Early warning method and system for transaction result abnormity of electric power spot market, medium and processor

Abstract

The invention relates to the technical field of transaction abnormity prediction, and discloses an early warning method and system for transaction result abnormity of an electric power spot market, a medium and a processor, and the method comprises the steps: obtaining a plurality of typical operation modes in historical operation modes of a power grid and corresponding characteristic indexes; obtaining a prediction attribute scalar of a power grid operation day, and calculating a characteristic index of the operation day; carrying out matching analysis on the corresponding feature index under each typical operation mode and the feature index of the operation day, and screening a first transaction result abnormal set meeting a matching condition under the typical operation mode; outputting a second transaction result exception set and an exception probability of the operation day through the trained deep neural network model; and according to a preset screening principle, screening the first transaction result exception set and the second transaction result exception set to obtain a transaction exception early warning result. According to the invention, accurate early warning of power spot transaction clearing result abnormity is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of transaction result prediction, and in particular to an early warning method, system, medium and processor for abnormal transaction results in a power spot market. Background Art

[0002] The electricity spot market is an important part of the electricity market. In the spot market, electricity prices fluctuate according to various factors such as real-time supply and demand, changes in power load, and power generation costs. This volatility and uncertainty makes the transaction results of the electricity spot market vulnerable to various external factors, which may lead to abnormalities.

[0003] In order to deal with the abnormalities in the transaction results of the power spot market, it is particularly important to establish an effective early warning method. The early warning method can help market players to timely discover potential risks and problems, take measures to prevent and respond in advance, and thus avoid or reduce the adverse effects of abnormal transactions on the market.

[0004] In view of this, there is a need for an early warning method, system, medium and processor for abnormal transaction results in the electricity spot market. Summary of the invention

[0005] In view of the need to deal with the abnormal problems that may occur in the transaction results of the power spot market in the prior art, the present invention provides an early warning method, system, medium and processor for abnormal transaction results in the power spot market, which can realize accurate early warning analysis of abnormal clearing results of power spot transactions, timely discover potential risks and problems, and avoid or reduce the adverse effects of abnormal transactions on the market. The specific technical solution is as follows:

[0006] An early warning method for abnormal transaction results in a power spot market, comprising:

[0007] Obtain several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid;

[0008] Obtain the predicted attribute scalar of the power grid operation day and calculate the characteristic index of the operation day;

[0009] Matching and analyzing the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and screening the transaction results that meet the matching conditions under the typical operation mode, and recording them as the first transaction result abnormal set;

[0010] The trained deep neural network model is used to output the abnormal trading results and abnormal probabilities of the operating day, which are recorded as the second abnormal trading result set;

[0011] According to a preset screening principle, the transaction abnormality warning result is obtained by screening the first transaction result abnormality set and the second transaction result abnormality set.

[0012] Preferably, the obtaining of several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid includes:

[0013] Acquire attribute variables characterizing the operation mode of the power grid, and perform standardization processing on the attribute variables;

[0014] The similarity algorithm is used to extract features from the standardized attribute variables, and cluster analysis is performed to extract typical operating modes and corresponding characteristic indicators.

[0015] Preferably, matching and analyzing the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and screening the transaction results that meet the matching conditions under the typical operation mode, which are recorded as the first transaction result abnormal set, includes:

[0016] The characteristic indexes corresponding to each typical operation mode are combined to form a corresponding first characteristic vector;

[0017] Combining the characteristic indicators of the operation day to form a second characteristic vector;

[0018] Calculate the similarity between the first eigenvector and the second eigenvector in each typical operation mode;

[0019] A selected date that meets a preset condition in a typical operation mode is selected according to the similarity, and the transaction result anomalies within the selected date are taken as a first transaction result anomaly set.

[0020] Preferably, the abnormal transaction results and abnormal probability of the operation day output by the trained deep neural network model, recorded as the second abnormal transaction result set, include:

[0021] Obtain historical power grid operation data and transaction data, and pre-process them respectively;

[0022] The processed data is used as sample data to input into the deep neural network model for training, and a trained model is generated that can output the abnormal trading results and abnormal probability of the operating day;

[0023] The prediction data of the operation day is input into the trained deep neural network model to obtain the abnormal transaction results and abnormal probabilities of the operation day, which are recorded as the second transaction result abnormal set.

[0024] Preferably, the abnormal transaction warning result obtained by screening the first abnormal transaction result set and the second abnormal transaction result set according to the preset screening principle includes:

[0025] Screening out the intersection of the first transaction result abnormal set and the second transaction result abnormal set, and selecting the intersection whose abnormal probability is greater than a first preset threshold as a third transaction result abnormal set;

[0026] Filtering out the second transaction result abnormality set whose abnormal probability is greater than a second preset threshold as a fourth transaction result abnormality set, and the second preset threshold is greater than the first preset threshold;

[0027] The third transaction result abnormality set and the fourth transaction result abnormality set are used as transaction abnormality warning results.

[0028] An early warning system for abnormal transaction results in the power spot market, applied to the aforementioned early warning method for abnormal transaction results in the power spot market, comprises:

[0029] A first characteristic index acquisition unit is used to acquire several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid;

[0030] A second characteristic index acquisition unit is used to acquire the predicted attribute scalar of the power grid operation day and calculate the characteristic index of the operation day;

[0031] A first transaction result abnormal set generating unit is used to match and analyze the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and select the transaction results that meet the matching conditions under the typical operation mode, and record them as the first transaction result abnormal set;

[0032] A second transaction result anomaly set generating unit, configured to output the transaction result anomalies and anomaly probabilities of the operation day through the trained deep neural network model, which are recorded as the second transaction result anomaly set;

[0033] The warning result generating unit is used to filter the first transaction result abnormality set and the second transaction result abnormality set to obtain a transaction abnormality warning result according to a preset screening principle.

[0034] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to provide the aforementioned early warning method for abnormal transaction results in the electricity spot market.

[0035] A processor is used to run a program, wherein the program executes the aforementioned early warning method for abnormal transaction results in the electricity spot market when running.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention provides an early warning method for abnormal transaction results in the power spot market by obtaining several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid; obtaining the predicted attribute scalar of the power grid operation day, and calculating the characteristic indicators of the operation day; matching and analyzing the corresponding characteristic indicators under each typical operation mode with the characteristic indicators of the operation day, screening the transaction results that meet the matching conditions under the typical operation mode, and recording them as the first abnormal transaction result set; outputting the abnormal transaction results and abnormal probability of the operation day through the trained deep neural network model, and recording them as the second abnormal transaction result set; and screening the abnormal transaction warning results from the first abnormal transaction result set and the second abnormal transaction result set according to the preset screening principle. The present invention aims at abnormal transaction results, and summarizes the abnormal transaction result set of historical characteristic days and the abnormal transaction result probability based on deep neural network mapping through similar day matching, so as to form an abnormal transaction result set, and thus output the early warning results. Not only considering various characteristic quantities of power grid operation, but also combining the artificial intelligence algorithm predicted by the deep neural network model, the probabilistic early warning of the transaction results of the operation day is carried out, and the accurate early warning analysis of the abnormal clearing results of power spot transactions is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0039] Figure 1 The present invention is a flow chart of an early warning method for abnormal transaction results in the electricity spot market.

[0040] Figure 2 The present invention is a flow chart of an embodiment of a method for early warning of abnormal transaction results in the electricity spot market.

[0041] Figure 3 The present invention is a flow chart of another embodiment of a method for early warning of abnormal transaction results in the electricity spot market.

[0042] Figure 4 The present invention is a schematic diagram of an early warning system for abnormal transaction results in the electricity spot market. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0045] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0046] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0047] See the following examples Figures 1 to 4 .

[0048] The embodiment of the present application provides an early warning method for abnormal transaction results in the power spot market, including:

[0049] Step S01, obtaining several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid;

[0050] From the perspective of grid operation feature extraction, attribute variables that characterize the grid operation mode are selected, and after the variables are standardized, a similarity-based algorithm is used to extract feature variables from the attribute variables. Subsequently, the grid operation status at each moment is used as a sample point, and the active power of the unit, line and load is used as an attribute to perform cluster analysis and extract typical operation modes.

[0051] Step S02: Obtain the predicted attribute scalar of the power grid operation day and calculate the characteristic index of the operation day;

[0052] The predicted attribute variables include the system supply-demand ratio at peak times, system load, voltage, frequency and other parameters. Through these predicted parameters, the operating status of the power grid can be calculated and evaluated, and the characteristic indicators of the operating day can be calculated.

[0053] Step S03, matching and analyzing the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and screening the transaction results that meet the matching conditions under the typical operation mode, and recording them as the first transaction result abnormal set;

[0054] By combining the characteristic indicators into a feature vector and calculating the similarity of the two vectors, all sample sets under typical operation modes are selected, and some dates with the highest similarity of typical operation mode samples are selected. The union of the abnormal transaction results on these dates is used as the initial abnormal transaction result set, which is recorded as the first abnormal transaction result set.

[0055] Step S04: Output the abnormal transaction results and abnormal probability of the operation day through the trained deep neural network model, which is recorded as the second abnormal transaction result set;

[0056] For deep neural network training, we should first collect historical power grid operation data, load demand, and unit output to obtain a large amount of sample data; then preprocess the samples and divide the processed samples into sample sets, verification sets, and test sets; then randomly initialize the deep neural network parameters; use the optimization algorithm to update the deep neural network parameters, and stop updating the parameters when the number of iterations exceeds the threshold or the early stopping conditions are met.

[0057] After the deep neural network has been effectively trained, the "mapping relationship between the input feature vector and the output feature vector" is obtained. Based on the day-ahead market clearing boundary conditions, the deep neural network directly maps out the abnormal trading results and their probabilities.

[0058] Specifically, the abnormal transaction results and abnormal probability of the operation day output by the trained deep neural network model are recorded as the second abnormal transaction result set, including:

[0059] Obtain historical power grid operation data and transaction data, and pre-process them respectively;

[0060] The processed data is used as sample data to input into the deep neural network model for training, and a trained model is generated that can output the abnormal trading results and abnormal probability of the operating day;

[0061] The prediction data of the operation day is input into the trained deep neural network model to obtain the abnormal transaction results and abnormal probabilities of the operation day, which are recorded as the second transaction result abnormal set.

[0062] Step S05: According to a preset screening principle, the first transaction result abnormality set and the second transaction result abnormality set are screened to obtain a transaction abnormality warning result.

[0063] The abnormal transaction warning results include abnormal warnings for multi-dimensional transaction results such as abnormal reserve, abnormal electricity price, and abnormal output.

[0064] Specifically, the abnormal transaction warning result obtained by screening the first abnormal transaction result set and the second abnormal transaction result set according to the preset screening principle includes:

[0065] Screening out the intersection of the first transaction result abnormal set and the second transaction result abnormal set, and selecting the intersection whose abnormal probability is greater than a first preset threshold (set to 50%) as the third transaction result abnormal set;

[0066] Screening out the second transaction result abnormality set whose abnormal probability is greater than a second preset threshold (set to 90%) as a fourth transaction result abnormality set, and the second preset threshold is greater than the first preset threshold;

[0067] The third transaction result abnormality set and the fourth transaction result abnormality set are used as transaction abnormality warning results.

[0068] The present invention provides an early warning method for abnormal transaction results in the power spot market by obtaining several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid; obtaining the predicted attribute scalar of the power grid operation day, and calculating the characteristic indicators of the operation day; matching and analyzing the corresponding characteristic indicators under each typical operation mode with the characteristic indicators of the operation day, screening the transaction results that meet the matching conditions under the typical operation mode, and recording them as the first abnormal transaction result set; outputting the abnormal transaction results and abnormal probability of the operation day through the trained deep neural network model, and recording them as the second abnormal transaction result set; according to the preset screening principle, screening the abnormal transaction warning results from the first abnormal transaction result set and the second abnormal transaction result set. The present invention takes into account various characteristic quantities of power grid operation, and according to the similarity between the operation day and the historical date, combined with the artificial intelligence algorithm predicted by the deep neural network model, carries out probabilistic early warning of the transaction results of the operation day, realizes the early warning analysis of abnormal clearing results of power spot transactions, and includes multi-dimensional abnormal transaction result warning.

[0069] Specifically, in a preferred implementation of the present application, the obtaining of several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid includes:

[0070] Step S11, obtaining attribute variables characterizing the operation mode of the power grid, and standardizing the attribute variables;

[0071] Step S12: Use a similarity algorithm to extract features from the standardized attribute variables, and perform cluster analysis to extract typical operating modes and corresponding feature indicators.

[0072] In this embodiment, a similarity-based algorithm is used to extract characteristic variables from attribute variables, and then the grid operation status at each moment is used as a sample point, and the active power of the unit, line and load is used as an attribute to perform cluster analysis and extract typical operation modes.

[0073] In the specific implementation, cluster analysis is performed on the attribute variables of the selected power grid operation mode. After determining the initial cluster center, the cluster center is corrected and iterated according to the time period samples, and finally a sample set representing m types of typical operation modes and their corresponding operation modes is output. Among them, m represents the number of clusters determined by the clustering algorithm (that is, the number of types of typical power grid operation modes). Based on the cluster analysis, typical operation modes and their corresponding characteristic indicators can be extracted. These characteristic indicators can reflect the key characteristics of the power grid under different typical operation modes.

[0074] Specifically, in a preferred implementation of the present application, the characteristic indicators corresponding to each typical operation mode are matched and analyzed with the characteristic indicators of the operation day, and the transaction results that meet the matching conditions under the typical operation mode are screened, which are recorded as the first transaction result abnormal set, including:

[0075] Step S31: Combining the corresponding characteristic indexes under each typical operation mode to form a corresponding first characteristic vector W i ;

[0076] Step S32: Combining the characteristic indexes of the operation day to form a second characteristic vector W j ;

[0077] Step S33: Calculate the first eigenvector W under each typical operation mode i With the second eigenvector W j similarity;

[0078] Step S34: selecting a selected date that meets a preset condition in a typical operation mode according to the similarity, and taking the transaction result anomalies within the selected date as a first transaction result anomaly set.

[0079] For example, assuming that under the i-th typical operation mode, the characteristic vectors of the output of the hydropower unit, wind power and photovoltaic unit are W hi , W wi , W si ; The characteristic vector representing the commissioning and decommissioning of the line is W li , the characteristic vector of the main transformer commissioning and decommissioning is W ti ; The characteristic vector of the partition load is W ldi The characteristic vector of the sample power grid operation mode is expressed as W i =W hi , W wi , W si , Wli , W ti , W ldi ). Similarly, the feature vector of the next day's operation mode can be expressed as W j , based on the Euclidean distance calculation W i With W j , to determine whether it belongs to the i-th typical operation mode.

[0080] Consider the actual (or predicted) output of hydropower, wind turbines, and photovoltaic power stations as the power grid operation attribute variables. The attribute variable selection object of hydropower units is based on the standard of a single installed capacity of 300MW. The number of hydropower units is N. w , wind power and photovoltaic power stations can use the actual output value (or predicted value) of the entire network as a unified standard. The number of attribute variables on the power generation side is N w +2.

[0081] The commissioning / decommissioning of lines and main transformers at 220 kV voltage level (or above) is used as the basis for the power grid operation attribute variables. In order to reflect the difference in the importance of components on power grid operation, the historical average transmission capacity of each component is used as the weight to modify the attributes. l , N t They are the number of main network lines and main transformers respectively.

[0082] The actual value or predicted value of the load in the sub-area (or the whole area) above 220kV is used as one of the selection criteria for the power grid operation attributes. The number of attribute variables on the load side is the same as the number of main grid sub-areas (or the whole area), which is N. m .

[0083] Traverse all sample sets under the i-th typical operation mode, select some dates with the highest similarity to the typical mode samples, and use the union of the abnormal transaction results on these dates as the initial abnormal transaction result set.

[0084] Since the abnormal transaction warning results include abnormal reserve, abnormal electricity price, abnormal output and other multi-dimensional abnormal transaction result warnings, the abnormal electricity price is used as an example for explanation. The details are as follows:

[0085] Step 1: Use K-Mediods clustering algorithm to complete the clustering of power grid operation characteristics. According to the power grid operation mode and node load on the operation day, calculate the value of the power grid operation characteristic index on the operation day.

[0086] Step 2: Use Mahalanobis distance as a similarity measurement method to compare the characteristic indicators of the operating day with those of each typical class. After determining that the operating day belongs to a typical class, find the N samples with the highest similarity in the historical operating days of that class to form the initial screening set Ω1 for electricity price anomalies.

[0087] Step 3: Based on the forecast data of the operation day (system supply-demand ratio at peak time, system load) and the electricity price statistics of the previous day, the deep neural network is trained and the parameters of the deep neural network are updated using the SGD optimization algorithm.

[0088] The operating day forecast data includes forecast data, plan data and constraint data. The forecast data is predicted by various professionals in the power system based on historical data and power grid planning; the plan data is formulated by the Power Grid Mode Department and the Operation Policy Department based on the actual operation needs and planning of the power grid; the constraint data is formulated by the Spot Market Department based on the actual operation status of the power grid. All data can be obtained from the spot market clearing system.

[0089] Step 4: Use the trained deep neural network model (the model represents the mapping relationship between the system supply-demand ratio at peak hours, system load, electricity price statistics of the previous day and the electricity price during operation) to estimate the electricity price on the operating day, and based on this, obtain the electricity price anomaly set Ω2 and the probability of electricity price anomaly.

[0090] Step 5: Preliminary screening of the intersection of set Ω1 and set Ω2, the electricity price anomaly set Ω3 with an anomaly probability greater than 50%, and the anomaly set Ω4 with an anomaly probability greater than 80% in set Ω2, and {Ω3, Ω4}∈{Ω2}.

[0091] Step 6: Output the electricity price warning result {Ω3, Ω4}, including the results of the initial and fine screening of electricity price anomalies, as well as the limit and anomaly probability.

[0092] The present invention provides an early warning system for abnormal transaction results in the power spot market, which is applied to the above-mentioned early warning method for abnormal transaction results in the power spot market, and comprises:

[0093] A first characteristic index acquisition unit is used to acquire several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid;

[0094] A second characteristic index acquisition unit is used to acquire the predicted attribute scalar of the power grid operation day and calculate the characteristic index of the operation day;

[0095] A first transaction result abnormal set generating unit is used to match and analyze the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and select the transaction results that meet the matching conditions under the typical operation mode, and record them as the first transaction result abnormal set;

[0096] A second transaction result anomaly set generating unit, configured to output the transaction result anomalies and anomaly probabilities of the operation day through the trained deep neural network model, which are recorded as the second transaction result anomaly set;

[0097] The warning result generating unit is used to filter the first transaction result abnormality set and the second transaction result abnormality set to obtain a transaction abnormality warning result according to a preset screening principle.

[0098] The functional explanation of each unit of this embodiment is the same as the explanation of an embodiment of a method for abnormal transaction results in the electricity spot market. At the same time, the technical effect of this embodiment is the same as the technical effect of an embodiment of a method for abnormal transaction results in the electricity spot market, and will not be repeated here.

[0099] The present application also provides a computer-readable storage medium, which includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to provide the aforementioned early warning method for abnormal transaction results in the electricity spot market.

[0100] The technical effect of this embodiment is the same as the technical effect of an embodiment of a method for abnormal transaction results in the electricity spot market, and will not be repeated here.

[0101] The present invention can be used in many general or special computer system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like.

[0102] The present application also provides a processor, which is used to run a program, wherein the program executes the aforementioned early warning method for abnormal transaction results in the electricity spot market when running.

[0103] The technical effect of this embodiment is the same as the technical effect of an embodiment of a method for abnormal transaction results in the electricity spot market, and will not be repeated here.

[0104] The processor in this embodiment may be a central processing unit (CPU), a controller, a microcontroller, or other data processing chips.

[0105] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0106] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0107] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0108] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for early warning of abnormal transaction results in the electricity spot market, characterized in that: include: Obtain several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid; Obtain the predicted attribute scalar of the power grid operation day and calculate the characteristic index of the operation day; Matching and analyzing the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and screening the transaction results that meet the matching conditions under the typical operation mode, and recording them as the first transaction result abnormal set; The trained deep neural network model is used to output the abnormal trading results and abnormal probabilities of the operating day, which are recorded as the second abnormal trading result set; According to a preset screening principle, the transaction abnormality warning result is obtained by screening the first transaction result abnormality set and the second transaction result abnormality set.

2. The method for early warning of abnormal transaction results in the electricity spot market according to claim 1, characterized in that: The acquisition of several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid includes: Acquire attribute variables characterizing the operation mode of the power grid, and perform standardization processing on the attribute variables; The similarity algorithm is used to extract features from the standardized attribute variables, and cluster analysis is performed to extract typical operating modes and corresponding characteristic indicators.

3. The method for early warning of abnormal transaction results in the electricity spot market according to claim 1, characterized in that: The matching analysis of the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and the screening of the transaction results that meet the matching conditions under the typical operation mode, recorded as the first transaction result abnormal set, includes: The characteristic indexes corresponding to each typical operation mode are combined to form a corresponding first characteristic vector; Combining the characteristic indicators of the operation day to form a second characteristic vector; Calculate the similarity between the first eigenvector and the second eigenvector in each typical operation mode; A selected date that meets a preset condition in a typical operation mode is selected according to the similarity, and the transaction result anomalies within the selected date are taken as a first transaction result anomaly set.

4. The method for early warning of abnormal transaction results in the electricity spot market according to claim 1, characterized in that: The abnormal trading results and abnormal probability of the operating day output by the trained deep neural network model are recorded as the second abnormal trading result set, which includes: Obtain historical power grid operation data and transaction data, and pre-process them respectively; The processed data is used as sample data to input into the deep neural network model for training, and a trained model is generated that can output the abnormal trading results and abnormal probability of the operating day; The prediction data of the operation day is input into the trained deep neural network model to obtain the abnormal transaction results and abnormal probabilities of the operation day, which are recorded as the second transaction result abnormal set.

5. The method for early warning of abnormal transaction results in the electricity spot market according to claim 1, characterized in that: The abnormal transaction warning result obtained by screening the first abnormal transaction result set and the second abnormal transaction result set according to the preset screening principle includes: Screening out the intersection of the first transaction result abnormal set and the second transaction result abnormal set, and selecting the intersection whose abnormal probability is greater than a first preset threshold as a third transaction result abnormal set; Filtering out the second transaction result abnormality set whose abnormal probability is greater than a second preset threshold as a fourth transaction result abnormality set, and the second preset threshold is greater than the first preset threshold; The third transaction result abnormality set and the fourth transaction result abnormality set are used as transaction abnormality warning results.

6. An early warning system for abnormal trading results in the electricity spot market, characterized in that: The early warning method for abnormal transaction results in the power spot market as described in any one of claims 1 to 5 comprises: A first characteristic index acquisition unit is used to acquire several typical operation modes and corresponding characteristic indicators in the historical operation modes of the power grid; A second characteristic index acquisition unit is used to acquire the predicted attribute scalar of the power grid operation day and calculate the characteristic index of the operation day; A first transaction result abnormal set generating unit is used to match and analyze the characteristic indicators corresponding to each typical operation mode with the characteristic indicators of the operation day, and select the transaction results that meet the matching conditions under the typical operation mode, and record them as the first transaction result abnormal set; A second transaction result anomaly set generating unit, configured to output the transaction result anomalies and anomaly probabilities of the operation day through the trained deep neural network model, which are recorded as the second transaction result anomaly set; The warning result generating unit is used to filter the first transaction result abnormality set and the second transaction result abnormality set to obtain a transaction abnormality warning result according to a preset screening principle.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the early warning method for abnormal transaction results in the electricity spot market as described in any one of claims 1 to 5.

8. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the early warning method for abnormal transaction results in the electricity spot market as described in any one of claims 1 to 5.

Images