes the following steps:
s41, constructing and estimating a Lasso-logit model for the abuse identification of the market force of the generator set according to the screening result and the generator set sample set;
s42, the significance of the model indexes is checked, and whether the indexes in the model can play a good explaining role in the abuse identification of the market force of the generator set is judged.
As a preferred scheme of the Lasso-logit model-based electric generating set market force abuse identification method and the establishment method, the method comprises the following steps: step S5 specifically includes the following steps:
s51, carrying out market force abuse identification analysis on the generator set in the electric power spot market according to the Lasso-logit model;
s52, analyzing the marginal contribution and influence of each index on market abuse and the influence of market abuse probability, analyzing whether the market abuse probability accords with the actual electric power market trading situation, and correspondingly providing a suppression measure.
As a preferred scheme of the Lasso-logit model-based electric generating set market force abuse identification method and the establishment method, the method comprises the following steps: step S6 specifically includes the following steps:
s61, calculating the average value and standard deviation of the total accuracy, the I-type errors and the II-type errors based on the two types of errors, and verifying the identification performance and stability of the model;
and S62, based on the ROC curve, calculating a corresponding AUC value, and verifying the effectiveness of the model.
The invention has the beneficial effects that:
1. the existing game theory derivation method has the problems that the model assumption does not conform to the actual power market running condition, the multi-solution of balance points and the like, and is difficult to be applied to practice. Or the comprehensive evaluation is carried out on the market abuse of the generator set through indexes, but certain subjectivity exists in index selection and weight setting, and the traditional comprehensive evaluation method has overlarge calculated amount and lower efficiency along with the fact that the spot trading volume is larger and larger due to typical large samples and high-dimensional characteristics of data. Based on the problems, the invention adopts the Logit model which is widely applied to risk identification in other fields, and combines Lasso variable selection to identify market force abuse.
2. On the basis of referring to the supervision indexes of the trading behavior of the generator set in the domestic and foreign power market, the invention comprehensively considers the internal factors of the trading behavior of the generator set and the external factors of the market environment, constructs a system of the market abuse identification index of the generator set from three aspects of the structural index, the behavior index and the influence index, and provides a basis for effectively identifying the market abuse in the power market.
3. In the face of complex mass electric power spot data, the Lasso variable selection can screen out important indexes for market abuse identification from a plurality of indexes and eliminate the indexes with collinearity or irrelevance, so that the dimensionality is effectively reduced, and the supervision cost of a manager is reduced.
4. Compared with other machine learning methods, the method provided by the invention can not only give model explanation and clearly express marginal contribution of each index to the recognition result, but also can realize effective control on the abuse behavior of the market power of the generator set by inhibiting or promoting corresponding index values.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
FIG. 1 is a flow diagram of an abuse identification method and an establishment method of a power generation unit market force based on a Lasso-logit model.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
Furthermore, the present invention is described in detail with reference to the drawings, and in the detailed description of the embodiments of the present invention, the cross-sectional view illustrating the structure of the device is not enlarged partially according to the general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Example 1
Referring to fig. 1, an embodiment of the present invention provides a method for identifying abuse of market force of a generator set based on a Lasso-logit model, which specifically includes the following steps:
s1, collecting data information such as basic information, operation information and annual report information related to the power market; wherein:
the basic information of the electric power market comprises a power generation group to which a power generation enterprise belongs, installed capacity, the number of units, single unit capacity, market share, power transmission and distribution price, transaction rules, transaction implementation rules, laws and regulations applicable to the electric power market and relevant regulation specifications;
the operation information of the electric power market comprises unit operation condition, market subject declaration information, medium-long term and spot electric energy market trading result information, medium-long term and operation day system operation actual information, market subject trading settlement information, market management information and operation summary information;
further:
the annual report information of the electric power market comprises the financial health status of an enterprise, the signing and performing conditions of a trade contract, the conditions of complying with a scheduling discipline and market rules, the conditions of whether administrative penalties or other regulatory measures exist by a regulatory agency, and the conditions of whether other conditions are processed by a government department or a judicial department in violation of laws and regulations.
S2, according to the data obtained in S1, a power generation unit market abuse identification index system is constructed from the three aspects of structure indexes, behavior indexes and influence indexes; referring to table 1, the identification index system for abusing the market force of the generator set includes a structure index, a behavior index and an influence index;
TABLE 1
Wherein,
the preset structure index comprises market share X1 and residual supply index X2; market share X1 reflects the market force of the generator set; the remaining supply index X2 reflects the status of the generator set in the market;
the behavior indexes comprise a declaration stage and a clearing stage, wherein the declaration stage comprises a weighted average quotation X3, a high price declaration rate X4, a quotation fluctuation rate X5 and a volume index X6, and the clearing stage comprises a medium rate X7, a high price medium rate X8, a dynamic market share X9, an out-of-order capacity index X10, a relative share index X11 and a marginal unit limit rate X12;
the weighted average quotation X3 in the declaration phase reflects the quotation level of the generator set; the high price declaration rate X4 reflects the behavior of raising the price of the generator set; the quotation fluctuation rate X5 reflects the change condition of the generator set quotation; the quantitative price index X6 reflects the behavior of the generator set reporting high price;
the winning rate X7 of the clearing stage reflects the potential market force of the generator set; the high price winning rate X8 reflects that the generator set raises the quoted price and obtains the condition close to the declared capacity; the dynamic market share X9 reflects the degree of the bid amount in the total bid amount of the generator set; the out-of-sequence capacity index X10 reflects the level of the generating set failing to bid for capacity; the relative share index X11 reflects the degree to which the share in the genset deviates from market share; the limit reaching rate X12 of the marginal unit reflects the times of the generator set serving as the marginal unit;
the impact type index includes market clearing X13. The market clearing price X13 reflects the electricity price level of the market.
S3, selecting to screen indexes of the electric generating set market abuse identification index system by using a Lasso variable, carrying out collinearity inspection, and rejecting the indexes which are collinearity or irrelevant; the method specifically comprises the following steps:
and S31, constructing a sample set by adopting a hierarchical sampling method, and avoiding model identification errors caused by data imbalance. The invention adopts a layered sampling method aiming at the unbalanced data, and respectively randomly samples the same sample data from the two categories without repetition as a sample set constructed by a model, thereby ensuring that the two categories have the same ratio.
S32, selecting based on Lasso variables, obtaining the mean square error under a plurality of different punishment coefficients by adopting a cross validation method, and screening the market abuse identification indexes of the generator set by selecting proper punishment coefficients. The Lasso variable selection is realized by constructing a penalty function and enabling coefficients of some indexes to be 0, so that the purpose of variable selection is realized, and the method can be used for solving the problem of multiple collinearity in regression analysis. The key of the Lasso variable selection lies in the selection of penalty parameters, and the invention adopts a cross verification method to determine the penalty parameters.
Statistics defining cross-validation are:
wherein,
an index vector of the generator set i; y isiThe category of the generator set i; λ is a penalty parameter;
the fitting parameters after the K-th subset is deleted are indicated, where K is 1, 2.
In the cross validation process, C isVAnd when the minimum value is reached, the corresponding punishment parameter is optimal, and the index of which the coefficient corresponding to the optimal punishment parameter is not 0 is the screened index.
And S33, further adopting the co-linearity test to verify the co-linearity among the indexes, and providing a basis for constructing the identification model. If the tolerance in the screened index collinearity test result is less than 0.1 or the variance expansion factor is more than 10, the collinearity exists among the indexes, and the corresponding indexes are removed.
S4, establishing a Lasso-logit model for market abuse identification of the generator set according to the screening result, and carrying out significance test on model indexes; the method specifically comprises the following steps:
s41, constructing and estimating a Lasso-logit model for the abuse identification of the market force of the generator set according to the screening result and the generator set sample set; firstly, constructing a two-classification Logit model by combining a screening index obtained by Lasso variable selection:
wherein, P (y)i1) probability of a genset being determined as market force abuse, yiThe value is 0 or 1 for the category of the generator set i, wherein 0 represents a generator set which is not judged to be market abuse, and 1 represents a generator set which is judged to be market abuse; x is the number ofiAn index vector of the generator set i; beta is a0Is a model parameter; beta is a parameter vector corresponding to the index.
And then substituting the constructed generator set sample set into the model to estimate parameters of the model.
S42, the significance of the model indexes is checked, and whether the indexes in the model can play a good explaining role in the abuse identification of the market force of the generator set is judged. The model indexes are verified by adopting the Wald significance test, the Wald test is a test for testing the significance of the explanatory variables in the statistical model, if the explanatory variables are important after the Wald test, parameters corresponding to the variables are not zero, and significant correlation exists between the variables and the dependent variables.
S5, according to the constructed recognition model, recognizing and analyzing the market abuse of the generator set, so as to obtain a set of the market abuse generator set; the method specifically comprises the following steps:
s51, carrying out market force abuse identification analysis on the generator set in the electric power spot market according to the Lasso-logit model; applying the Lasso-location model to the electric power market transaction, calculating and substituting corresponding index values according to the acquired data, and setting the model probability P (y)iWhen the number is 1) > 0.5, the generator set is considered to have market force abuse, otherwise, the abuse does not exist.
S52, analyzing the marginal contribution and influence of each index on market abuse and the influence of market abuse probability, analyzing whether the market abuse probability accords with the actual electric power market trading situation, and correspondingly providing a suppression measure. Firstly, analyzing the positive index coefficient according to the coefficient estimation value of the corresponding index in the model, wherein the larger the index value is, the more likely the generator set is to be abused by market force; if the index coefficient is negative, the smaller the index value is, the more likely the generator set is to be abused by market force; the larger the analysis index coefficient is, the maximum marginal contribution of the abuse probability of the market force of the generator set is shown; the smaller the index coefficient is, the minimum marginal contribution of the market abuse probability of the generator set is indicated, namely whether the generator set has the minimum correlation between the market abuse and the index; and then, whether the analysis result is consistent with the actual power market is further verified by analyzing the relation between the single index and the market abuse probability. And finally, according to the analysis result, the effective control of the market abuse behavior of the generator set is realized by inhibiting or promoting the corresponding index value.
S6, performing performance evaluation on the model by adopting two types of errors and an ROC curve; the method specifically comprises the following steps:
s61, calculating the average value and standard deviation of the total accuracy, the I-type errors and the II-type errors based on the two types of errors, and verifying the identification performance and stability of the model; two types of errors are used to evaluate the accuracy and reliability of model identification. These two types of errors refer to: a category I error identifies a market power generating unit that is abused as an unabused market power generating unit, which increases the risk to the market. The type II error is to identify the unmisused market power generator set as the misused market power generator set, so that the supervision is excessive and the market reform is not facilitated. Both types of errors are unacceptable to the electricity market. The overall accuracy, the mean and standard deviation of class I errors and class II errors were calculated by constructing a confusion matrix, as shown in table 2.
Confusion matrix of Lasso-logit model
TABLE 2
And comparing the common Logit model, randomly extracting 75 percent of the training set and the rest testing set by using the constructed sample set, respectively operating 100 times by using the common Logit model and the Lasso-Logit model, and calculating the average value and the standard deviation of the total accuracy, the type I errors and the type II errors. The larger the average value of the total accuracy is, the smaller the average values of the type I errors and the type II errors are, and the better the model performance is; the smaller the overall accuracy, class I errors, and class II errors, the more stable the model.
And S62, based on the ROC curve, calculating a corresponding AUC value, and verifying the effectiveness of the model. The ROC curve is a common evaluation standard of a binary classification model and is used for evaluating the classification performance of the model. For a good classification model, its ROC curve should be as close as possible to the upper left corner of the graph. AUC values are complementary to ROC curves, and often represent the results of ROC curves as values, ranging from 0 to 1. Generally, the higher the AUC value of the model, the higher the classification precision, and the better the classification effect.
In conclusion, the method for identifying the abuse of the market force of the generator set based on the Lasso-logit model provides a basis for identifying the abuse of the market force of the generator set by constructing a system of indexes for identifying the abuse of the market force of the generator set; index screening and collinearity inspection are carried out on a power generation unit market abuse identification index system by using Lasso variable selection, so that the dimension and the supervision cost of a market manager are effectively reduced, and the problems of dimension disaster and overlarge calculation amount are avoided; and establishing a Lasso-logic model for market abuse identification of the generator set according to the screening result, carrying out significance inspection, identifying and analyzing the generator set, effectively obtaining a market abuse generator set, and realizing risk supervision and prevention of the power market.
It is important to note that the construction and arrangement of the present application as shown in the various exemplary embodiments is illustrative only. Although only a few embodiments have been described in detail in this disclosure, those skilled in the art who review this disclosure will readily appreciate that many modifications are possible (e.g., variations in sizes, dimensions, structures, shapes and proportions of the various elements, values of parameters (e.g., temperatures, pressures, etc.), mounting arrangements, use of materials, colors, orientations, etc.) without materially departing from the novel teachings and advantages of the subject matter recited in this application. For example, elements shown as integrally formed may be constructed of multiple parts or elements, the position of elements may be reversed or otherwise varied, and the nature or number of discrete elements or positions may be altered or varied. Accordingly, all such modifications are intended to be included within the scope of this invention. The order or sequence of any process or method steps may be varied or re-sequenced according to alternative embodiments. In the claims, any means-plus-function clause is intended to cover the structures described herein as performing the recited function and not only structural equivalents but also equivalent structures. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the exemplary embodiments without departing from the scope of the present inventions. Therefore, the present invention is not limited to a particular embodiment, but extends to various modifications that nevertheless fall within the scope of the appended claims.
Moreover, in an effort to provide a concise description of the exemplary embodiments, all features of an actual implementation may not be described (i.e., those unrelated to the presently contemplated best mode of carrying out the invention, or those unrelated to enabling the invention).
It should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort might be complex and time consuming, but would nevertheless be a routine undertaking of design, fabrication, and manufacture for those of ordinary skill having the benefit of this disclosure, without undue experimentation.
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (7)
1. A generating set market force abuse identification method based on a Lasso-logit model is characterized by comprising the following steps: the method comprises the following steps:
s1, collecting data information such as basic information, operation information and annual report information related to the power market;
s2, according to the data obtained in S1, a power generation unit market abuse identification index system is constructed from the three aspects of structure indexes, behavior indexes and influence indexes;
s3, selecting to screen indexes of the electric generating set market abuse identification index system by using a Lasso variable, carrying out collinearity inspection, and rejecting the indexes which are collinearity or irrelevant;
s4, establishing a Lasso-logit model for market abuse identification of the generator set according to the screening result, and carrying out significance test on model indexes;
s5, according to the constructed recognition model, recognizing and analyzing the market abuse of the generator set, so as to obtain a set of the market abuse generator set;
and S6, performing performance evaluation on the model by adopting two types of errors and an ROC curve.
2. The Lasso-logit model-based power generation unit market force abuse identification method of claim 1, wherein:
in step S1, the basic information includes the power generation group to which the power generation enterprise belongs, installed capacity, the number of units, stand-alone capacity, market share, power transmission and distribution price, transaction rules, transaction implementation rules, and laws and regulations and relevant regulation specifications applicable to the power market;
the operation information comprises unit operation condition, market subject declaration information, medium and long term and spot electric energy market trading result information, medium and long term and operation day system operation actual information, market subject trading settlement information, market management information and operation summary information;
the annual report information comprises the financial health state of an enterprise, the signing and performing conditions of a trade contract, the conditions of complying with a scheduling discipline and a market rule, the conditions of whether administrative penalties or other regulatory measures exist by a regulatory agency, and the conditions of whether other conditions of violation of laws and regulations are processed by a government department or a judicial department.
3. The Lasso-logit model-based power generation unit market force abuse identification method of claim 1, wherein:
in step S2, the electric generating set market force abuse identification index system includes a structure index, a behavior index and an influence index, wherein:
the structural class indicators include market share and remaining supply index;
the behavior index comprises a declaration stage and a clearing stage, wherein the declaration stage comprises weighted average quotation, high price declaration rate, quotation fluctuation rate and volume price index, and the clearing stage comprises medium-price rate, high price medium-price rate, dynamic market share, out-of-sequence capacity index, relative share index and marginal unit reaching limit rate;
the impact type index includes a market clearing price.
4. The Lasso-logit model-based power generation unit market force abuse identification method of claim 1, wherein:
step S3 specifically includes the following steps:
s31, constructing a sample set by adopting a hierarchical sampling method, and avoiding model identification errors caused by data imbalance;
s32, selecting based on Lasso variables, obtaining mean square errors under a plurality of different penalty coefficients by adopting a cross validation method, and screening market abuse identification indexes of the generator set by selecting proper penalty coefficients;
and S33, further adopting the co-linearity test to verify the co-linearity among the indexes, and providing a basis for constructing the identification model.
5. The Lasso-logit model-based power generation unit market force abuse identification method of claim 1, wherein:
step S4 specifically includes the following steps:
s41, constructing and estimating a Lasso-logit model for the abuse identification of the market force of the generator set according to the screening result and the generator set sample set;
s42, the significance of the model indexes is checked, and whether the indexes in the model can play a good explaining role in the abuse identification of the market force of the generator set is judged.
6. The Lasso-logit model-based power generation unit market force abuse identification method of claim 1, wherein:
step S5 specifically includes the following steps:
s51, carrying out market force abuse identification analysis on the generator set in the electric power spot market according to the Lasso-logit model;
s52, analyzing the marginal contribution and influence of each index on market abuse and the influence of market abuse probability, analyzing whether the market abuse probability accords with the actual electric power market trading situation, and correspondingly providing a suppression measure.
7. The Lasso-logit model-based power generation unit market force abuse identification method of claim 1, wherein: step S6 specifically includes the following steps:
s61, calculating the average value and standard deviation of the total accuracy, the I-type errors and the II-type errors based on the two types of errors, and verifying the identification performance and stability of the model;
and S62, based on the ROC curve, calculating a corresponding AUC value, and verifying the effectiveness of the model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011369681.0A CN112365187A (en) | 2020-11-30 | 2020-11-30 | Generating set market force abuse identification method based on Lasso-logit model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011369681.0A CN112365187A (en) | 2020-11-30 | 2020-11-30 | Generating set market force abuse identification method based on Lasso-logit model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112365187A true CN112365187A (en) | 2021-02-12 |
Family
ID=74536465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011369681.0A Pending CN112365187A (en) | 2020-11-30 | 2020-11-30 | Generating set market force abuse identification method based on Lasso-logit model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112365187A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112950046A (en) * | 2021-03-18 | 2021-06-11 | 广东电网有限责任公司电力调度控制中心 | Identification method for key units and units in electric power market |
| CN113077165A (en) * | 2021-04-15 | 2021-07-06 | 广东电力交易中心有限责任公司 | Method for judging market force abuse of generator set |
-
2020
- 2020-11-30 CN CN202011369681.0A patent/CN112365187A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112950046A (en) * | 2021-03-18 | 2021-06-11 | 广东电网有限责任公司电力调度控制中心 | Identification method for key units and units in electric power market |
| CN113077165A (en) * | 2021-04-15 | 2021-07-06 | 广东电力交易中心有限责任公司 | Method for judging market force abuse of generator set |
| CN113077165B (en) * | 2021-04-15 | 2024-03-26 | 广东电力交易中心有限责任公司 | Generator set market force abuse discrimination method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Angelos et al. | Detection and identification of abnormalities in customer consumptions in power distribution systems | |
| Spathis | Audit qualification, firm litigation, and financial information: an empirical analysis in Greece | |
| Geiger et al. | Going‐concern opinions in the “new” legal environment | |
| CN103366123B (en) | Software hazard appraisal procedure based on defect analysis | |
| CN106709818A (en) | Power consumption enterprise credit risk evaluation method | |
| CN114065223A (en) | Multi-dimensional software security risk assessment method based on CVSS | |
| CN111695830A (en) | Power resource allocation method, system and equipment | |
| CN112365187A (en) | Generating set market force abuse identification method based on Lasso-logit model | |
| CN111242424B (en) | Quality data processing method and device | |
| CN111062500B (en) | Power equipment evaluation method based on discrete fuzzy number and analytic hierarchy process | |
| Ioana | STUDY REGARDING THE IMPACT OF THE AUDIT COMMITTEE CHARACTERISTICS ON COMPANY PERFORMANCE. | |
| CN106530139A (en) | Method for calculating the index parameter of grid investment analysis model | |
| CN106295858A (en) | A kind of electric energy meter non-health degree Forecasting Methodology | |
| CN112001644A (en) | Power distribution network operation reliability detection method, device, terminal and storage medium | |
| CN106056274A (en) | Power construction main body benefit analysis method based on PCA-DEA two-dimensional comprehensive evaluation model | |
| CN110782163A (en) | Enterprise data processing method and device | |
| Curti et al. | Benchmarking operational risk stress testing models | |
| CN111143763A (en) | Method and device for evaluating state of power equipment and storage medium thereof | |
| CN115829334A (en) | Risk assessment method and system for power grid service | |
| Abbas et al. | Financial Factors and Mandatory Disclosures | |
| Barrett | Detecting bias in jury selection | |
| CN113269578A (en) | Improved support vector machine-based violation identification method for abuse market force of power generation enterprise | |
| Talwar | Dynamics of Firm Value, Financial Performance, Leverage, and Governance: A Panel Data Analysis of Listed Indian Firms | |
| Li et al. | Research on Evaluation of Power Supply Companies External Service Quality Based on improved grey Interrelated Analysis Method. | |
| CN112613696B (en) | Vendor processing method, device, storage medium and processor |
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 |