Method and device for identifying target enterprise

Abstract

The present application discloses a method and apparatus for identifying a target enterprise. Among them, the method includes: obtaining power consumption behavior data and external influencing factor data of multiple enterprises, where the power consumption behavior data at least includes: power consumption and power consumption power, and the external influencing factor data at least includes: historical temperature data, historical wind force data, and historical economic data; constructing power consumption behavior characteristics of multiple enterprises respectively according to the power consumption behavior data and the external influencing factor data; respectively inputting the power consumption behavior characteristics of multiple enterprises into a preset algorithm for identification, and identifying a target enterprise from multiple enterprises, where the target enterprise is an enterprise with abnormal production. The present application solves the technical problem of being unable to accurately identify enterprises with abnormal production due to the fact that most enterprises do not install pollution situation monitoring equipment terminals.

Description

Technical Field

[0001] This application relates to the field of identifying abnormal production enterprises. Specifically, it relates to a method and device for identifying target enterprises. Background Art

[0002] With the development of the global industry and economy, the consumption of resources by humans has been increasing continuously, and environmental problems of varying degrees have emerged in countries around the world. In recent years, while China's economy has been developing rapidly, environmental pollution problems have gradually become prominent. In order to strictly control the deterioration of environmental pollution, China has promulgated a series of policy measures, monitored and supervised some key polluting enterprises, and regarded key enterprises as the focus of implementing pollution discharge permit management, supervision and monitoring, law enforcement supervision, environmental statistics, pollution control, and pollution source management. The key polluting enterprises are monitored from multiple aspects, and the electricity consumption behavior is an important factor reflecting the production situation of key polluting enterprises.

[0003] At present, most enterprises have not fully installed real-time pollution monitoring equipment terminals and cannot accurately obtain the production situation of key polluting enterprises. To supervise and monitor these key polluting enterprises or conduct on-site law enforcement requires a large amount of manpower, material resources and time, and the results are affected by human factors because law enforcement officers are involved. To accurately identify abnormal production enterprises from key polluting enterprises, it is necessary to study a method of unsupervised analysis through the electricity consumption behavior of key polluting enterprises to obtain abnormal production enterprises among them. At present, most solutions use traditional statistical methods, but for the characteristics of large electricity consumption data volume, numerous types, and high growth rate, they cannot extract the deep-seated reasons hidden behind abnormal data and have low computing efficiency.

[0004] For the above problems, no effective solutions have been proposed yet. Summary of the Invention

[0005] The embodiments of this application provide a method and device for identifying target enterprises to at least solve the technical problem that abnormal production enterprises cannot be accurately identified due to most enterprises not installing pollution monitoring equipment terminals.

[0006] According to one aspect of the embodiments of this application, a method for identifying target enterprises is provided, including: obtaining electricity consumption behavior data and external influencing factor data of multiple enterprises, where the electricity consumption behavior data at least includes: electricity consumption and electricity power, and the external influencing factor data at least includes: historical temperature data, historical wind data, and historical economic data; constructing electricity consumption behavior characteristics of multiple enterprises respectively based on the electricity consumption behavior data and the external influencing factor data; respectively inputting the electricity consumption behavior characteristics of multiple enterprises into a preset algorithm for identification, and identifying target enterprises from multiple enterprises, where the target enterprises are enterprises with abnormal production.

[0007] Optionally, after obtaining the electricity consumption behavior data and external influencing factor data of multiple enterprises, the method further includes: preprocessing the electricity consumption behavior data and external influencing factor data; preprocessing the electricity consumption behavior data and external influencing factor data includes: removing the unique attributes in the electricity consumption behavior data and external influencing factor data; processing the missing values in the electricity consumption behavior data and external influencing factor data.

[0008] Optionally, processing the missing values in the electricity consumption behavior data and external influencing factor data includes: using the features including missing values; deleting the features including missing values; filling in the missing values of the features.

[0009] Optionally, based on the electricity consumption behavior data and external influencing factor data, the electricity consumption behavior characteristics of multiple enterprises are respectively constructed, including: using the association rule mining algorithm to analyze the association relationship between the external influencing factor data and the electricity consumption behavior data of multiple enterprises; determining the target factor data in the external influencing factor data that has the greatest impact on the electricity consumption behavior data of multiple enterprises according to the association relationship; predicting and analyzing the electricity consumption fluctuations caused by the impact of the target factor data through the BP neural network, and removing the electricity consumption fluctuations to obtain the processed electricity consumption behavior data; constructing the electricity consumption behavior characteristics of multiple enterprises by using the processed electricity consumption behavior data.

[0010] Optionally, the characteristic indicators of the electricity consumption behavior characteristics of multiple enterprises include: electricity consumption activity and electricity consumption level.

[0011] Optionally, the electricity consumption behavior characteristics of multiple enterprises are respectively input into a preset algorithm for identification, and the target enterprise is identified from multiple enterprises, including: inputting the characteristic indicators of the electricity consumption behavior characteristics of multiple enterprises into the preset algorithm; calculating the local outlier factor by the local outlier factor detection method; identifying the target enterprise from multiple enterprises in combination with the local outlier factor.

[0012] Optionally, identifying the target enterprise from multiple enterprises in combination with the local outlier factor includes: when the value of the local outlier factor is less than or equal to 1.5, regarding the enterprise as a normal production enterprise; when the value of the local outlier factor is greater than 1.5 and less than or equal to 2, regarding the enterprise as a suspected abnormal production enterprise; when the value of the local outlier factor is greater than 2, regarding the enterprise as the target enterprise.

[0013] According to another aspect of the embodiments of the present application, there is also provided a recognition device for a target enterprise, including: an acquisition module, configured to acquire power consumption behavior data and external influencing factor data of multiple enterprises, where the power consumption behavior data at least includes: power consumption and power consumption power, and the external influencing factor data at least includes: historical temperature data, historical wind force data, and historical economic data; a construction module, configured to respectively construct power consumption behavior characteristics of multiple enterprises based on the power consumption behavior data and the external influencing factor data; and an identification module, configured to respectively input the power consumption behavior characteristics of multiple enterprises into a preset algorithm for identification, and identify a target enterprise from multiple enterprises, where the target enterprise is an enterprise with abnormal production.

[0014] According to still another aspect of the embodiments of the present application, there is also provided a non-volatile storage medium, and the non-volatile storage medium includes a stored program, where when the program runs, it controls the device where the non-volatile storage medium is located to execute the above-mentioned recognition method for a target enterprise.

[0015] According to still another aspect of the embodiments of the present application, there is also provided a processor, and the processor is configured to run a program stored in a memory, where when the program runs, it executes the above-mentioned recognition method for a target enterprise.

[0016] In the embodiments of the present application, by acquiring power consumption behavior data and external influencing factor data of multiple enterprises, where the power consumption behavior data at least includes: power consumption and power consumption power, and the external influencing factor data at least includes: historical temperature data, historical wind force data, and historical economic data; respectively constructing power consumption behavior characteristics of multiple enterprises based on the power consumption behavior data and the external influencing factor data; and respectively inputting the power consumption behavior characteristics of multiple enterprises into a preset algorithm for identification, and identifying a target enterprise from multiple enterprises, where the target enterprise is an enterprise with abnormal production, the purpose of identifying an enterprise with abnormal production is achieved by constructing the power consumption behavior characteristics of the enterprise and inputting them into a preset algorithm, thereby realizing the technical effect of accurately and efficiently detecting users of enterprises with abnormal production, and further solving the technical problem of being unable to accurately identify enterprises with abnormal production due to the fact that most enterprises do not install pollution monitoring equipment terminals. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0018] Figure 1 is a flowchart of a recognition method for a target enterprise according to an embodiment of the present application;

[0019] Figure 2It is a structural block diagram of an identification device for a target enterprise according to an embodiment of the present application. Detailed implementation manners

[0020] In order to enable those skilled in the art of the present technology to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present application.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order different from those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0022] According to an embodiment of the present application, an embodiment of a method for identifying a target enterprise is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described here can be executed in an order different from that here.

[0023] Figure 1 It is a flowchart of a method for identifying a target enterprise according to an embodiment of the present application. As Figure 1 shown, the method includes the following steps:

[0024] Step S102, obtain the electricity consumption behavior data and external influencing factor data of multiple enterprises. Among them, the electricity consumption behavior data at least includes: electricity consumption and electricity power, and the external influencing factor data at least includes: historical temperature data, historical wind power data, and historical economic data;

[0025] In this step, the electricity consumption behavior information and electricity customer information in the electricity collection system and the marketing business application system are associated to obtain the user information (user number, user name, user address, substation area information, contract capacity, etc.) and electricity consumption information (metering point information, electricity meter information, daily electricity consumption, power, etc.) of one year; external data such as historical temperature, historical wind force, and historical economic situation are obtained through external data on the Internet.

[0026] Step S104: Construct the electricity consumption behavior characteristics of multiple enterprises respectively based on the electricity consumption behavior data and external influencing factor data.

[0027] According to an optional embodiment of the present application, when performing step S104, the Apriori algorithm is combined to analyze the correlation between external factors such as temperature and wind force and electricity consumption behavior, and the BP neural network is used to predict and analyze the electricity consumption fluctuations of enterprises caused by the external factors with greater influence, and the electricity consumption characteristics of enterprise users are calculated by removing and combining the analysis of user electricity consumption behavior.

[0028] Step S106: Input the electricity consumption behavior characteristics of multiple enterprises into a preset algorithm for identification respectively, and identify the target enterprises from multiple enterprises, where the target enterprises are the enterprises with abnormal production.

[0029] When performing step S106, the LOF algorithm is used to analyze and identify abnormal production enterprises by selecting appropriate features.

[0030] Through the above steps, by constructing the electricity consumption behavior characteristics of enterprises and inputting them into a preset algorithm, the purpose of identifying abnormal production enterprises is achieved, thus realizing the technical effect of accurately and efficiently detecting abnormal production enterprise users.

[0031] According to an optional embodiment of the present application, after performing step S102 to obtain the electricity consumption behavior data and external influencing factor data of multiple enterprises, the method further includes: preprocessing the electricity consumption behavior data and external influencing factor data; preprocessing the electricity consumption behavior data and external influencing factor data includes: removing the unique attributes in the electricity consumption behavior data and external influencing factor data; processing the missing values in the electricity consumption behavior data and external influencing factor data.

[0032] Preprocess the data, that is, clean the extracted data. The original data should mainly have problems in two aspects: integrity and accuracy. Integrity mainly checks whether the units or individuals to be investigated are missing, and whether all the investigation items or indicators are filled in completely. Accuracy mainly includes two aspects: one is to check whether the data reflects the objective reality truthfully and whether the content conforms to the reality; the other is to check whether there are errors in the data and whether the calculations are correct, etc. The main methods for verifying data accuracy are logical check and calculation check. Logical check mainly verifies whether the data conforms to logic, whether the content is reasonable, and whether there are contradictions between each item or number. This method is mainly suitable for the verification of qualitative data. Calculation check is to check whether there are errors in the calculation results and calculation methods of each item of data in the questionnaire, and is mainly used for the verification of quantitative data. Preprocess the data, specifically including: removing unique attributes, usually some id attributes, which cannot describe the distribution law of the sample itself, so simply delete these attributes; handling missing values.

[0033] According to an optional embodiment of the present application, handling missing values in the electricity consumption behavior data and external influencing factor data includes: using features including missing values; deleting features including missing values; and filling in the features with missing values.

[0034] The three methods for handling missing values include: directly using features containing missing values; deleting features containing missing values (this method is effective when the attribute containing missing values contains a large number of missing values and only a very small number of valid values); filling in missing values. Filling in missing values means filling the missing values with a unified default value or filling them with the statistic of the feature (such as mean, minimum value, median, etc.). Common methods for filling in missing values include: mean imputation, similar mean imputation, modeling prediction, high-dimensional mapping, multiple imputation, maximum likelihood estimation, compressive sensing, and matrix completion.

[0035] In some other optional embodiments of the present application, performing step S104 to construct the electricity consumption behavior characteristics of multiple enterprises based on the electricity consumption behavior data and external influencing factor data includes: using the association rule mining algorithm to analyze the association relationship between the external influencing factor data and the electricity consumption behavior data of multiple enterprises; determining the target factor data in the external influencing factor data that has the greatest influence on the electricity consumption behavior data of multiple enterprises according to the association relationship; predicting and analyzing the electricity consumption fluctuations generated by the influence of the target factor data through a BP neural network, and removing the electricity consumption fluctuations to obtain the processed electricity consumption behavior data; constructing the electricity consumption behavior characteristics of multiple enterprises using the processed electricity consumption behavior data.

[0036] After data cleaning, the electricity customer information, metering point information, electricity meter information, and substation area information are associated. Combining multi-dimensional data such as enterprise electricity consumption data, external temperature data, external wind data, and external economic data, the external influencing factors of electricity consumption behavior are analyzed from multiple perspectives of temperature, economy, and wind. The Apriori algorithm is used to analyze the correlation between external factors and electricity consumption behavior. The analysis results of the Apriori algorithm are shown in the following table:

[0037] Factor name Support degree Confidence degree Lift degree Wind force 0.00291 0.58333 6.09312 Air temperature 0.07605 0.77778 75.28103 Economy 0.00582 0.69230 11.95101

[0038] Combined with the analysis results in the above table, temperature is an important factor affecting the electricity consumption of enterprises. To accurately identify abnormal production enterprises in the later stage, the BP neural network is used to combine the electricity consumption of enterprises in various industries and the temperature changes in the past year to predict and analyze the electricity consumption fluctuations of enterprises caused by temperature factors. After removing part of the electricity consumption, characteristic indicators of enterprise production electricity consumption behavior (such as electricity consumption activity, regional electricity consumption level, industry electricity consumption level, overall electricity consumption level, etc.) are constructed.

[0039] Analyze the data of 632 enterprises in a certain area. Take 80% of 632 (a total of 506 households, including 25 abnormal production enterprise users, 56 suspected abnormal production enterprise users, and 425 normal enterprise users) as training data, and 20% as test data (a total of 126 households, including 7 abnormal production enterprise users, 15 suspected abnormal production enterprise users, and 104 normal enterprise users). Remove the electricity consumption fluctuations of enterprises caused by temperature influence from the training data and test data, and calculate the electricity consumption characteristics of enterprise users by combining user electricity consumption behavior analysis.

[0040] This invention considers that the electricity consumption of enterprises includes living electricity consumption and production electricity consumption. To ensure the accuracy of the model, the Apriori algorithm is used to analyze the correlation of external factors for all enterprise user information and user electricity consumption information. The BP neural network is used for prediction and analysis to remove the electricity consumption fluctuations of enterprises caused by temperature factors, so as to more accurately and efficiently identify abnormal production enterprise users.

[0041] According to an optional embodiment of the present application, the characteristic indicators of the electricity consumption behavior of multiple enterprises include: electricity consumption activity and electricity consumption level.

[0042] Among them, the electricity consumption activity refers to the fluctuation of an enterprise's electricity consumption on the current day relative to its own electricity consumption. The specific calculation method is the ratio of the enterprise's electricity consumption on the current day to the average electricity consumption in the previous year and the ratio of the enterprise's electricity consumption on the current day to the enterprise's contract capacity, which is calculated according to a weight ratio of 8:2. The electricity consumption level is the influence weight of the enterprise's own electricity consumption situation among all key enterprises. Each enterprise has 3 electricity consumption levels, namely the regional electricity consumption level (the proportion of the enterprise's electricity consumption on the current day in the median of the electricity consumption of key enterprises in the region), the industry electricity consumption level (the proportion of the enterprise's electricity consumption on the current day in the median of the electricity consumption of key enterprises in the industry), and the overall electricity consumption level (the proportion of the enterprise's electricity consumption on the current day in the median of the electricity consumption of all key enterprises).

[0043] The present invention constructs indicators such as electricity consumption activity, regional electricity consumption level, industry electricity consumption level, and overall electricity consumption level from three perspectives of the enterprise user's electricity consumption behavior data, namely itself, region, and industry, accurately describes the user's electricity consumption behavior, constructs enterprise user production electricity consumption behavior characteristic indicators in multiple aspects, and greatly improves the model recognition accuracy.

[0044] In some other alternative embodiments of the present application, when performing step S106, the electricity consumption behavior characteristics of multiple enterprises are respectively input into a preset algorithm for identification, and target enterprises are identified from multiple enterprises, including: inputting the characteristic indicators of the electricity consumption behavior characteristics of multiple enterprises into the preset algorithm; calculating the local outlier factor by the local outlier factor detection method; and identifying target enterprises from multiple enterprises in combination with the local outlier factor.

[0045] Taking the constructed electricity consumption behavior characteristic indicators as input data, calculating the local outlier factor through the LOF algorithm, and outputting a list of abnormal production enterprises in combination with this factor.

[0046] The principle of the LOF algorithm is as follows:

[0047] (1) K-distance: The k-distance of a data object q is defined as the distance from the k-th point closest to the data object q in the data set to q, denoted as k -distance(q), where the distance refers to the Euclidean distance, that is, the straight-line distance.

[0048] (2) K-distance neighborhood: The set of data points in the data set whose distance from the data object q is not greater than the k-distance, that is

[0049] N k -distance(q)(p) = {p ∈ D\{q}|d(q, p) ≤ k-distance(q)} (1)

[0050] (3) Reachability distance: For any two points p and q in the data set, the reachability distance from p to q is defined as:

[0051] reach-distk(p,q) = max{d(p,q) ≤ k - distance(q)} (2)

[0052] Among them, d(p, q) represents the Euclidean distance between point p and point q.

[0053] (4) Local reachability density: The local reachability density of q refers to the reciprocal of the average reachable distance from q to all points within its neighborhood. It is commonly represented by density, and the calculation method is as follows:

[0054]

[0055] Among them, N k (q) is the number of points within the k - neighborhood of q. Since there may be several points with the same distance to q, the k - nearest neighbor points may be more than one, so N k (q) ≥ k. If lrd k (q) is larger, it indicates that the density of q is higher and the q point is more normal.

[0056] (5) Local outlier factor: Characterizes the degree of outliers in the data, and the calculation method is as follows:

[0057]

[0058] (6) If the LOF value is much larger than 1, it indicates that the density of the q point is quite different from the overall data density and is regarded as an outlier. The closer the LOF value is to 1, the more normal the point q is.

[0059] The present invention uses the LOF algorithm to analyze abnormal production enterprises. This algorithm combines the data point q with the surrounding k points for analysis, making the finally obtained outlier factor value more reasonable, reducing the influence of the maximum and minimum density values on the overall data, and representing the degree of outliers of the data points in numerical form, which is easier to understand. Only one parameter k needs to be set, and it is easy to operate and implement.

[0060] According to an optional embodiment of the present application, identifying target enterprises from multiple enterprises in combination with the local outlier factor includes: when the value of the local outlier factor is less than or equal to 1.5, regarding the enterprise as a normal production enterprise; when the value of the local outlier factor is greater than 1.5 and less than or equal to 2, regarding the enterprise as a suspected abnormal production enterprise; when the value of the local outlier factor is greater than 2, regarding the enterprise as a target enterprise.

[0061] The local outlier factor is defined as follows: if the factor value is much greater than 1, it indicates that the density of this point is quite different from the overall data density, and this point is regarded as an outlier. The closer the factor value is to 1, the more normal this point is. In this embodiment, it is defined that if the factor value is greater than 2, it indicates that the density of this point is quite different from the overall data density, and this point is regarded as an outlier, and this enterprise is an abnormal production enterprise; if the factor value is greater than 1.5, then this enterprise is a suspected abnormal production enterprise. Putting the aforementioned training data into the abnormal production enterprise user identification model, and combining with the test data, it is predicted that there are 9 abnormal production enterprise users in the test data. By comparing with the test data, it is found that 9 are predicted to be abnormal production enterprise users (6 of them are real abnormal production enterprise users), and the accuracy rate reaches 85.71%.

[0062] The abnormal production enterprise user identification model established by the present invention is applied to the abnormal production enterprise user identification scenario. On the one hand, it reduces the input of human and material resources for the investigation of abnormal production enterprise users through system automatic identification, and on the other hand, it can improve the accuracy rate of abnormal power consumption identification and detection.

[0063] The above-mentioned identification method of the target enterprise provided by this application has been proved feasible through experiments, simulations and use. By analyzing the electricity consumption behavior data of enterprise users, removing the electricity consumption fluctuations of enterprises caused by external factors with greater influence, and combining the analysis of user electricity consumption behavior to calculate the electricity consumption characteristics of enterprise users, appropriate features are selected and the LOF algorithm is used to analyze and identify abnormal production enterprises. The method has been tested in multiple regions through experiments and can more accurately and effectively identify abnormal production enterprises and improve the investigation efficiency of abnormal production enterprises.

[0064] Figure 2 It is the structural block diagram of an identification device of a target enterprise according to an embodiment of the present application, as Figure 2 shown, the device includes:

[0065] An acquisition module 20, configured to acquire the electricity consumption behavior data and external influence factor data of multiple enterprises, wherein the electricity consumption behavior data at least includes: electricity consumption and electricity power, and the external influence factor data at least includes: historical temperature data, historical wind power data and historical economic data;

[0066] A construction module 22, configured to respectively construct the electricity consumption behavior characteristics of multiple enterprises according to the electricity consumption behavior data and external influence factor data;

[0067] An identification module 24, configured to respectively input the electricity consumption behavior characteristics of multiple enterprises into a preset algorithm for identification, and identify the target enterprise from multiple enterprises, wherein the target enterprise is an enterprise with abnormal production.

[0068] It should be noted that Figure 2 The preferred implementation manners of the illustrated embodiments can be referred to Figure 1The relevant descriptions of the illustrated embodiments will not be elaborated here.

[0069] According to another aspect of the embodiments of the present application, a non-volatile storage medium is further provided. The non-volatile storage medium includes a stored program, wherein when the program runs, it controls the device where the non-volatile storage medium is located to execute the above-mentioned identification method of the target enterprise.

[0070] The above non-volatile storage medium is used to store a program for performing the following functions: obtaining the electricity consumption behavior data and external influencing factor data of multiple enterprises, wherein the electricity consumption behavior data at least includes: electricity consumption and electricity power, and the external influencing factor data at least includes: historical temperature data, historical wind power data, and historical economic data; respectively constructing the electricity consumption behavior characteristics of multiple enterprises according to the electricity consumption behavior data and external influencing factor data; respectively inputting the electricity consumption behavior characteristics of multiple enterprises into a preset algorithm for identification, and identifying the target enterprise from multiple enterprises, wherein the target enterprise is an enterprise with abnormal production.

[0071] The embodiments of the present application further provide a processor, and the processor is used to run a program stored in a memory, wherein when the program runs, it executes the above-mentioned identification method of the target enterprise.

[0072] The above processor is used to run a program for performing the following functions: obtaining the electricity consumption behavior data and external influencing factor data of multiple enterprises, wherein the electricity consumption behavior data at least includes: electricity consumption and electricity power, and the external influencing factor data at least includes: historical temperature data, historical wind power data, and historical economic data; respectively constructing the electricity consumption behavior characteristics of multiple enterprises according to the electricity consumption behavior data and external influencing factor data; respectively inputting the electricity consumption behavior characteristics of multiple enterprises into a preset algorithm for identification, and identifying the target enterprise from multiple enterprises, wherein the target enterprise is an enterprise with abnormal production.

[0073] The serial numbers of the embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0074] In the above embodiments of the present application, the descriptions of the respective embodiments have their own emphases. For the parts not elaborated in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0075] In several embodiments provided in the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units can be a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of units or modules can be in electrical or other forms.

[0076] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can physically exist alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0078] If the above-mentioned 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 such an understanding, the technical solution of the present application, in essence, or the part that contributes to the related technology, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs and other various media that can store program codes.

[0079] The above is only the preferred embodiment of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A method for identifying a target enterprise, characterized in that, it includes: Obtain the electricity consumption behavior data and external influencing factor data of multiple enterprises. Among them, the electricity consumption behavior data at least includes: electricity consumption and electricity power, and the external influencing factor data at least includes: historical temperature data, historical wind power data, and historical economic data; According to the electricity consumption behavior data and the external influencing factor data, construct the electricity consumption behavior characteristics of the multiple enterprises respectively, including: electricity consumption activity and electricity consumption level; Input the electricity consumption behavior characteristics of the multiple enterprises into a preset algorithm for identification respectively, and identify the target enterprise from the multiple enterprises, including: input the characteristic indexes of the electricity consumption behavior characteristics of the multiple enterprises into the preset algorithm; calculate the local outlier factor by the local outlier factor detection method; combine the local outlier factor to identify the target enterprise from the multiple enterprises. Among them, when the value of the local outlier factor is greater than 2, the enterprise is regarded as the target enterprise, and the target enterprise is an enterprise with abnormal production.

2. The method according to claim 1, characterized in that, after obtaining the electricity consumption behavior data and external influencing factor data of multiple enterprises, the method further includes: preprocessing the electricity consumption behavior data and external influencing factor data; Preprocessing the electricity consumption behavior data and external influencing factor data includes: Removing the unique attributes in the electricity consumption behavior data and external influencing factor data; Processing the missing values in the electricity consumption behavior data and external influencing factor data.

3. The method according to claim 2, characterized in that, Processing the missing values in the electricity consumption behavior data and external influencing factor data includes: Using the features including the missing values; Deleting the features including the missing values; Completing the features of the missing values.

4. The method according to claim 1, characterized in that, According to the electricity consumption behavior data and the external influencing factor data, construct the electricity consumption behavior characteristics of the multiple enterprises respectively, including: Using the association rule mining algorithm to analyze the association relationship between the external influencing factor data and the electricity consumption behavior data of the multiple enterprises; Determine the target factor data in the external influencing factor data that has the greatest influence on the electricity consumption behavior data of the multiple enterprises according to the association relationship; Predict and analyze the electricity consumption fluctuations caused by the influence of the target factor data through a BP neural network, and remove the electricity consumption fluctuations to obtain the processed electricity consumption behavior data; Use the processed electricity consumption behavior data to construct the electricity consumption behavior characteristics of the multiple enterprises.

5. The method according to claim 1, characterized in that, Combining the local outlier factor to identify the target enterprise from the multiple enterprises further includes: When the value of the local outlier factor is less than or equal to 1.5, the enterprise is regarded as a normal production enterprise; When the value of the local outlier factor is greater than 1.5 and less than or equal to 2, the enterprise is regarded as a suspected abnormal production enterprise.

6. An identification device for a target enterprise, characterized in that, it includes: An acquisition module, configured to acquire power consumption behavior data and external influencing factor data of multiple enterprises, wherein the power consumption behavior data at least includes: power consumption and power consumption power, and the external influencing factor data at least includes: historical temperature data, historical wind power data, and historical economic data; A construction module, configured to respectively construct power consumption behavior characteristics of the multiple enterprises according to the power consumption behavior data and the external influencing factor data, including: power consumption activity and power consumption level; An identification module, configured to respectively input the power consumption behavior characteristics of the multiple enterprises into a preset algorithm for identification, and identify target enterprises from the multiple enterprises, including: inputting characteristic indexes of the power consumption behavior characteristics of the multiple enterprises into the preset algorithm; calculating a local outlier factor by a local outlier factor detection method; identifying target enterprises from the multiple enterprises in combination with the local outlier factor, wherein when the value of the local outlier factor is greater than 2, the enterprise is regarded as the target enterprise, and the target enterprise is an enterprise with abnormal production.

7. A non-volatile storage medium, characterized in that, the non-volatile storage medium includes a stored program, wherein when the program runs, it controls the device where the non-volatile storage medium is located to execute the identification method of the target enterprise according to any one of claims 1 to 5.

8. A processor, characterized in that, the processor is used to run a program stored in a memory, wherein when the program runs, it executes the identification method of the target enterprise according to any one of claims 1 to 5.

Images