CN113313407A - Enterprise power utilization behavior identification method and device - Google Patents
Enterprise power utilization behavior identification method and device Download PDFInfo
- Publication number
- CN113313407A CN113313407A CN202110666573.8A CN202110666573A CN113313407A CN 113313407 A CN113313407 A CN 113313407A CN 202110666573 A CN202110666573 A CN 202110666573A CN 113313407 A CN113313407 A CN 113313407A
- Authority
- CN
- China
- Prior art keywords
- enterprise
- data
- power data
- illegal
- field
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 238000001514 detection method Methods 0.000 claims abstract description 8
- 230000005611 electricity Effects 0.000 claims description 33
- 230000006399 behavior Effects 0.000 claims description 23
- 238000004458 analytical method Methods 0.000 claims description 16
- 238000010219 correlation analysis Methods 0.000 claims description 11
- 230000001174 ascending effect Effects 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000012800 visualization Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000007794 visualization technique Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000006835 compression Effects 0.000 claims description 3
- 238000007906 compression Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000149 penetrating effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an enterprise power consumption behavior identification method and equipment, wherein the enterprise power consumption behavior identification method comprises the following steps: preprocessing the power data of an enterprise to obtain characteristic data; performing model training by taking the obtained characteristic data as input and output of a deep automatic coding network, and training the deep automatic coding network to obtain a power data detection model; and when new enterprise power data appear, repeatedly preprocessing the enterprise power data, detecting model reconstruction data by using the trained power data, calculating a reconstruction error, and judging whether the enterprise has illegal investment risk or not by comparing the reconstruction error with a set threshold value. The invention judges whether the illegal funding risk exists or not through the enterprise power data, and has higher reliability.
Description
Technical Field
The invention relates to the technical field of electric power data processing and analysis, in particular to a method and equipment for identifying enterprise electricity utilization behaviors.
Background
Illegal funding refers to the act of a unit or person raising funds to the public in a manner of issuing stocks, bonds, tickets, fund investing securities or other bond vouchers, without being approved by the relevant departments according to legal procedures, and committing to pay money or give rewards to payers in money, objects and other manners within a certain period of time.
However, financial innovation is deepened continuously, financial disordering is caused, online illegal collection is forbidden frequently, stock right people develop and run frequently, regional financial asset transaction centers are disordering, a plurality of difficulties such as heavy supervision tasks, difficult risk discovery, difficult study and judgment of financial cases, difficult illegal asset disposal and the like are brought to financial supervision, a financial governing department cannot master all information and usually intervenes passively after risk outbreak, and as illegal collection institutions often set up empty-shell companies, exaggerate and create operational benefits, investors are difficult to judge by themselves from a small amount of information visible on the internet.
The power utilization data of the power grid has the characteristics of multiple label points, large data throughput, stable equipment, high reliability and the like, and has unique advantages in monitoring power utilization abnormal enterprises such as vacant shells and the like, for example: the investigation shows that the historical electricity consumption and the electricity charge of the vacant shell enterprise are generally in a lower level, are obviously lower than the distribution condition of the industry electricity consumption, and have lower relevance with the industry. Therefore, on the basis of the electricity utilization characteristic data of the vacant enterprises, an electricity utilization abnormal enterprise monitoring model can be constructed on the basis of multidimensional electricity data, and the illegal funding early warning monitoring scheme based on the electricity data can improve the capacity of financial monitoring in the aspects of wide coverage, penetrability, agility, traceability and the like, so that the introduction of the electricity data has great practical significance for finding illegal funding organizations with high concealment.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a method and equipment for identifying enterprise electricity utilization behaviors with higher reliability.
In order to solve the problems, the technical scheme of the invention is as follows:
an enterprise electricity consumption behavior identification method comprises the following steps:
preprocessing the power data of an enterprise to obtain characteristic data;
performing model training by taking the obtained characteristic data as input and output of a deep automatic coding network, and training the deep automatic coding network to obtain a power data detection model; and
when new enterprise power data appear, the enterprise power data are repeatedly preprocessed, the trained power data detection model is used for reconstructing data, reconstruction errors are calculated, and whether the enterprises have illegal investment risk or not is judged by comparing the reconstruction errors with a set threshold value.
Optionally, the step of preprocessing the power data of the enterprise to obtain the feature data specifically includes:
carrying out mean value completion on vacant numerical values in the electric power data of the enterprise;
carrying out standardized processing on the power data of the enterprise;
performing linear correlation analysis and time trend visualization analysis on the field subjected to data preprocessing;
and performing feature extraction on the power utilization field which is obtained after linear correlation analysis and time trend visual analysis and is easy to distinguish illegal collection to obtain feature data.
Optionally, the step of performing mean value completion on the vacant numerical values in the power data of the enterprise specifically includes: and designing a level mapping table, converting all level fields into numerical values, and filling the average values of field information recorded at other moments in the vacant moments after the conversion is finished.
Optionally, the step of normalizing the power data of the enterprise specifically includes: for numerical variables, the min-max normalization formula is adopted:
data′＝data-min
max-min, where max and min are the maximum and minimum values of the variable, respectively; for the grade variable, the grade is converted to a numerical value.
Optionally, the step of performing linear correlation analysis and time trend visualization analysis on the field after data preprocessing specifically includes: calculating the correlation among different field data, and selecting one from a correlation field set with a larger absolute value of a correlation coefficient for reservation, so as to reduce overfitting of a model to redundant field data; and selecting the characteristic fields for effectively distinguishing the illegal fundraising companies from the normal enterprises by using a time trend graph visualization method.
Optionally, the step of extracting features of the power utilization field which is obtained after the linear correlation analysis and the time trend visual analysis and is easy to distinguish illegal funding includes: extracting characteristics of mean value, variance, ascending frequency ratio, descending frequency ratio, ascending average gradient and descending average gradient for the power utilization field represented or recorded by using continuous numerical values; for the rank field, the rank frequency ratio is adopted as its characteristic.
Optionally, the deep automatic coding network training formula is:
wherein the content of the first and second substances,
is the input vector of the input vector,
is the output vector, y ═ y1,y2,L,ym}∈RmM < n + s is the representation of the characteristics of x after DAE compression, f is the activation function, W and W 'are the weights of the encoding and decoding processes, respectively, and B' are the biases of the hidden and output layers.
Optionally, if the reconstruction error is smaller than a set threshold, the new enterprise power data is considered to be normal, and if the reconstruction error is larger than the set threshold, the new enterprise has an illegal funding risk.
Further, the present invention also provides an enterprise power consumption behavior identification device, which includes at least one processor and at least one memory storing at least one program, when the at least one program is executed by the at least one processor, the at least one processor is enabled to implement the enterprise power consumption behavior identification method as described above.
Compared with the prior art, the enterprise electricity consumption behavior identification method and the enterprise electricity consumption behavior identification equipment have the advantages that:
1. compared with the existing method that whether the enterprise is an illegal collective resource enterprise is judged through financial data, the method that the enterprise power data is used for judging is higher in reliability, and the credibility of the judgment on the illegal collective resource risk of the enterprise is improved.
2. The illegal funding early warning and monitoring scheme based on the enterprise power data improves the capacity of financial monitoring in the aspects of wide coverage, penetrating power, agility, traceability and the like, and has great practical significance for finding illegal funding institutions with high concealment.
3. The invention adopts a deep self-coding network algorithm, deeply excavates the implicit power utilization rule of illegal collection, does not need to learn the characteristics of negative samples and avoids the problem of insufficient quantity of negative samples.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
fig. 1 is a flow chart of an enterprise electricity consumption behavior identification method according to an embodiment of the present invention;
FIG. 2 is a block diagram of a process for pre-processing power data of an enterprise according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a deep self-coding network structure model according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an enterprise electricity consumption behavior identification device according to an embodiment of the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
The invention constructs an accurate enterprise electricity consumption behavior identification method by taking the electric power big data of the national power grid big data center as a core. The method comprises the steps of carrying out deep analysis and research on historical electricity utilization rules of enterprises judged to be illegal collective resources, effectively extracting data features on important power fields by using a data mining and visualization method, providing a self-optimization training scheme based on a deep self-coding network, carrying out feature modeling on most normal enterprise electricity utilization features existing in a data set by using model characteristics of the self-coding network which can carry out unsupervised training under the condition of no large number of illegal collective resources and no manual class labels of normal enterprises, comparing a reconstruction error with a set threshold value, and screening companies with illegal collective resources.
Specifically, the invention provides an enterprise electricity consumption behavior identification method, as shown in fig. 1, the method includes the following steps:
s1: preprocessing the power data of an enterprise to obtain characteristic data;
specifically, the power data of the enterprise is huge and numerous in labels, and the data needs to be preprocessed, so that the data quality is improved, and the calculated amount of the deep self-coding network is reduced. As shown in fig. 2, the preprocessing of the power data of the enterprise specifically includes the following steps:
s11: carrying out mean value completion on vacant numerical values in the electric power data of the enterprise;
specifically, there are many redundant data in hundreds of fields contained in the national grid big data center database, and these fields can be roughly divided into: the system comprises more than 10 major categories of chargeable electric charge level, real chargeable electric charge level, chargeable electric charge fluctuation condition, real chargeable electric charge fluctuation condition, chargeable electric charge trend condition, electric power consumption characteristics, overdue electric charge, electric charge payment level, electric power safety, illegal electric power consumption, service communication and the like. For classified power fields, we find that part of field information is still missing due to various reasons, in order to fill up the fields, and convert some level (such as "a", "B", "C", etc.) field information into numerical indexes, specifically, we design a level mapping table first, convert all level fields into numerical values, and after conversion, perform mean filling up on the missing numerical values, that is: the mean value of the field information recorded at other times is filled in at the vacant time.
S12: carrying out standardized processing on the power data of the enterprise;
specifically, for numerical variables, the min-max normalization formula is used as follows:
where max and min are the maximum and minimum values of the variables, respectively.
For level variables, for example: variables such as "a", "B", "risk", etc., translate the scale to a numerical value. Ranking the grades according to the risk index from low to high, for example, "a" indicates that the grade risk is lowest, then "a" ranks first, "risk" indicates that the grade risk is greatest, and ranks last, and the value corresponding to the ith grade is:
where n represents the number of levels.
S13: performing linear correlation analysis and time trend visualization analysis on the field subjected to data preprocessing;
specifically, in order to reduce the interference of redundant field information on a model and the additional calculation amount brought by the interference, linear correlation analysis is carried out on the field after data preprocessing, power data of 1000 enterprises are selected, the correlation among different field data is calculated, one of the sets of the relevant fields with large absolute values of correlation coefficients is selected for reservation, and overfitting of the model on the redundant field data is reduced; in order to further find effective fields which are beneficial to distinguishing, the power utilization data of illegal funding companies and normal companies can be visually analyzed in a time dimension, and the characteristic fields for effectively distinguishing the illegal funding companies from the normal enterprises are selected by utilizing a time trend graph visualization method.
S14: and performing feature extraction on the power utilization field which is obtained after linear correlation analysis and time trend visual analysis and is easy to distinguish illegal collection to obtain feature data.
Specifically, in the time trend visualization analysis, it can be found that the overall trend and fluctuation of the effective field of the illegal fundamentals are obviously different from those of the normal companies, so that the method extracts the characteristics of the mean value, the variance, the ascending frequency proportion, the descending frequency proportion, the ascending average gradient and the descending average gradient of the power utilization field expressed or recorded by using continuous numerical values in the database; for the rank field, the rank frequency ratio is adopted as its characteristic.
For fields with continuous numerical values, 2-dimensional plane analysis is carried out on every two dimensions of 6-dimensional features extracted from the continuous numerical values, the inter-class distance of company distribution of different industries is obtained through analysis, and similarly, for the fields with level power utilization, visual analysis is carried out by parallel coordinate axes, so that the industries of the fields are not greatly different. Therefore, the influence of the industry factors on the algorithm precision can be not considered in the subsequent modeling process of the enterprise electricity utilization data, namely, a special model does not need to be designed and trained independently for each industry.
S2: performing model training by taking the obtained characteristic data as input and output of a deep automatic coding network, and training the deep automatic coding network to obtain a power data detection model;
specifically, a Deep self-encoding network (DAE) structure model is shown in fig. 3, where the number of neuron nodes in the output layer is equal to the number of neuron nodes in the input layer, which is n, and the number of neuron nodes in the bottleneck layer is m. Through training of the DAE model, a compact representation of the input data may be obtained.
The deep automatic coding network training formula is shown as follows:
wherein the content of the first and second substances,
is the input vector of the input vector,
is the output vector, y ═ y1,y2,L,ym}∈RmM < n + s is the representation of the characteristics of x after DAE compression, f is the activation function, W and W 'are the weights of the encoding and decoding processes, respectively, and B' are the biases of the hidden and output layers.
The reconstruction Error, Mean Square Error (MSE), is defined to characterize the Error between the reconstructed data and the input data, and is expressed as follows:
the method of DAE network training is as follows: firstly, pre-training layer by utilizing a restricted Boltzmann machine to obtain an initial weight W0And W0' and initial bias B0And B0' then, model fine tuning is carried out through a back propagation algorithm to obtain final W, W ', B and B ', DAE network training is completed, model parameters are fixed, and a power data core detection model is generated.
S3: when new enterprise power data appear, the enterprise power data are repeatedly preprocessed, the trained power data detection model is used for reconstructing data, reconstruction errors are calculated, and whether the enterprises have illegal investment risk or not is judged by comparing the reconstruction errors with a set threshold value.
Specifically, the model is verified by using collected sample data containing normal enterprises and illegal funding enterprises. If the reconstruction error is smaller than the set threshold, the new enterprise power data is considered to be normal, and if the reconstruction error is larger than the set threshold, the new enterprise has illegal resource collection risk.
Further, as shown in fig. 4, an embodiment of the present invention further provides an illegal funding enterprise electricity consumption behavior identification device, which includes at least one processor and at least one memory storing at least one program, where when the at least one program is executed by the at least one processor, the at least one processor is enabled to implement the illegal funding enterprise electricity consumption behavior identification method as described above.
Compared with the prior art, the invention has the advantages that: compared with the existing method for judging whether the enterprise is an illegal collective resource enterprise or not through financial data, the method for judging the illegal collective resource enterprise through the enterprise power data has higher reliability and improves the reliability of judging the illegal collective resource risk of the enterprise. In addition, the illegal collection early warning monitoring scheme based on the enterprise power data improves the capacity of financial monitoring in the aspects of wide coverage, penetrating power, agility, traceability and the like, and has great practical significance for finding illegal collection institutions with high concealment. Furthermore, the invention adopts a deep self-coding network algorithm, deeply excavates the implicit power utilization rule which is difficult to find visually by illegal funding enterprises, and avoids the problem of insufficient data volume of negative samples (illegal funding enterprises).
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by those skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (9)
1. An enterprise electricity consumption behavior identification method is characterized by comprising the following steps:
preprocessing the power data of an enterprise to obtain characteristic data;
performing model training by taking the obtained characteristic data as input and output of a deep automatic coding network, and training the deep automatic coding network to obtain a power data detection model; and
when new enterprise power data appear, the enterprise power data are repeatedly preprocessed, the trained power data detection model is used for reconstructing data, reconstruction errors are calculated, and whether the enterprises have illegal investment risk or not is judged by comparing the reconstruction errors with a set threshold value.
2. The method for identifying the enterprise electricity consumption behavior according to claim 1, wherein the step of preprocessing the power data of the enterprise to obtain the characteristic data specifically comprises:
carrying out mean value completion on vacant numerical values in the electric power data of the enterprise;
carrying out standardized processing on the power data of the enterprise;
performing linear correlation analysis and time trend visualization analysis on the field subjected to data preprocessing;
and performing feature extraction on the electricity utilization field which is obtained after the linear correlation analysis and the time trend visual analysis and is easy to distinguish illegal collection to obtain feature data.
3. The method for identifying the enterprise electricity consumption behavior according to claim 2, wherein the step of performing mean value filling on the vacant numerical values in the enterprise electricity data specifically comprises the following steps: and designing a level mapping table, converting all level fields into numerical values, and filling the average values of field information recorded at other moments in the vacant moments after the conversion is finished.
4. The method for identifying the enterprise electricity consumption behavior according to claim 2, wherein the step of standardizing the power data of the enterprise specifically comprises: for numerical variables, the min-max normalization formula is adopted:
5. The method for identifying the enterprise electricity consumption behavior according to claim 2, wherein the step of performing linear correlation analysis and time trend visualization analysis on the field after data preprocessing specifically comprises the following steps: calculating the correlation among different field data, and selecting one from a correlation field set with a larger absolute value of a correlation coefficient for reservation, so as to reduce overfitting of a model to redundant field data; and selecting the characteristic fields for effectively distinguishing the illegal fundraising companies from the normal enterprises by using a time trend graph visualization method.
6. The method for identifying the enterprise electricity consumption behavior according to claim 2, wherein the step of extracting the characteristics of the electricity consumption field which is obtained after the linear correlation analysis and the time trend visual analysis and is easy to distinguish illegal funding specifically comprises the following steps of: extracting features of a mean value, a variance, an ascending frequency ratio, a descending frequency ratio, an ascending average gradient and a descending average gradient for the electricity utilization field represented or recorded by using continuous numerical values; for the rank field, the rank frequency ratio is adopted as its characteristic.
7. The method for identifying enterprise electricity consumption behaviors of claim 1, wherein the deep automatic coding network training formula is as follows:
wherein the content of the first and second substances,
is the input vector of the input vector,
is the output vector, y ═ y1,y2,L,ym}∈RmM < n + s is the representation of the characteristics of x after DAE compression, f is the activation function, W and W 'are the weights of the encoding and decoding processes, respectively, and B' are the biases of the hidden and output layers.
8. The method according to claim 1, wherein if the reconstruction error is smaller than a set threshold, the new enterprise power data is considered to be normal, and if the reconstruction error is larger than the set threshold, the new enterprise has an illegal funding risk.
9. An enterprise power consumption behavior recognition apparatus, comprising at least one processor and at least one memory storing at least one program that, when executed by the at least one processor, causes the at least one processor to implement the enterprise power consumption behavior recognition method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110666573.8A CN113313407A (en) | 2021-06-16 | 2021-06-16 | Enterprise power utilization behavior identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110666573.8A CN113313407A (en) | 2021-06-16 | 2021-06-16 | Enterprise power utilization behavior identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113313407A true CN113313407A (en) | 2021-08-27 |
Family
ID=77379008
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110666573.8A Pending CN113313407A (en) | 2021-06-16 | 2021-06-16 | Enterprise power utilization behavior identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113313407A (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108334591A (en) * | 2018-01-30 | 2018-07-27 | 天津中科智能识别产业技术研究院有限公司 | Industry analysis method and system based on focused crawler technology |
| CN108510006A (en) * | 2018-04-08 | 2018-09-07 | 重庆邮电大学 | A kind of analysis of business electrical amount and prediction technique based on data mining |
| CN108765004A (en) * | 2018-05-28 | 2018-11-06 | 贵州黔驰信息股份有限公司 | A method of user's electricity stealing is identified based on data mining |
| CN109524139A (en) * | 2018-10-23 | 2019-03-26 | 中核核电运行管理有限公司 | A kind of real-time device performance monitoring method based on equipment working condition variation |
| CN110929843A (en) * | 2019-10-29 | 2020-03-27 | 国网福建省电力有限公司 | Abnormal electricity consumption behavior identification method based on improved deep self-coding network |
| CN110942171A (en) * | 2019-09-12 | 2020-03-31 | 中电科新型智慧城市研究院有限公司 | Enterprise labor and resource dispute risk prediction method based on machine learning |
| CN111310275A (en) * | 2020-03-31 | 2020-06-19 | 南京智行信息科技有限公司 | Bridge disease big data analysis method |
| CN111738462A (en) * | 2020-06-08 | 2020-10-02 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN111861758A (en) * | 2020-07-22 | 2020-10-30 | 浙江中医药大学 | Botnet enterprise identification method and system based on full-connection neural network |
| CN112419040A (en) * | 2020-10-31 | 2021-02-26 | 国家电网有限公司 | Credit anti-fraud identification method, credit anti-fraud identification device and storage medium |
| CN112686491A (en) * | 2020-10-30 | 2021-04-20 | 国网浙江省电力有限公司台州供电公司 | Enterprise power data analysis method based on power consumption behavior |
| CN112712423A (en) * | 2020-12-29 | 2021-04-27 | 广州金融科技股份有限公司 | Suspected illegal fundraising item judgment method and device, computer equipment and storage medium |
| CN112749846A (en) * | 2021-01-14 | 2021-05-04 | 国网吉林省电力有限公司 | Big data platform-based financial cloud picture construction analysis method for electric power system |
-
2021
- 2021-06-16 CN CN202110666573.8A patent/CN113313407A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108334591A (en) * | 2018-01-30 | 2018-07-27 | 天津中科智能识别产业技术研究院有限公司 | Industry analysis method and system based on focused crawler technology |
| CN108510006A (en) * | 2018-04-08 | 2018-09-07 | 重庆邮电大学 | A kind of analysis of business electrical amount and prediction technique based on data mining |
| CN108765004A (en) * | 2018-05-28 | 2018-11-06 | 贵州黔驰信息股份有限公司 | A method of user's electricity stealing is identified based on data mining |
| CN109524139A (en) * | 2018-10-23 | 2019-03-26 | 中核核电运行管理有限公司 | A kind of real-time device performance monitoring method based on equipment working condition variation |
| CN110942171A (en) * | 2019-09-12 | 2020-03-31 | 中电科新型智慧城市研究院有限公司 | Enterprise labor and resource dispute risk prediction method based on machine learning |
| CN110929843A (en) * | 2019-10-29 | 2020-03-27 | 国网福建省电力有限公司 | Abnormal electricity consumption behavior identification method based on improved deep self-coding network |
| CN111310275A (en) * | 2020-03-31 | 2020-06-19 | 南京智行信息科技有限公司 | Bridge disease big data analysis method |
| CN111738462A (en) * | 2020-06-08 | 2020-10-02 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN111861758A (en) * | 2020-07-22 | 2020-10-30 | 浙江中医药大学 | Botnet enterprise identification method and system based on full-connection neural network |
| CN112686491A (en) * | 2020-10-30 | 2021-04-20 | 国网浙江省电力有限公司台州供电公司 | Enterprise power data analysis method based on power consumption behavior |
| CN112419040A (en) * | 2020-10-31 | 2021-02-26 | 国家电网有限公司 | Credit anti-fraud identification method, credit anti-fraud identification device and storage medium |
| CN112712423A (en) * | 2020-12-29 | 2021-04-27 | 广州金融科技股份有限公司 | Suspected illegal fundraising item judgment method and device, computer equipment and storage medium |
| CN112749846A (en) * | 2021-01-14 | 2021-05-04 | 国网吉林省电力有限公司 | Big data platform-based financial cloud picture construction analysis method for electric power system |
Non-Patent Citations (1)
| Title |
|---|
| 张春梅等: "基于集成学习的电网用电量预测***研究", 《现代计算机(专业版)》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113657545B (en) | User service data processing method, device, equipment and storage medium | |
| CN109657947B (en) | Enterprise industry classification-oriented anomaly detection method | |
| CN110852856B (en) | Invoice false invoice identification method based on dynamic network representation | |
| CN110751557B (en) | Abnormal fund transaction behavior analysis method and system based on sequence model | |
| CN112150298B (en) | Data processing method, system, device and readable medium | |
| CN111738843B (en) | Quantitative risk evaluation system and method using running water data | |
| CN110222733B (en) | High-precision multi-order neural network classification method and system | |
| CN111192133A (en) | Method and device for generating risk model after user loan and electronic equipment | |
| CN113240527A (en) | Bond market default risk early warning method based on interpretable machine learning | |
| CN116861331A (en) | Expert model decision-fused data identification method and system | |
| CN115081641A (en) | Model training method, estimation result prediction method, device and storage medium | |
| CN115439150B (en) | Bargaining method and device for data asset | |
| CN113313407A (en) | Enterprise power utilization behavior identification method and device | |
| CN115271442A (en) | Modeling method and system for evaluating enterprise growth based on natural language | |
| CN115618297A (en) | Method and device for identifying abnormal enterprise | |
| CN117455681A (en) | Service risk prediction method and device | |
| Yang et al. | An evidential reasoning rule-based ensemble learning approach for evaluating credit risks with customer heterogeneity | |
| KR20220074327A (en) | Loan regular auditing system using artificia intellicence | |
| CN116051296B (en) | Customer evaluation analysis method and system based on standardized insurance data | |
| CN117972595A (en) | Method, system, device and medium for analyzing electric charge abnormality | |
| CN116976737A (en) | Method, system and computer readable storage medium for evaluating enterprise scientific creation capability | |
| Shen et al. | Modelling the predictive performance of credit scoring | |
| CN116739756A (en) | Financial data processing method, apparatus, device, medium and program product | |
| Zhang et al. | Application of Adaboost Algorithm in Enterprise Financial Risk Analysis Model | |
| CN118333739A (en) | Method for constructing retail credit risk prediction model and retail credit business Scoremult model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210827 |
|
| RJ01 | Rejection of invention patent application after publication |