Intelligent power grid transaction abnormity early warning method and device based on real-time data

Abstract

The invention relates to the technical field of power grid transactions, in particular to an intelligent power grid transaction abnormity early warning method and device based on real-time data. The method comprises the following steps: acquiring real-time transaction data of smart grid transaction, and at least one group of historical transaction data and at least one group of predicted transaction data corresponding to the real-time transaction data; performing serialized feature alignment on the historical transaction data, the real-time transaction data and the predicted transaction data according to the time sequence features to obtain a power grid transaction detection sequence; carrying out fluctuation evaluation on the power grid transaction detection sequence to obtain a power grid transaction fluctuation index; the preset power grid transaction fluctuation threshold value is used for carrying out abnormity judgment on the power grid transaction fluctuation index to obtain the intelligent power grid transaction abnormity result, the transaction behavior can be analyzed more comprehensively, and the potential risk can be early warned in time.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid transactions, and in particular to a method and device for warning abnormality of smart grid transactions based on real-time data. Background Art

[0002] With the rapid development of modern power systems, smart grids, as a new type of power network integrating advanced technologies, are gradually being widely used around the world; smart grids not only improve the efficiency and reliability of power systems, but also promote the access to renewable energy and enhance user participation; however, in such a highly complex and dynamically changing environment, the stability and security of the power trading market become particularly important; the power trading market involves multi-party interactions among power generation, transmission, distribution and end users, and any abnormal behavior of any party may have an impact on the entire system.

[0003] However, the existing smart grid transaction anomaly warning methods mainly rely on static rules and post-audits, ignoring the changes in transaction behavior in time series, resulting in limited accuracy of warning results; lack of effective time series analysis means makes it difficult to capture the dynamic change characteristics of transaction data, thus affecting the timeliness and sensitivity of anomaly warnings.

[0004] Therefore, there is an urgent need to provide a smart grid transaction abnormality warning method and device based on real-time data to solve the above technical problems. Summary of the invention

[0005] The embodiments of the present invention provide a smart grid transaction abnormality warning method and device based on real-time data, which can more comprehensively analyze transaction behaviors and timely warn of potential risks.

[0006] In a first aspect, an embodiment of the present invention provides a smart grid transaction abnormality early warning method based on real-time data, comprising: Acquire real-time transaction data of smart grid transactions, and at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data; According to the time series characteristics, the historical transaction data, the real-time transaction data and the predicted transaction data are serialized and aligned to obtain a power grid transaction detection sequence; Performing volatility assessment on the power grid transaction detection sequence to obtain a power grid transaction volatility index; The preset power grid transaction fluctuation threshold is used to perform abnormal judgment on the power grid transaction fluctuation index to obtain the abnormal result of smart grid transaction.

[0007] In a second aspect, the present application also provides a smart grid transaction abnormality warning device based on real-time data, the device comprising: A data acquisition module, used to acquire real-time transaction data of smart grid transactions, and at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data; A serialization processing module, used for performing serialization feature alignment on the historical transaction data, the real-time transaction data and the predicted transaction data according to time series features to obtain a power grid transaction detection sequence; A volatility assessment module, used for performing a volatility assessment on the power grid transaction detection sequence to obtain a power grid transaction volatility index; The abnormality judgment module is used to use a preset power grid transaction fluctuation threshold to perform abnormality judgment on the power grid transaction fluctuation index to obtain a smart grid transaction abnormality result.

[0008] In a third aspect, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any one of the above methods.

[0009] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.

[0010] The embodiments of the present invention provide a method, device, electronic device and storage medium for warning abnormalities in smart grid transactions based on real-time data. By integrating historical, real-time and predicted transaction data and aligning time series features, the dynamic change characteristics of transaction data can be effectively captured, and the accuracy of warnings can be significantly improved. At the same time, the volatility assessment is used to obtain the power grid transaction volatility index, and abnormal judgment is made in combination with a preset volatility threshold, which not only enhances the timeliness of the warning, but also improves its sensitivity to abnormal transactions. Compared with traditional methods that rely on static rules and post-audit, this method can more comprehensively analyze transaction behaviors, timely warn of potential risks, and provide strong guarantees for the stability and security of the power trading market. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 is a flow chart of a smart grid transaction abnormality early warning method based on real-time data provided by an embodiment of the present invention; Figure 2 is a hardware architecture diagram of an electronic device provided by an embodiment of the present invention; Figure 3 It is a structural diagram of a smart grid transaction abnormality warning device based on real-time data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0014] Please refer to Figure 1 The embodiment of the present invention provides a smart grid transaction abnormality early warning method based on real-time data, the method comprising: Step 100: acquiring real-time transaction data of smart grid transactions, and at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data; Step 200: align the historical transaction data, the real-time transaction data and the predicted transaction data according to the time series characteristics to obtain a power grid transaction detection sequence; Step 300: performing a volatility assessment on the power grid transaction detection sequence to obtain a power grid transaction volatility index; Step 400: Use a preset power grid transaction fluctuation threshold to perform abnormality judgment on the power grid transaction fluctuation index to obtain abnormal results of smart grid transactions.

[0015] In this implementation, by integrating historical, real-time and predicted transaction data and aligning time series features, the dynamic change characteristics of transaction data are effectively captured, and the accuracy of early warning is significantly improved; at the same time, the volatility assessment is used to obtain the power grid transaction volatility index, and the preset volatility threshold is combined for abnormal judgment, which not only enhances the timeliness of the early warning, but also improves its sensitivity to abnormal transactions; compared with the traditional method that relies on static rules and post-audit, this method can more comprehensively analyze transaction behavior, timely warn of potential risks, and provide strong guarantees for the stability and security of the power trading market.

[0016] In one embodiment of the present invention, real-time transaction data includes transaction time (specific time when the transaction occurs, usually recorded in the form of a timestamp), transaction volume (amount of electricity traded, usually in megawatt-hours) and transaction price (price of electricity traded, usually in yuan / megawatt-hour); its acquisition process needs to rely on the transaction monitoring system, data acquisition equipment and communication network of the smart grid, and the power trading platform in the smart grid is the main source of real-time transaction data; the trading platform records information such as the time, transaction volume and transaction price of each transaction; smart meters and sensors are deployed on the power generation side, transmission side and user side to collect relevant data on power transactions in real time; smart meters can record users' electricity consumption and transaction prices, and sensors can monitor the operating status of the power grid; power transaction data is collected in real time through Internet of Things devices, and the data is transmitted to the central data processing platform; the received data is preliminarily cleaned and formatted, invalid or erroneous records are removed, and the timestamp is adjusted to a unified format; key features such as transaction time, transaction volume and transaction price are extracted from the original data; By combining multiple data collection methods, we can ensure the real-time nature of data and reduce the time difference from data generation to system utilization; by cleaning, standardizing and timestamp calibrating data, we can improve data quality and enhance the reliability of analysis results; by integrating data sources from different channels, we can form a comprehensive and detailed real-time transaction data set, which will help to more fully understand the dynamic changes in the electricity market.

[0017] Specifically, the method of obtaining at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data in step S100 includes: Step 110: Obtain a transaction environment feature set corresponding to the real-time transaction data of smart grid transactions; the transaction environment feature set is a multi-dimensional feature set that describes the smart grid transaction environment and can fully reflect various environmental factors that affect power transactions; specifically, it includes but is not limited to the following subsets: a. A subset of power supply characteristics, which includes: The proportion of power generation types is used to indicate the proportion of each type of power generation (such as thermal power, hydropower, wind power, photovoltaic power, etc.) in the total power generation; Power generation capacity, which is used to indicate the real-time power generation capacity of various power generation methods; Available capacity, which is used to represent the generation capacity available for dispatch in the grid; b. Electricity demand characteristic subset, which includes: Time period characteristics, used to indicate the specific time period when the transaction occurs (such as peak, flat, and trough); Seasonal characteristics are used to indicate the season in which the transaction occurs (such as summer and winter), reflecting seasonal changes in electricity demand; c. Market policy feature subset, which includes: The number of market entities participating in power transactions is used to indicate the number of power generation enterprises, distribution companies, users and other entities participating in the transaction; Market share distribution, used to indicate the share of each market player in the electricity trading market; Policy subsidy, used to indicate the subsidy policy for renewable energy power generation in the region; Policy restrictions, which are used to indicate the impact of policies such as carbon emission restrictions and power generation quotas on electricity trading; d. Meteorological environment feature subset, which includes: Temperature, which is used to indicate the real-time temperature and its impact on electricity demand; Humidity, which is used to indicate real-time humidity and its impact on the operation of power equipment; Light intensity, used to indicate real-time light intensity and its impact on photovoltaic power generation; Air pressure, used to indicate real-time air pressure and its impact on wind power generation.

[0018] In step 110, the transaction environment feature set covers multi-dimensional features such as power supply, demand, market policy and meteorological environment, which can comprehensively reflect various environmental factors affecting power trading and significantly improve the accuracy and reliability of anomaly detection; by real-time collection and extraction of power supply features such as power generation type ratio, power generation capacity, available capacity, and power demand features such as time period characteristics and seasonal characteristics, it can dynamically capture changes in power supply and demand and provide real-time data support for abnormal warning; in addition, the introduction of market policy feature subsets and meteorological environment feature subsets further enhances the environmental adaptability of the early warning method and can effectively respond to the impact of policy changes and meteorological conditions on power trading; the multi-dimensional design of the transaction environment feature set provides comprehensive input for subsequent historical data screening and prediction models, supports multi-dimensional data fusion and dynamic analysis, and significantly improves the timeliness, sensitivity and foresight of abnormal warnings.

[0019] Step 120: Using the transaction environment feature set as a screening condition, search in the smart grid transaction database to obtain at least one set of historical transaction data; the at least one set of historical transaction data includes transaction data that is closest to the real-time transaction data in time series and matches the transaction environment feature set; the core goal of step S120 is to screen out historical transaction data similar to the current transaction environment through the transaction environment feature set, so as to provide a reference for subsequent anomaly detection; the specific implementation is as follows: Step 121, in the smart grid transaction database, using the transaction environment feature set as a screening condition, querying historical transaction data similar to the current transaction environment; the query conditions include: power supply characteristics such as power generation type ratio, power generation capacity, and available capacity; power demand characteristics such as time period characteristics and seasonal characteristics; market policy characteristics such as the number of market entities, market share distribution, policy subsidies, and policy restrictions; meteorological environment characteristics such as temperature, humidity, light intensity, and air pressure; the specific screening method can adopt similarity matching, and calculate the feature similarity of historical transaction data and real-time transaction data through methods such as Euclidean distance and cosine similarity; set a similarity threshold to filter out historical transaction data with a similarity higher than the threshold; Step 122: Perform time series matching on the historical transaction data selected by the screening conditions, and the data should be closest to the real-time transaction data in terms of time series to ensure the timeliness and relevance of the data; for example, if the real-time transaction data occurs during the summer peak period, the historical transaction data during the summer peak period is preferentially screened; Step 123, the result obtained by time series matching includes at least one set of historical transaction data, which is closest to the real-time transaction data in time series and highly matches the transaction environment feature set; each set of historical transaction data includes information such as transaction time, transaction volume, and transaction price.

[0020] In step 120, historical transaction data similar to the current transaction environment is screened out through the transaction environment feature set to ensure the representativeness and reliability of the data; the screened historical transaction data is closest to the real-time transaction data in terms of time series, ensuring the timeliness and relevance of the data; the screened historical transaction data provides a reliable reference for subsequent serialization feature alignment and anomaly detection, significantly improving the accuracy and sensitivity of anomaly detection.

[0021] Step 130: input the transaction environment feature set as input into a preset smart grid transaction prediction model to obtain at least one set of predicted transaction data; the at least one set of predicted transaction data includes transaction data that is closest to the real-time transaction data in time sequence and matches the transaction environment feature set in the future transaction process of the smart grid; Specifically, a machine learning model (such as LSTM, GRU, random forest, etc.) or a deep learning model (such as Transformer, CNN, etc.) is used as a smart grid transaction prediction model, which is trained by historical transaction data and a transaction environment feature set, and can predict future transaction data that is highly consistent with the transaction environment feature set; the input of the smart grid transaction prediction model is the transaction environment feature set, and the input data needs to be standardized before training to ensure that the dimensions of different features are consistent; The output of the smart grid transaction prediction model is at least one set of predicted transaction data, including: transaction time, i.e., the specific time when future transactions occur; transaction volume, i.e., the electricity transaction volume of future transactions; transaction price, i.e., the electricity transaction price of future transactions; it should be noted that the generated predicted transaction data must be closest to the real-time transaction data in terms of time series to ensure the timeliness and relevance of the data; for example, if the real-time transaction data occurs at the current time, the transaction data for the next hour, day or week is predicted; In the process of constructing the smart grid transaction prediction model, historical transaction data and transaction environment feature sets are used to train the smart grid transaction prediction model and optimize model parameters; cross-validation, grid search and other methods are used to improve the prediction accuracy of the model; the transaction environment feature set is input into the trained model to generate predicted transaction data; real-time prediction is achieved through the model reasoning interface (such as RESTful API); and the generated predicted transaction data is stored in a temporary database.

[0022] In step 130, future transaction data is generated through the smart grid transaction prediction model to provide forward-looking support for anomaly detection and significantly improve the timeliness and sensitivity of early warning. The transaction environment feature set is used as input to drive the model to generate predicted transaction data that is highly matched with the current transaction environment, ensuring the accuracy and reliability of the prediction. The prediction model can respond to changes in transaction environment characteristics in real time, dynamically generate predicted transaction data, and support the dynamic adaptability and robustness of anomaly detection.

[0023] In one embodiment of the present invention, the smart grid transaction database is used to store, manage and analyze smart grid transaction data, aiming to support real-time monitoring, historical data analysis and future trend prediction of the power transaction market; adopts a relational database (such as MySQL, PostgreSQL) or a time series database (such as InfluxDB, TimescaleDB) to support efficient data storage and query; provides a RESTful API or GraphQL interface to support external system access and operation of the database; implements data encryption, access control and audit logs to ensure data security and privacy; specifically includes the following functions: Data storage: storing real-time transaction data, historical transaction data, and predicted transaction data of smart grid transactions, and storing transaction environment characteristic data, including power supply characteristics, power demand characteristics, market policy characteristics, and meteorological environment characteristics; Data management: Provides data addition, deletion, modification and query functions, supports dynamic data update and maintenance, implements data backup and recovery, and ensures data security and reliability; Data analysis: supports statistical analysis, time series analysis, and volatility analysis of transaction data, and provides data visualization functions to facilitate users to intuitively understand data characteristics and trends; Data query: supports complex query operations such as multi-condition screening, time series matching and similarity calculation, provides efficient query interface, and supports real-time data retrieval and analysis.

[0024] The smart grid transaction database provides data support for smart grid transaction abnormality warning, significantly improving the timeliness, accuracy and adaptability of the warning method. Through efficient data storage, management and analysis, the database can fully capture the dynamic changes in the power trading market and provide guarantees for the stable operation of the power system.

[0025] In one embodiment of the present invention, a coherent power grid transaction detection sequence is constructed by performing serialized feature alignment on key features (transaction time, transaction volume, and transaction price) in historical transaction data, real-time transaction data, and predicted transaction data. The specific implementation is as follows: Extract the three key features of transaction time, transaction volume and transaction price from historical transaction data, real-time transaction data and predicted transaction data respectively; for example, each record in historical transaction data contains multiple fields, but only these three features directly related to the transaction are selected for subsequent processing; ensure the consistency of the formats of the three types of data; for transaction time, uniformly convert it into a specific timestamp format (such as UNIX timestamp); transaction volume and transaction price must also ensure the same data type, for example, transaction volume is numerical data, and the unit is consistent (such as megawatt-hour), and transaction price is uniformly RMB / MWh; Arrange the transaction time, transaction volume and transaction price in the historical transaction data in ascending or descending order according to the transaction time; for example, if the recent trend is to be studied, the transaction time may be sorted from recent to distant; find the position closest to the transaction time of the real-time transaction data in the sorted historical transaction data; this is achieved by traversing the timestamps of the historical transaction data and comparing them with the real-time transaction time; similarly sort the predicted transaction data according to the transaction time, and ensure that the order of its transaction time matches that of the historical transaction data and the real-time transaction data for subsequent alignment operations; Arrange the transaction time, transaction volume and transaction price of several records (determined according to the set time series range) in the historical transaction data that are close to the real-time transaction time in sequence, then insert the three characteristic values ​​of the real-time transaction data, and then arrange the transaction time, transaction volume and transaction price in the predicted transaction data that are close to the real-time transaction time to form a complete power grid transaction detection sequence; for example, if the three records closest to the real-time transaction time in the historical transaction data are (t1, q1, p1), (t2, q2, p2), and (t3, q3, p3), the real-time transaction data is (t0, q0, p0), and the record closest to the real-time transaction time in the predicted transaction data is (t4, q4, p4), then the constructed detection sequence is (t1, q1, p1), (t2, q2, p2), (t3, q3, p3), (t0, q0, p0), and (t4, q4, p4).

[0026] In step 200, by accurately extracting the three key features of transaction time, transaction volume and transaction price from a variety of transaction data and unifying the format, the standardization and comparability of the data are guaranteed, laying a solid foundation for subsequent processing; sorting and alignment operations are performed according to transaction time, so as to construct a coherent power grid transaction detection sequence, which helps to accurately reflect the changing trend of transaction data over time, and is more intuitive for analyzing the evolution of historical transactions and predicting the direction of transactions; when constructing the detection sequence, historical, real-time and predicted data are reasonably incorporated so that the sequence contains sufficient information to comprehensively evaluate the fluctuation of power grid transactions.

[0027] In one embodiment of the present invention, in order to evaluate the volatility of the power grid transaction detection sequence, the following method is used to calculate the power grid transaction volatility index; Step 310: In the power grid transaction detection sequence, the range of transaction volume and transaction price is calculated respectively. For example, for the sequence Q corresponding to the transaction volume, Q={q1,q2,⋯,q n}, the transaction volume is extremely poor R Q =max(Q)−min(Q), the larger the range, the larger the fluctuation range of the data; for the sequence P={p1,p2,⋯,p n}; Similarly, the transaction price range R P =max(P)−min(P), which intuitively shows the maximum fluctuation range of the transaction price in the sequence.

[0028] Step 320: For the transaction prices in the power grid transaction detection sequence, calculate their mean, variance and standard deviation. The transaction price variance represents the average value of the square of the difference between each transaction price and the average transaction price. The transaction price standard deviation represents the average degree to which the transaction price deviates from the average price. The larger the standard deviation, the more drastic the fluctuation of the transaction price. The specific calculation formula is: ; in, represents the standard deviation of transaction prices, represents the number of transaction prices in the power grid transaction detection sequence, represents the transaction price at the i-th time point in the power grid transaction detection sequence, represents the average value of transaction prices in the power grid transaction detection sequence; Similarly, the standard deviation of the transaction volume in the power grid transaction detection sequence is calculated, and the specific calculation formula is: ; in, represents the standard deviation of trading volume, represents the number of transactions in the power grid transaction detection sequence, represents the transaction volume at the i-th time point in the power grid transaction detection sequence, Represents the average value of transaction volume in the power grid transaction detection sequence.

[0029] Step 330: Determine weights for different volatility indicators based on the importance of transaction volume and transaction price in the power grid transaction process to the market. For example, in the electricity market, fluctuations in transaction prices may have a greater impact on market stability, so indicators related to transaction price fluctuations are given relatively high weights; set weight coefficients corresponding to the transaction volume range, transaction price range, transaction price standard deviation, and transaction volume standard deviation.

[0030] Step 340: Based on the weight coefficient determined in step 330, weighted calculation and analysis are performed on the transaction volume range, transaction price range, transaction price standard deviation, and transaction volume standard deviation calculated in steps 310 and 320 to obtain the power grid transaction volatility index; specifically, the calculation formula is: ; in, represents the power grid transaction volatility index; , , and Respectively represent the weight coefficients corresponding to the transaction volume range, transaction price range, transaction price standard deviation and transaction volume standard deviation; Indicates extremely poor trading volume; Indicates that the transaction price is extremely bad; Indicates the standard deviation of transaction prices; represents the standard deviation of trading volume; , , and They represent the standardized functions corresponding to the range of trading volume, the range of trading price, the standard deviation of trading price and the standard deviation of trading volume respectively.

[0031] In the above process of calculating the power grid transaction volatility index, by calculating the range and standard deviation of transaction volume and transaction price, the volatility characteristics of the data can be fully captured; the range reflects the maximum fluctuation range of the data, while the standard deviation measures the average deviation of the data. The combination of the two provides a more comprehensive volatility analysis; different weights are assigned according to the importance of transaction volume and transaction price to the market, so that the volatility index is closer to the actual market demand and the practicality of the index is enhanced; each indicator is processed by a standardized function to ensure that indicators of different dimensions contribute fairly when integrated, and avoid a single indicator dominating the result due to differences in magnitude, thereby improving the accuracy and reliability of the volatility index and helping to improve the stability and security of the power trading market.

[0032] In one embodiment of the present invention, the setting of the power grid transaction fluctuation threshold requires reference to a large amount of historical transaction data to analyze the normal range of the power grid transaction fluctuation index under different market conditions, seasons, time periods, etc.; for example, during the peak period of summer electricity consumption in the past years, the power load fluctuated greatly due to the demand for cooling. By statistically analyzing the distribution of the fluctuation index during this period, the upper and lower limits of the normal fluctuation are determined, so that the early warning mechanism is more in line with the actual situation of the power market and the accuracy of the early warning is improved; In addition, the risk tolerance of the electricity trading market must also be considered; if the electricity supply is relatively tight or the market structure is relatively fragile, then relatively small fluctuations may be regarded as abnormal; for example, in a region that is mainly supplied by a single energy source and has small backup capacity, even if the volatility index is slightly higher than the normal range, it may be judged as abnormal. Differentiated settings should be made for different regions and different market structures to enhance the pertinence and applicability of the early warning mechanism.

[0033] The power grid transaction fluctuation index obtained in step 300 is compared with a preset fluctuation threshold; if the fluctuation index exceeds the set upper threshold, it means that there is an abnormal fluctuation caused by a sudden large-scale increase in power demand (such as a sharp increase in air conditioning usage under extremely high temperatures) or a supply-side failure; similarly, when the fluctuation index is lower than the lower threshold, it means that there is a sharp drop in power demand (such as a sudden shutdown of a large enterprise) or oversupply. Abnormal results of smart grid transactions are determined based on the comparison results. Abnormal results can be divided into different levels, such as mild abnormality, moderate abnormality and severe abnormality. If the volatility index just exceeds the upper threshold and the excess is small, it can be determined as a mild abnormality. If the excess is large and lasts for a long time, it is determined as a moderate or severe abnormality, which helps to take timely response measures of varying degrees and effectively prevent transaction risks.

[0034] like Figure 2 , Figure 3 As shown, the embodiment of the present invention provides a smart grid transaction abnormality warning device based on real-time data. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. From the hardware level, Figure 2 As shown, it is a hardware architecture diagram of an electronic device where a smart grid transaction abnormality warning device based on real-time data is provided in an embodiment of the present invention. Figure 2 In addition to the processor, memory, network interface, and non-volatile memory shown, the electronic device in the embodiment may also include other hardware, such as a forwarding chip responsible for processing messages, etc. Taking software implementation as an example, Figure 3 As shown, as a device in a logical sense, the CPU of the electronic device in which it is located reads the corresponding computer program in the non-volatile memory into the internal memory and runs it.

[0035] like Figure 3 As shown, the present invention provides a smart grid transaction abnormality early warning device based on real-time data, comprising: A data acquisition module 500, for acquiring real-time transaction data of smart grid transactions, and at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data; A serialization processing module 502 is used to perform serialization feature alignment on the historical transaction data, the real-time transaction data and the predicted transaction data according to time series features to obtain a power grid transaction detection sequence; The volatility assessment module 504 is used to perform a volatility assessment on the power grid transaction detection sequence to obtain a power grid transaction volatility index; The abnormality judgment module 506 is used to use the preset power grid transaction fluctuation threshold to perform abnormality judgment on the power grid transaction fluctuation index to obtain the smart grid transaction abnormality result.

[0036] In an embodiment of the present invention, the data acquisition module 500 can be used to execute step 100 in the above method embodiment, the serialization processing module 502 can be used to execute step 200 in the above method embodiment, the volatility assessment module 504 can be used to execute step 300 in the above method embodiment, and the abnormality judgment module 506 can be used to execute step 400 in the above method embodiment.

[0037] In this embodiment, firstly, the transaction data is collected in real time by the data acquisition module and combined with the historical and predicted data to realize the fusion of multi-dimensional data, which can comprehensively capture the dynamic change characteristics of transaction behavior and significantly improve the accuracy and timeliness of early warning; secondly, the serialization processing module aligns the historical, real-time and predicted transaction data based on the time series characteristics, constructs the power grid transaction detection sequence, effectively solves the problem that the traditional method is difficult to capture the time series changes, and enhances the sensitivity of anomaly detection; in addition, the volatility assessment module quantifies the transaction volatility index and combines the threshold analysis of the anomaly judgment module to realize real-time monitoring and early warning of transaction anomalies, thereby improving the stability and security of the smart grid transaction market; through multi-dimensional data fusion, time series feature analysis and volatility assessment, the accuracy, timeliness and foresight of abnormal warning are significantly improved, providing strong support for the reliable operation of the power system.

[0038] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on a smart grid transaction abnormality warning device based on real-time data. In other embodiments of the present invention, a smart grid transaction abnormality warning device based on real-time data may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0039] The information interaction, execution process and other contents between the modules in the above-mentioned device are based on the same concept as the embodiment of the method of the present invention. For the specific contents, please refer to the description in the embodiment of the method of the present invention, and no further description is given here.

[0040] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, a smart grid transaction abnormality warning method based on real-time data in any embodiment of the present invention is implemented.

[0041] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the processor executes a smart grid transaction abnormality warning method based on real-time data in any embodiment of the present invention.

[0042] Specifically, a system or device equipped with a storage medium can be provided, on which software program code that implements the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device can be enabled to read and execute the program code stored in the storage medium.

[0043] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0044] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0045] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0046] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0047] It should be noted that, in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical factors in the process, method, article or device including the elements.

[0048] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0049] 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 make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart grid transaction abnormality early warning method based on real-time data, characterized in that: The method comprises: Acquire real-time transaction data of smart grid transactions, and at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data; According to the time series characteristics, the historical transaction data, the real-time transaction data and the predicted transaction data are serialized and aligned to obtain a power grid transaction detection sequence; Performing a volatility assessment on the power grid transaction detection sequence to obtain a power grid transaction volatility index; The preset power grid transaction fluctuation threshold is used to perform abnormal judgment on the power grid transaction fluctuation index to obtain the abnormal result of smart grid transaction.

2. The smart grid transaction abnormality early warning method based on real-time data according to claim 1 is characterized in that: The method for obtaining at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data comprises: Obtaining a transaction environment feature set corresponding to real-time transaction data of smart grid transactions; Using the transaction environment feature set as a screening condition, searching in a smart grid transaction database to obtain at least one set of historical transaction data; the at least one set of historical transaction data includes transaction data that is closest to the real-time transaction data in time sequence and matches the transaction environment feature set; The transaction environment feature set is used as input and input into a preset smart grid transaction prediction model to obtain at least one set of predicted transaction data; the at least one set of predicted transaction data includes transaction data that is closest to the real-time transaction data in time series and matches the transaction environment feature set during future transactions of the smart grid.

3. The smart grid transaction abnormality early warning method based on real-time data as claimed in claim 2 is characterized in that: The transaction environment feature set includes at least one of a power supply feature subset, a power demand feature subset, a market policy feature subset, and a meteorological environment feature subset; Wherein, the power supply feature subset includes at least any one of the proportion of power generation types, power generation capacity and available capacity; The power demand feature subset includes at least one of a time period feature and a season feature; The market policy feature subset includes at least one of the number of market entities participating in power trading, market share distribution, policy subsidies and policy restrictions; The meteorological environment feature subset includes at least one of temperature, humidity, light intensity and air pressure.

4. The smart grid transaction abnormality early warning method based on real-time data as claimed in claim 3 is characterized in that: The real-time transaction data at least includes real-time transaction time, real-time transaction volume and real-time transaction price; The historical transaction data at least includes historical transaction time, historical transaction volume and historical transaction price; The predicted transaction data at least includes future transaction time, future transaction volume and future transaction price.

5. The smart grid transaction abnormality early warning method based on real-time data as claimed in claim 4 is characterized in that: The method for calculating the power grid transaction fluctuation index includes: In the power grid transaction detection sequence, the range of transaction volume and transaction price are calculated respectively; For the transaction price and transaction volume in the power grid transaction detection sequence, respectively calculate the corresponding standard deviation; According to the importance of transaction volume and transaction price in the power grid transaction process to the power transaction market, the weight coefficients corresponding to the transaction volume range, transaction price range, transaction price standard deviation and transaction volume standard deviation are set respectively; Based on the determined weight coefficient, the calculated transaction volume range, transaction price range, transaction price standard deviation and transaction volume standard deviation are weighted and analyzed to obtain the power grid transaction volatility index.

6. The smart grid transaction abnormality early warning method based on real-time data as claimed in claim 5, characterized in that: The calculation formula of the power grid transaction fluctuation index is: ； in, represents the power grid transaction volatility index; , , and Respectively represent the weight coefficients corresponding to the transaction volume range, transaction price range, transaction price standard deviation and transaction volume standard deviation; Indicates extremely poor trading volume; Indicates that the transaction price is extremely bad; Indicates the standard deviation of transaction prices; represents the standard deviation of trading volume; , , and They represent the standardized functions corresponding to the range of trading volume, the range of trading price, the standard deviation of trading price and the standard deviation of trading volume respectively.

7. The smart grid transaction abnormality early warning method based on real-time data as claimed in claim 2, characterized in that: The smart grid transaction prediction model adopts any one of a machine learning model and a deep learning model. The input of the smart grid transaction prediction model is a transaction environment feature set, and the output of the smart grid transaction prediction model is at least one set of predicted transaction data.

8. A smart grid transaction abnormality warning device based on real-time data, characterized in that: The device comprises: A data acquisition module, used to acquire real-time transaction data of smart grid transactions, and at least one set of historical transaction data and at least one set of predicted transaction data corresponding to the real-time transaction data; A serialization processing module, used for performing serialization feature alignment on the historical transaction data, the real-time transaction data and the predicted transaction data according to time series features to obtain a power grid transaction detection sequence; A volatility assessment module, used for performing a volatility assessment on the power grid transaction detection sequence to obtain a power grid transaction volatility index; The abnormality judgment module is used to use a preset power grid transaction fluctuation threshold to perform abnormality judgment on the power grid transaction fluctuation index to obtain a smart grid transaction abnormality result.

9. An electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 7 are implemented.

Images