CN117039850A - Abnormal electricity consumption analysis method and system based on space-time ground feature characteristics - Google Patents
Abnormal electricity consumption analysis method and system based on space-time ground feature characteristics Download PDFInfo
- Publication number
- CN117039850A CN117039850A CN202310850043.8A CN202310850043A CN117039850A CN 117039850 A CN117039850 A CN 117039850A CN 202310850043 A CN202310850043 A CN 202310850043A CN 117039850 A CN117039850 A CN 117039850A
- Authority
- CN
- China
- Prior art keywords
- electricity
- electricity consumption
- data
- user
- users
- 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
- 230000005611 electricity Effects 0.000 title claims abstract description 478
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 120
- 238000004458 analytical method Methods 0.000 title claims abstract description 21
- 238000012544 monitoring process Methods 0.000 claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000013499 data model Methods 0.000 claims abstract description 7
- 230000005612 types of electricity Effects 0.000 claims abstract description 6
- 238000005065 mining Methods 0.000 claims description 15
- 230000006399 behavior Effects 0.000 claims description 14
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000000306 recurrent effect Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 4
- 230000009191 jumping Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 6
- 231100000656 threshold model Toxicity 0.000 abstract description 2
- 238000010924 continuous production Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Computational Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Evolutionary Biology (AREA)
- Remote Sensing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Algebra (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Probability & Statistics with Applications (AREA)
- Human Computer Interaction (AREA)
- Computing Systems (AREA)
- Operations Research (AREA)
- Bioinformatics & Computational Biology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
Abstract
The invention provides a method and a system for abnormal electricity analysis based on space-time ground feature characteristics, wherein the method comprises the following steps: implementing multistage grid division on the monitoring area; collecting and sorting historical electricity consumption information of users in an electricity consumption monitoring area, wherein the historical electricity consumption information comprises user types, electricity consumption behavior characteristics of the users and historical weather information; constructing an electricity data model of the regional users based on regional space-time data and historical electricity information, and realizing prediction of the electricity data of the regional users; setting thresholds of power consumption of users of different categories and power consumption of users in different weather conditions in different periods of time; and according to the predicted electricity consumption, analyzing whether the electricity consumption is within a reasonable range of a calibrated threshold value, if so, not using the electricity abnormally, otherwise, recording the electricity abnormally. The method and the device support the quick indexing of the regional abnormal electricity users by combining the mode of space multi-level grid division; through the calibration of different types of electricity utilization users, different threshold models are set, and the accuracy of abnormal electricity utilization detection is improved.
Description
Technical Field
The invention relates to the field of artificial intelligence and the field of geographic information science, in particular to a method and a system for abnormal electricity analysis based on space-time ground feature characteristics.
Background
Regional power consumption real-time monitoring is one of the core works of power consumption management departments. Accurate and instant regional power consumption monitoring can provide effective data support and auxiliary decision for power supply dispatching and power consumption risk prevention and control. The existing power consumption monitoring system can monitor the whole power consumption situation of the area on a large scale, and on the other hand, single-user power consumption monitoring of a user focusing attention is realized. However, in the process of monitoring the single-user power consumption of the heavy-point user, intelligent early warning with different scales and different thresholds is not realized according to the difference of user types (industrial power consumption or residential power consumption); and the regional multistage hierarchical monitoring is realized without according to the geospatial information, so that the monitoring and decision-making efficiency is improved. In summary, the existing regional power consumption real-time monitoring cannot fully meet the intelligent supervision requirement of the power utilization department, and the monitoring method and the monitoring process have a larger optimization space.
Disclosure of Invention
The invention mainly solves the technical problems existing in the prior art, and provides a method and a system for abnormal electricity analysis based on space-time ground feature characteristics. The method supports the rapid indexing of the regional abnormal electricity users by combining the mode of space multi-level grid division; through the calibration of different types of electricity utilization users, different threshold models are set, and the accuracy of abnormal electricity utilization detection is improved.
An abnormal electricity analysis method based on space-time ground feature characteristics comprises the following steps:
step one, realizing multistage grid division on a monitoring area;
step two, collecting and sorting historical electricity consumption information of users in an electricity consumption monitoring area, wherein the historical electricity consumption information comprises user types, user electricity consumption behavior characteristics and historical weather information;
thirdly, constructing an electricity data model of the regional users based on regional space-time data and historical electricity consumption, and realizing prediction of the electricity data of the regional users;
setting thresholds of power consumption of users in different categories, power consumption of users in different periods of time and power consumption of users in different weather conditions;
and fifthly, analyzing whether the power consumption is within a reasonable range of the threshold value calibrated in the fourth step according to the power consumption predicted in the third step, if the power consumption is within the reasonable range of the threshold value calibrated in the fourth step, not using the power consumption abnormally, otherwise recording the power consumption abnormally.
Further, the first step specifically includes:
step 1.1, in ArcGIS software, a base map layer is newly built, and the electricity consumption monitoring area range is defined;
step 1.2, collecting user data in the range of an electricity consumption monitoring area, newly establishing an attribute map layer in ArcGIS software, and storing the position of an electricity consumption user in the newly established attribute map layer in the form of point elements;
step 1.3, dividing the electricity consumption monitoring area into square grid areas with the same size;
step 1.4, counting the number of users in the square grid, and if the number of users is higher than a set threshold value, continuously dividing the grid into small grids;
step 1.5, repeatedly dividing small grids according to the step 1.4, and ensuring that the number of users in each grid area is not more than 10;
and step 1.6, numbering all grids to realize accurate indexing to the corresponding grid area according to the numbering.
Further, the second step specifically includes:
step 2.1, collecting historical electricity utilization information of all users in an electricity utilization monitoring area, wherein the historical electricity utilization information comprises the date of the last three years and the electricity utilization data of the time period;
step 2.2, dividing users in grids with all numbers in step 1.6 into two types by combining the statistical data of the power utilization departments: industrial electricity users and residential electricity users;
step 2.3, associating all the statistical data with the point elements in the attribute layer in the step one by one, namely associating the user information with the user points in the space-time information;
and 2.4, collecting time-sharing weather data of the area for nearly three years, and recording.
Further, step 2.1 further includes: if the acquired data is partially missing, an average interpolation method is adopted, and the average value is obtained according to the power consumption data of the same time period of the previous and the next date, and the method is specifically described as follows:
a) If the electricity consumption data of a user is lost in the last three years, firstly, checking whether the electricity consumption data loss period of the user is no electricity consumption behavior or data acquisition is omitted;
b) If no electricity consumption behavior exists, marking the electricity consumption data of the period as 0, terminating the subsequent step, and directly jumping to the step 2.2;
c) If the user electricity consumption data is missing because the user electricity consumption data is not acquired or is not acquired, taking the average value of the electricity consumption data of the same time period of the previous day and the next day;
d) If the electricity data of the user in a certain period are not collected or are not collected for a plurality of continuous days, the electricity data of the user in the period are marked as abnormal, and are marked as-1.
Further, the third step specifically includes:
step 3.1, constructing a deep learning RNN prediction model by using a universal recurrent neural network RNN;
step 3.2, predicting the total power consumption of each grid by using the RNN prediction model constructed in step 3.1 according to the grid sequence constructed in step 1.6, wherein the method specifically comprises the following steps:
a) Firstly, carrying out time-division addition on the historical electricity consumption data of each grid to obtain the sum data of the electricity consumption of each grid per hour;
b) Then, in each grid, carrying out time-period addition on historical electricity consumption data of each user to obtain historical time-period user data of each user in the grid;
c) Predicting the current power consumption data of each grid by using the RNN prediction model constructed in the step 3.1, if the prediction data are consistent with the real-time data, skipping the step 3.2, and continuing the step 3.3;
d) If the predicted data and the real-time data are inconsistent, predicting the electricity consumption of users in the grid one by using an RNN prediction model, and marking the electricity users inconsistent with the real-time electricity consumption data;
step 3.3, when the predicted data marked in the step 3.2 do not accord with the real-time user data, backtracking and retrieving the historical electricity consumption data of the user, and inquiring whether the situation of data insufficiency exists or not: if the power consumption data is missing because the power consumption data is not acquired or is not acquired, taking the average value of the power consumption data of the same time period of the previous day and the next day; if not, the step is ended and recorded as an abnormal condition.
Further, the step four specifically includes:
step 4.1, for industrial electricity users, the specific threshold calibration method is described as follows:
a) Dividing industrial electricity into two types of electricity consumption in holidays and non-holiday;
b) For electricity consumption in non-holidays, the calibration threshold is 10%, namely if the difference between the predicted electricity consumption and the actual electricity consumption is more or less than 10%, the power consumption is calibrated as abnormal electricity consumption;
c) The electricity consumption conditions of holidays of industrial electricity users are further divided into normalized electricity consumption and non-normalized electricity consumption;
d) The normalized electricity consumption is compared with the time period of the non-holiday at ordinary times, if the electricity consumption difference is smaller than 20%, the normalized electricity consumption is defined as normalized electricity consumption, and the same threshold value calibration as the non-holiday is adopted;
e) The rest conditions of electricity consumption in holidays are marked as abnormal electricity consumption; if the electricity consumption is 20% of the electricity consumption in the same period of the non-holiday at ordinary times, marking the abnormal electricity consumption; if the electricity consumption is far smaller than the electricity consumption in the same period of the usual non-holiday, marking as scattered electricity consumption, wherein the electricity consumption data is not used as electricity consumption evaluation data; wherein "much less than" is defined as less than 1/10 of the amount of electricity used at ordinary times;
step 4.2, for domestic electricity consumption, the specific threshold calibration method is described as follows:
a) Setting a threshold value to be 30%, if the difference between the predicted power consumption and the real-time power consumption is less than 30%, calibrating the power consumption to be normal power consumption, and continuing the subsequent step 4.2;
b) If the predicted electricity consumption and the real-time electricity consumption are more than 30% and less than 100%, calculating the temperature difference between the current predicted electricity consumption period and the historical data at the same time, and if the temperature difference is more than 8 ℃, calibrating as: the power consumption of the user is changed due to weather change, so that abnormal power consumption is avoided;
c) If the temperature is lower than 8 ℃, reading real-time electricity consumption data of the resident electricity consumption user, and if the real-time electricity consumption is lower than 1 kilowatt hour, calibrating that the resident electricity consumption is lower, the fluctuation range is larger, and the resident electricity consumption is not abnormal electricity consumption;
d) Locating the resident user as an abnormal electricity utilization situation in real time according to the rest of the conditions defined in the steps b) and c).
Further, the fifth step specifically includes:
step 5.1, analyzing the type of each electricity user in each numbered grid as an industrial electricity user or a resident electricity user;
step 5.2, predicting the total power consumption of each numbered grid in each period, and comparing the total power consumption with real-time power consumption data;
step 5.3, if the electricity consumption of a certain grid is abnormal in a certain period, indicating that the grid has potential risks of abnormal electricity consumption;
step 5.4, aiming at all power utilization users of the grid, detecting whether the real-time power utilization of each user is abnormal or not by adopting the dynamic threshold calibration method set in the step four;
step 5.5, for the user marked as abnormal electricity consumption, the user needs to be verified manually, and if the user is actually abnormal in electricity consumption, the mining work of the abnormal electricity consumption user related to the invention is completed; if the user power consumption is verified to be normal, the real-time power consumption data is updated to the historical power consumption data base so as to improve the accuracy of subsequent prediction.
An abnormal electricity consumption analysis device based on space-time ground feature, comprising:
the geographic grid division module is used for realizing multistage grid division on the monitoring area and ensuring that the number of users in each grid is relatively fixed;
the regional historical electricity consumption information arrangement module is used for collecting and arranging historical electricity consumption information of users in the electricity consumption monitoring region, wherein the historical electricity consumption information comprises user types, user electricity consumption behavior characteristics and historical weather information;
the regional user electricity consumption data prediction module is used for constructing a regional user electricity consumption data model based on regional space-time data and historical electricity consumption information, and realizing prediction of regional user electricity consumption data;
the dynamic threshold calibration module is used for setting different thresholds and realizing the dynamic threshold setting of the power consumption of users in different categories, the power consumption of users in different time periods and the power consumption of users in different weather conditions;
and the regional abnormal electricity utilization user mining module is used for analyzing whether the regional abnormal electricity utilization user mining module is in a reasonable range of the threshold value calibrated by the dynamic threshold value calibration module according to the electricity utilization quantity predicted by the regional user electricity utilization data prediction module, if the regional abnormal electricity utilization user mining module is in the reasonable range of the threshold value calibrated by the dynamic threshold value calibration module, the regional abnormal electricity utilization user mining module is not used for abnormal electricity utilization, and otherwise, the regional abnormal electricity utilization is marked as abnormal electricity utilization.
Further, the area history electricity consumption information arrangement module is specifically configured to: collecting historical electricity consumption data of all users in an electricity consumption monitoring area, wherein the historical electricity consumption data comprise the date of minute and the time of minute of the last three years; users within all numbered grids are classified into two categories: industrial electricity users and residential electricity users; associating all the statistical data with the point elements in the attribute layer one by one, namely associating the user information with the user points in the space-time information; and collecting the time-sharing weather data of the region for nearly three years, and recording to finish the arrangement of the historical electricity utilization information of the region.
Further, the regional abnormal electricity user mining module is specifically configured to: for each numbered grid, analyzing that the respective type of the electricity utilization user in the grid is an industrial electricity utilization user or a resident electricity utilization user; for each numbered grid, predicting the total electricity consumption of the grid in each period, comparing the total electricity consumption with real-time electricity consumption data, and if the electricity consumption of a certain grid in a certain period is abnormal, indicating that the grid has potential risk of abnormal electricity consumption; aiming at all power utilization users of the grid, a dynamic threshold calibration method set by a dynamic threshold calibration module is adopted to detect whether the power utilization of each user is abnormal in real time, and for the users calibrated to be abnormal power utilization, the manual verification is needed, and if the users are indeed abnormal power utilization, the mining work of the abnormal power utilization users related to the invention is completed; if the user is verified to be normal in electricity consumption, the real-time electricity consumption data is updated to the historical electricity consumption database.
The invention has the following beneficial effects:
1. the regional electricity users are displayed on the map, and grid division (multi-level grid) with relatively fixed scale of the number of the electricity users (not more than 10 users) is realized in a space multi-level grid mode, so that the subsequent abnormal electricity users can be found out accurately;
2. dividing power utilization users in the grid into two types of industrial power utilization users and residential power utilization users, and formulating different power utilization prediction methods through different power utilization behavior characteristics of the two types of users;
3. the method has the advantages that by adopting a dynamic threshold mode, real-time electricity utilization of industrial electricity users and residential electricity users is evaluated whether abnormal or not, and different criteria such as thresholds, judgment factors (such as residential electricity needs to consider weather conditions) and the like are adopted for the two evaluations, so that electricity utilization characteristics of different users can be effectively fitted, and the accuracy of abnormal electricity detection is improved;
4. in the RNN electricity consumption prediction model, the influence of weather data is put in a threshold calibration step without considering the weather data of time intervals, so that the convergence effect of the electricity consumption prediction model is effectively improved, and meanwhile, the influence factors of weather on the electricity consumption of residents are fully utilized.
Drawings
FIG. 1 is a flow chart of the abnormal electricity analysis method based on space-time clutter characteristics of the present invention;
FIG. 2 is a block diagram of an anomaly power analysis system based on space-time clutter features of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention. The whole system is realized in a software mode.
Referring to fig. 1, an embodiment of the present invention provides a method for abnormal electricity analysis based on space-time ground feature, including the following steps:
step one, a geographic grid dividing step, which is used for realizing multistage grid division on a monitoring area, ensuring that each grid has a relatively fixed number of users, and specifically comprises the following steps:
and 1.1, creating a base map layer in ArcGIS software, and delineating the electricity monitoring area.
And 1.2, collecting user data in the power consumption monitoring area, newly building an attribute layer in ArcGIS software, and storing the position of the power consumption user in the newly built attribute layer in the form of point elements.
And 1.3, dividing the electricity consumption monitoring area into square grid areas with the same size, wherein the grid size can be manually set according to factors such as the density of electricity consumption users, the size of the whole electricity consumption monitoring area and the like, for example, the area is grid into 8x8 grids.
And 1.4, counting the number of users in the square grid, and if the number of users is higher than a set threshold value, continuously dividing the grid into small grids, such as 8x8 small grids.
Step 1.5, repeatedly dividing small grids according to the step 1.4, and ensuring that the number of users in each grid area is not more than 10.
And step 1.6, numbering all grids, secondary grids (small grids divided again), tertiary grids and the like, so as to ensure that corresponding grid areas can be accurately indexed according to the numbers.
Step two, a historical electricity consumption information collecting step of collecting and collecting historical electricity consumption information of users in an electricity consumption monitoring area, wherein the historical electricity consumption information comprises user types, user electricity consumption behavior characteristics, historical weather information and the like, and the method comprises the following specific implementation steps of:
step 2.1, collecting historical electricity utilization information (user category, user electricity utilization behavior characteristics and historical weather information) of all users in an electricity utilization monitoring area, wherein the historical electricity utilization information comprises the date of last three years and the electricity utilization data of time intervals; in particular, if some data is missing, an average interpolation method is adopted to obtain an average value according to the electricity consumption data of the same time period of the previous date and the next date, and the method is specifically described as follows:
a) If the electricity consumption data of a user is lost in the last three years, the user electricity consumption data loss period needs to be verified, and the user does not have electricity consumption behavior or is missing in data acquisition.
b) If the electricity consumption behavior is not generated, marking the electricity consumption data of the period as 0, terminating the subsequent steps, and directly jumping to the step 2.2.
c) If the user electricity consumption data is missing because the user electricity consumption data is not acquired or is not acquired, taking the average value of the electricity consumption data of the same time period of the previous day and the next day;
d) If the electricity data of the user in a certain period are not collected or are not collected for a plurality of continuous days, the electricity data of the user in the period are marked as abnormal, and are marked as-1.
Step 2.2, dividing users in grids with all numbers in step 1.6 into two types by combining the statistical data of the power utilization departments: industrial electricity users and residential electricity users.
And 2.3, associating all the statistical data with the point elements in the attribute layer in the first step one by one, namely associating the user information with the user points in the space-time information.
And 2.4, collecting the time-sharing weather data of the region, mainly the air temperature data, of the last three years, and recording.
So far, finishing the regional historical electricity consumption information, wherein the data comprise: 1) The time-period data and the instant weather data of the electricity consumption of all users in the area in the last three years; 2) The user type is industrial electricity users or resident electricity users; 3) The electricity usage data is associated with a spatiotemporal location in a map of the user.
Step three, a regional user electricity consumption data prediction step, which is used for constructing a regional user electricity consumption data model based on regional space-time data and historical electricity consumption information, so as to realize the prediction of regional user electricity consumption data, and specifically comprises the following steps:
step 3.1, constructing a deep learning RNN prediction model by using a universal recurrent neural network (Recurrent Neural Network, RNN) for predicting new electricity utilization data based on historical electricity utilization data;
step 3.2, predicting the total power consumption of each grid by using the RNN prediction model constructed in step 3.1 according to the grid sequence constructed in step 1.6, wherein the method is described in detail as follows:
a) Firstly, carrying out time-division addition on the historical electricity consumption data of each grid to obtain the sum data of the electricity consumption of each grid per hour;
b) Then, in each grid (comprising a plurality of electricity utilization users), carrying out time-period addition on historical electricity utilization data of each user to obtain historical time-period user data of each user in the grid;
c) Predicting the current power consumption data of each grid by using the RNN prediction model constructed in the step 3.1, if the prediction data and the real-time data accord with each other (indicating that the prediction is accurate), skipping the step 3.2, and continuing the step 3.3;
d) If the predicted data and the real-time data are inconsistent, the predicted data are inconsistent with the data monitored by the power supply department in real time, the RNN prediction model is used for predicting the power consumption of users in the grid one by one, and the power consumption users inconsistent with the real-time power consumption data are marked;
step 3.3, when the predicted data marked in step 3.2 does not match the real-time user data, backtracking and retrieving the historical electricity consumption data of the user, and inquiring whether the data insufficiency condition in step 2.1 c) exists or not: if the power consumption data in the step 2.1 c) are missing because the power consumption data are not acquired or are acquired in a missing mode, taking the power consumption data of the same time period of the previous day and the next day to average; if not, the step is ended and recorded as an abnormal condition.
Thus, the prediction of the real-time electricity consumption data is completed, and the abnormal situation is recorded.
And step four, a dynamic threshold calibration step, which is used for setting different thresholds and realizing the dynamic threshold setting of the power consumption of users in different categories, the power consumption of users in different time periods and the power consumption of users in different weather conditions. The dynamic threshold is calibrated, so that on one hand, the false alarm rate can be effectively improved, namely, normal electricity utilization is marked as abnormal electricity utilization; meanwhile, the 'missing report rate' (missing report is that the abnormal electricity consumption condition is not detected) can be reduced, and finally the monitoring effect of the abnormal electricity consumption user is improved. The specific implementation steps are as follows:
step 4.1, for industrial electricity users, considering that the industrial electricity periodicity is strong (i.e. the electricity consumption period and the electricity consumption amount are consistent every day), the specific threshold calibration method is described as follows:
a) Dividing industrial electricity into two types of electricity consumption in holidays and non-holiday;
b) For electricity consumption in non-holidays, the calibration threshold is 10%, namely if the difference between the predicted electricity consumption and the actual electricity consumption is more or less than 10%, the power consumption is calibrated as abnormal electricity consumption;
c) The electricity consumption conditions of holidays of industrial electricity users are further divided into normalized electricity consumption and non-normalized electricity consumption;
d) The normalized electricity consumption is compared with the time period of the non-holiday at ordinary times, if the electricity consumption difference is smaller than 20%, the normalized electricity consumption is defined as normalized electricity consumption (holiday overtime and industrial continuous production), and the normalized electricity consumption is calibrated by adopting the same threshold value as the non-holiday (10%);
e) The rest conditions of electricity consumption in holidays are marked as abnormal electricity consumption; if the electricity consumption is 20% of the electricity consumption in the same period of the non-holiday at ordinary times, marking the abnormal electricity consumption; if the electricity consumption is far smaller than the electricity consumption in the same period of the usual non-holiday, marking as scattered electricity consumption, wherein the electricity consumption data is not used as electricity consumption evaluation data; wherein "much less than" is defined as less than 1/10 of the amount of electricity used at ordinary times;
step 4.2, for domestic electricity, considering that the electricity randomness is stronger (compared with industrial electricity), the specific threshold calibration method is described as follows:
a) Setting a threshold value to be 30%, if the difference between the predicted power consumption and the real-time power consumption is less than 30%, calibrating the power consumption to be normal power consumption, and continuing the subsequent step 4.2;
b) If the predicted electricity consumption and the real-time electricity consumption are more than 30% and less than 100%, calculating the temperature difference between the current predicted electricity consumption period and the historical data at the same time, and if the temperature difference is more than 8 ℃, calibrating as: the power consumption of the user is changed due to weather change, so that abnormal power consumption is avoided;
c) If the temperature is lower than 8 ℃, reading real-time electricity consumption data of the resident electricity consumption user, and if the real-time electricity consumption is lower than 1 kilowatt hour, calibrating that the resident electricity consumption is lower, the fluctuation range is larger, and the resident electricity consumption is not abnormal electricity consumption;
d) Locating the resident user as an abnormal electricity utilization situation in real time according to the rest of the conditions defined in the steps b) and c).
The thresholds in the step 4.1 and the step 4.2 can be combined with specific electricity utilization areas and area electricity utilization characteristics to adopt different calibration thresholds; meanwhile, the threshold value needs to be updated at regular time, so that the applicability of the calibration of the abnormal electricity consumption threshold value in different seasons is improved.
And fifthly, analyzing whether the power consumption is within a reasonable range of the threshold value calibrated in the fourth step according to the power consumption predicted in the third step, if the power consumption is within the reasonable range of the threshold value calibrated in the fourth step, not using the power consumption abnormally, otherwise recording the power consumption abnormally. The specific implementation steps are as follows:
step 5.1, analyzing the type of each electricity user in each numbered grid as an industrial electricity user or a resident electricity user;
step 5.2, predicting the total power consumption of each numbered grid in each period, and comparing the total power consumption with real-time power consumption data;
step 5.3, if the electricity consumption of a certain grid is abnormal (in order to avoid missing detection, the whole electricity consumption threshold of the grid is set to be 5%), indicating that the grid has potential risk of abnormal electricity consumption;
step 5.4, aiming at all power utilization users (not more than 10 users) of the grid, detecting whether the real-time power utilization of each user is abnormal or not by adopting the dynamic threshold calibration method set in the step four;
step 5.5, for the user marked as abnormal electricity consumption, the user needs to be verified manually, and if the user is actually abnormal in electricity consumption, the mining work of the abnormal electricity consumption user related to the invention is completed; if the user power consumption is verified to be normal, the real-time power consumption data is updated to the historical power consumption database designed in the second step so as to improve the accuracy of subsequent prediction.
Referring to fig. 2, the embodiment of the invention further provides an abnormal electricity analysis device based on space-time feature, which includes:
the geographic meshing module 10 is used for realizing multistage meshing of the monitoring area and ensuring that a relatively fixed number of users exist in each mesh. Firstly, in ArcGIS software, a base map layer is newly built and is used for delineating the range of an electricity monitoring area; then, collecting user data in an electricity consumption monitoring area, newly establishing an attribute layer in ArcGIS software, and storing the position of an electricity consumption user in the newly established attribute layer in the form of point elements; then dividing the electricity utilization area into square grid areas with the same size, wherein the grid size can be manually set according to factors such as the density of electricity utilization users and the size of the whole electricity utilization monitoring area, for example, the area is grid into 8x8 grids; meanwhile, counting the number of users in 8x8 grids, if the number of users is too large (higher than a threshold value), continuously dividing the grids into small grids, such as small grids with 8x 8; finally, numbering all grids, secondary grids (small grids divided again), tertiary grids and the like, and ensuring that the corresponding grid areas can be accurately indexed according to the numbers.
The regional historical electricity consumption information arrangement module 20 is used for collecting and arranging historical electricity consumption information of users in the electricity consumption monitoring region, including user types, user electricity consumption behavior characteristics and historical weather information. Firstly, collecting historical electricity consumption data of all users in an electricity consumption monitoring area, wherein the historical electricity consumption data comprise the date of the last three years and the electricity consumption data of the time period; particularly, if part of the data required to be acquired in the step is missing, an average interpolation method is adopted, and the average value of the data is obtained according to the power consumption data of the same time period of the previous and subsequent dates; next, the users within all numbered grids are divided into two categories: industrial electricity users and residential electricity users; then, all the statistical data are associated with the point elements in the attribute layer in the module 10 one by one, namely, the user information is associated with the user points in the space-time information; finally, the regional historical electricity consumption information is finished by collecting the regional time-sharing weather data, mainly air temperature data, of the last three years and recording.
And the regional user electricity consumption data prediction module 30 is used for constructing a regional user electricity consumption data model based on regional space-time data and historical electricity consumption information and realizing prediction of regional user electricity consumption data. Firstly, constructing a general recurrent neural network (Recurrent Neural Network, RNN) to construct a deep learning prediction framework for predicting new electricity utilization data based on historical electricity utilization data; then, carrying out time-period addition on the historical electricity consumption data of each grid to obtain the sum data of the electricity consumption of each grid per hour, and carrying out time-period addition on the historical electricity consumption data of each user in each grid (comprising a plurality of electricity consumption users) to obtain the historical time-period user data of each user in the grid; then, predicting the current electricity consumption data of each grid, if the predicted data and the real-time data are consistent (indicating that the prediction is accurate), predicting the electricity consumption of users in the grid one by using an RNN prediction model, and marking the electricity consumption users inconsistent with the real-time electricity consumption data; finally, the prediction of the real-time electricity consumption data is completed, and the abnormal situation is recorded.
The dynamic threshold calibration module 40 is configured to set different thresholds, and realize dynamic threshold setting for different types of users, users in different periods of time, and different weather conditions. For industrial electricity users, considering that the industrial electricity has stronger periodicity (namely, the electricity consumption period and the electricity consumption are consistent every day), firstly, the industrial electricity needs to be divided into two conditions of holiday electricity consumption and non-holiday electricity consumption, and for the non-holiday electricity consumption, the calibration threshold is 10 percent, namely, if the difference between the predicted electricity consumption and the actual electricity consumption is more than or less than 10 percent, the abnormal electricity consumption is calibrated; the electricity consumption conditions of holidays of industrial electricity users are further divided into normalized electricity consumption and non-normalized electricity consumption; the normalized electricity consumption is compared with the time period of the non-holiday at ordinary times, if the electricity consumption difference is smaller than 20%, the normalized electricity consumption is defined as normalized electricity consumption (holiday overtime and industrial continuous production), and the normalized electricity consumption is calibrated by adopting the same threshold value as the non-holiday (10%); the rest conditions of electricity consumption in holidays are marked as abnormal electricity consumption; if the electricity consumption is 20% of the electricity consumption in the same period of the non-holiday at ordinary times, marking the abnormal electricity consumption; if the electricity consumption is far smaller than the electricity consumption in the same period of the usual non-holiday, marking as scattered electricity consumption, wherein the electricity consumption data is not used as electricity consumption evaluation data; where "much less than" is defined as less than 1/10 of the amount of electricity used at ordinary times. For domestic electricity consumption, considering that the electricity consumption randomness is strong (compared with industrial electricity consumption), firstly, setting a threshold value to be 30%, and if the difference between the predicted electricity consumption and the real-time electricity consumption is less than 30%, calibrating to be normal electricity consumption; then, if the predicted electricity consumption and the real-time electricity consumption are more than 30 percent and less than 100 percent, calculating the temperature difference between the current predicted electricity consumption period and the historical data at the same period, and if the temperature difference is more than 8 ℃, calibrating as follows: the power consumption of the user is changed due to weather change, so that abnormal power consumption is avoided; if the temperature output is less than 8 ℃, the real-time electricity consumption data of the resident electricity consumption user is read, and if the real-time electricity consumption is less than 1 kilowatt hour, the resident electricity consumption is calibrated to be smaller, the fluctuation range is larger, and the resident electricity consumption is not abnormal electricity consumption; finally, the rest conditions mark the situation that the resident user is abnormal electricity.
The regional abnormal electricity consumption user discovery module 50 analyzes whether the regional abnormal electricity consumption user is within a reasonable range of the threshold value calibrated by the dynamic threshold value calibration module 40 according to the electricity consumption amount predicted by the regional user electricity consumption data prediction module 30, if the regional abnormal electricity consumption user is within the reasonable range of the calibrated threshold value, the regional abnormal electricity consumption user discovery module is not abnormal electricity consumption, otherwise, the regional abnormal electricity consumption user discovery module is marked as abnormal electricity consumption. The module has the main function of being used for grading abnormal electricity utilization users in the estimation area, and improving the identification accuracy of the abnormal electricity utilization users. Firstly, for each numbered grid, analyzing the respective types of electricity users (industrial electricity users or residential electricity users) in the grid; then, for each numbered grid, predicting the total electricity consumption of the grid in each period, comparing the total electricity consumption with real-time electricity consumption data, and if the electricity consumption of a certain grid in a certain period is abnormal (in order to avoid missing detection, the whole electricity consumption threshold of the grid is set to be 5%), indicating that the grid has potential risk of abnormal electricity consumption; then, aiming at all power utilization users (not more than 10 users) of the grid, detecting whether the power utilization of each user is abnormal in real time by adopting a dynamic threshold calibration method set by a dynamic threshold calibration module 40, and if the power utilization is abnormal, completing mining work of the abnormal power utilization users related to the invention; if the user power consumption is verified to be normal, the real-time power consumption data is updated to the historical power consumption data base so as to improve the accuracy of subsequent prediction.
The embodiment of the invention selects longitude and latitude coordinates of Qingshan areas in Wuhan city as follows:
[[114.425864,30.643199],
[114.434452,30.643199],
[114.434452,30.637327],
[114.425864,30.637327]]
the rectangular area of (1) is subjected to field verification, the area comprises a residential area and an industrial park, the total number of the areas is 131, the area is divided into 3 levels of grids, 39 cases of abnormal electricity consumption are detected through continuous 90-day 3-time period (9:00-10:00, 15:00-16:00, 20:00-21:00) time period test, and the areas are subjected to field verification of a power supply department. The theory, the method and the system related to the invention are further verified, the abnormal electricity utilization detection capability is stronger, and potential safety hazards of electricity utilization can be effectively found.
Those of skill would further appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the various illustrative components and steps have been described generally in terms of functionality in the foregoing description to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in random access memory, read only memory, electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
It should be understood that various other corresponding changes and modifications can be made by one skilled in the art according to the technical concept of the present invention, and all such changes and modifications should fall within the scope of the claims of the present invention.
The foregoing is merely illustrative embodiments of the present invention, and the present invention is not limited thereto, and any changes or substitutions that may be easily contemplated by those skilled in the art within the scope of the present invention should be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.
Claims (10)
1. The abnormal electricity utilization analysis method based on the space-time ground feature is characterized by comprising the following steps of:
step one, realizing multistage grid division on a monitoring area;
step two, collecting and sorting historical electricity consumption information of users in an electricity consumption monitoring area, wherein the historical electricity consumption information comprises user types, user electricity consumption behavior characteristics and historical weather information;
thirdly, constructing an electricity data model of the regional users based on regional space-time data and historical electricity consumption, and realizing prediction of the electricity data of the regional users;
setting thresholds of power consumption of users in different categories, power consumption of users in different periods of time and power consumption of users in different weather conditions;
and fifthly, analyzing whether the power consumption is within a reasonable range of the threshold value calibrated in the fourth step according to the power consumption predicted in the third step, if the power consumption is within the reasonable range of the threshold value calibrated in the fourth step, not using the power consumption abnormally, otherwise recording the power consumption abnormally.
2. The abnormal-state electrical analysis method based on space-time feature characteristics according to claim 1, wherein the first step specifically comprises:
step 1.1, in ArcGIS software, a base map layer is newly built, and the electricity consumption monitoring area range is defined;
step 1.2, collecting user data in the range of an electricity consumption monitoring area, newly establishing an attribute map layer in ArcGIS software, and storing the position of an electricity consumption user in the newly established attribute map layer in the form of point elements;
step 1.3, dividing the electricity consumption monitoring area into square grid areas with the same size;
step 1.4, counting the number of users in the square grid, and if the number of users is higher than a set threshold value, continuously dividing the grid into small grids;
step 1.5, repeatedly dividing small grids according to the step 1.4, and ensuring that the number of users in each grid area is not more than 10;
and step 1.6, numbering all grids to realize accurate indexing to the corresponding grid area according to the numbering.
3. The abnormal electricity analysis method based on space-time feature characteristics according to claim 2, wherein the second step specifically comprises:
step 2.1, collecting historical electricity utilization information of all users in an electricity utilization monitoring area, wherein the historical electricity utilization information comprises the date of the last three years and the electricity utilization data of the time period;
step 2.2, dividing users in grids with all numbers in step 1.6 into two types by combining the statistical data of the power utilization departments: industrial electricity users and residential electricity users;
step 2.3, associating all the statistical data with the point elements in the attribute layer in the step one by one, namely associating the user information with the user points in the space-time information;
and 2.4, collecting time-sharing weather data of the area for nearly three years, and recording.
4. The abnormal-state electrical analysis method based on space-time clutter characteristics of claim 3, wherein step 2.1 further comprises: if the acquired data is partially missing, an average interpolation method is adopted, and the average value is obtained according to the power consumption data of the same time period of the previous and the next date, and the method is specifically described as follows:
a) If the electricity consumption data of a user is lost in the last three years, firstly, checking whether the electricity consumption data loss period of the user is no electricity consumption behavior or data acquisition is omitted;
b) If no electricity consumption behavior exists, marking the electricity consumption data of the period as 0, terminating the subsequent step, and directly jumping to the step 2.2;
c) If the user electricity consumption data is missing because the user electricity consumption data is not acquired or is not acquired, taking the average value of the electricity consumption data of the same time period of the previous day and the next day;
d) If the electricity data of the user in a certain period are not collected or are not collected for a plurality of continuous days, the electricity data of the user in the period are marked as abnormal, and are marked as-1.
5. The abnormal-state electricity analysis method based on space-time feature characteristics according to claim 3, wherein the third step specifically comprises:
step 3.1, constructing a deep learning RNN prediction model by using a universal recurrent neural network RNN;
step 3.2, predicting the total power consumption of each grid by using the RNN prediction model constructed in step 3.1 according to the grid sequence constructed in step 1.6, wherein the method specifically comprises the following steps:
a) Firstly, carrying out time-division addition on the historical electricity consumption data of each grid to obtain the sum data of the electricity consumption of each grid per hour;
b) Then, in each grid, carrying out time-period addition on historical electricity consumption data of each user to obtain historical time-period user data of each user in the grid;
c) Predicting the current power consumption data of each grid by using the RNN prediction model constructed in the step 3.1, if the prediction data are consistent with the real-time data, skipping the step 3.2, and continuing the step 3.3;
d) If the predicted data and the real-time data are inconsistent, predicting the electricity consumption of users in the grid one by using an RNN prediction model, and marking the electricity users inconsistent with the real-time electricity consumption data;
step 3.3, when the predicted data marked in the step 3.2 do not accord with the real-time user data, backtracking and retrieving the historical electricity consumption data of the user, and inquiring whether the situation of data insufficiency exists or not: if the power consumption data is missing because the power consumption data is not acquired or is not acquired, taking the average value of the power consumption data of the same time period of the previous day and the next day; if not, the step is ended and recorded as an abnormal condition.
6. The abnormal-state electricity analysis method based on space-time feature characteristics according to claim 5, wherein the fourth step specifically comprises:
step 4.1, for industrial electricity users, the specific threshold calibration method is described as follows:
a) Dividing industrial electricity into two types of electricity consumption in holidays and non-holiday;
b) For electricity consumption in non-holidays, the calibration threshold is 10%, namely if the difference between the predicted electricity consumption and the actual electricity consumption is more or less than 10%, the power consumption is calibrated as abnormal electricity consumption;
c) The electricity consumption conditions of holidays of industrial electricity users are further divided into normalized electricity consumption and non-normalized electricity consumption;
d) The normalized electricity consumption is compared with the time period of the non-holiday at ordinary times, if the electricity consumption difference is smaller than 20%, the normalized electricity consumption is defined as normalized electricity consumption, and the same threshold value calibration as the non-holiday is adopted;
e) The rest conditions of electricity consumption in holidays are marked as abnormal electricity consumption; if the electricity consumption is 20% of the electricity consumption in the same period of the non-holiday at ordinary times, marking the abnormal electricity consumption; if the electricity consumption is far smaller than the electricity consumption in the same period of the usual non-holiday, marking as scattered electricity consumption, wherein the electricity consumption data is not used as electricity consumption evaluation data; wherein "much less than" is defined as less than 1/10 of the amount of electricity used at ordinary times;
step 4.2, for domestic electricity consumption, the specific threshold calibration method is described as follows:
a) Setting a threshold value to be 30%, if the difference between the predicted power consumption and the real-time power consumption is less than 30%, calibrating the power consumption to be normal power consumption, and continuing the subsequent step 4.2;
b) If the predicted electricity consumption and the real-time electricity consumption are more than 30% and less than 100%, calculating the temperature difference between the current predicted electricity consumption period and the historical data at the same time, and if the temperature difference is more than 8 ℃, calibrating as: the power consumption of the user is changed due to weather change, so that abnormal power consumption is avoided;
c) If the temperature is lower than 8 ℃, reading real-time electricity consumption data of the resident electricity consumption user, and if the real-time electricity consumption is lower than 1 kilowatt hour, calibrating that the resident electricity consumption is lower, the fluctuation range is larger, and the resident electricity consumption is not abnormal electricity consumption;
d) Locating the resident user as an abnormal electricity utilization situation in real time according to the rest of the conditions defined in the steps b) and c).
7. The abnormal-state electricity analysis method based on space-time feature characteristics according to claim 6, wherein the fifth step specifically comprises:
step 5.1, analyzing the type of each electricity user in each numbered grid as an industrial electricity user or a resident electricity user;
step 5.2, predicting the total power consumption of each numbered grid in each period, and comparing the total power consumption with real-time power consumption data;
step 5.3, if the electricity consumption of a certain grid is abnormal in a certain period, indicating that the grid has potential risks of abnormal electricity consumption;
step 5.4, aiming at all power utilization users of the grid, detecting whether the real-time power utilization of each user is abnormal or not by adopting the dynamic threshold calibration method set in the step four;
step 5.5, for the user marked as abnormal electricity consumption, the user needs to be verified manually, and if the user is actually abnormal in electricity consumption, the mining work of the abnormal electricity consumption user related to the invention is completed; if the user power consumption is verified to be normal, the real-time power consumption data is updated to the historical power consumption data base so as to improve the accuracy of subsequent prediction.
8. An abnormal electricity consumption analysis device based on space-time ground feature, characterized by comprising:
the geographic grid division module is used for realizing multistage grid division on the monitoring area and ensuring that the number of users in each grid is relatively fixed;
the regional historical electricity consumption information arrangement module is used for collecting and arranging historical electricity consumption information of users in the electricity consumption monitoring region, wherein the historical electricity consumption information comprises user types, user electricity consumption behavior characteristics and historical weather information;
the regional user electricity consumption data prediction module is used for constructing a regional user electricity consumption data model based on regional space-time data and historical electricity consumption information, and realizing prediction of regional user electricity consumption data;
the dynamic threshold calibration module is used for setting different thresholds and realizing the dynamic threshold setting of the power consumption of users in different categories, the power consumption of users in different time periods and the power consumption of users in different weather conditions;
and the regional abnormal electricity utilization user mining module is used for analyzing whether the regional abnormal electricity utilization user mining module is in a reasonable range of the threshold value calibrated by the dynamic threshold value calibration module according to the electricity utilization quantity predicted by the regional user electricity utilization data prediction module, if the regional abnormal electricity utilization user mining module is in the reasonable range of the threshold value calibrated by the dynamic threshold value calibration module, the regional abnormal electricity utilization user mining module is not used for abnormal electricity utilization, and otherwise, the regional abnormal electricity utilization is marked as abnormal electricity utilization.
9. The abnormal electricity consumption analysis device based on space-time feature characteristics according to claim 8, wherein the area history electricity consumption information sorting module is specifically configured to: collecting historical electricity consumption data of all users in an electricity consumption monitoring area, wherein the historical electricity consumption data comprise the date of minute and the time of minute of the last three years; users within all numbered grids are classified into two categories: industrial electricity users and residential electricity users; associating all the statistical data with the point elements in the attribute layer one by one, namely associating the user information with the user points in the space-time information; and collecting the time-sharing weather data of the region for nearly three years, and recording to finish the arrangement of the historical electricity utilization information of the region.
10. The abnormal electricity consumption analysis method based on space-time feature of claim 9, wherein the regional abnormal electricity consumption user discovery module is specifically configured to: for each numbered grid, analyzing that the respective type of the electricity utilization user in the grid is an industrial electricity utilization user or a resident electricity utilization user; for each numbered grid, predicting the total electricity consumption of the grid in each period, comparing the total electricity consumption with real-time electricity consumption data, and if the electricity consumption of a certain grid in a certain period is abnormal, indicating that the grid has potential risk of abnormal electricity consumption; aiming at all power utilization users of the grid, a dynamic threshold calibration method set by a dynamic threshold calibration module is adopted to detect whether the power utilization of each user is abnormal in real time, and for the users calibrated to be abnormal power utilization, the manual verification is needed, and if the users are indeed abnormal power utilization, the mining work of the abnormal power utilization users related to the invention is completed; if the user is verified to be normal in electricity consumption, the real-time electricity consumption data is updated to the historical electricity consumption database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310850043.8A CN117039850A (en) | 2023-07-12 | 2023-07-12 | Abnormal electricity consumption analysis method and system based on space-time ground feature characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310850043.8A CN117039850A (en) | 2023-07-12 | 2023-07-12 | Abnormal electricity consumption analysis method and system based on space-time ground feature characteristics |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117039850A true CN117039850A (en) | 2023-11-10 |
Family
ID=88641986
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310850043.8A Pending CN117039850A (en) | 2023-07-12 | 2023-07-12 | Abnormal electricity consumption analysis method and system based on space-time ground feature characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117039850A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117689070A (en) * | 2023-12-13 | 2024-03-12 | 北京朗杰科技有限公司 | Ammeter management system based on internet of things equipment |
| CN117788046A (en) * | 2024-01-25 | 2024-03-29 | 广东美电国创科技有限公司 | Power consumption monitoring and early warning method and device based on Internet of things |
-
2023
- 2023-07-12 CN CN202310850043.8A patent/CN117039850A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117689070A (en) * | 2023-12-13 | 2024-03-12 | 北京朗杰科技有限公司 | Ammeter management system based on internet of things equipment |
| CN117788046A (en) * | 2024-01-25 | 2024-03-29 | 广东美电国创科技有限公司 | Power consumption monitoring and early warning method and device based on Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117039850A (en) | Abnormal electricity consumption analysis method and system based on space-time ground feature characteristics | |
| CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
| CN110097297A (en) | A kind of various dimensions stealing situation Intellisense method, system, equipment and medium | |
| CN109828182A (en) | A kind of network system accident analysis method for early warning based on failure modes processing | |
| CN108268597A (en) | A kind of moving-target activity probability map construction and behavior intension recognizing method | |
| CN116797049B (en) | Quantitative evaluation method for differentiated energy-saving potential of power distribution network | |
| CN116976707B (en) | User electricity consumption data anomaly analysis method and system based on electricity consumption data acquisition | |
| CN116680658B (en) | Heat wave monitoring station site selection method and system based on risk evaluation | |
| Tang et al. | Enhancement of distribution load modeling using statistical hybrid regression | |
| CN113593191A (en) | Visual urban waterlogging monitoring and early warning system based on big data | |
| CN104715156A (en) | Building energy-saving potential dynamic assessment method | |
| Hartmann et al. | Suspicious electric consumption detection based on multi-profiling using live machine learning | |
| CN117538503A (en) | Real-time intelligent soil pollution monitoring system and method | |
| CN117390403B (en) | Power grid fault detection method and system for new energy lighthouse power station | |
| CN115049410A (en) | Electricity stealing behavior identification method and device, electronic equipment and computer readable storage medium | |
| Arsene et al. | Profiling consumers in a water distribution network using K-Means clustering and multiple pre-processing methods | |
| CN114169797A (en) | Energy management system and method | |
| CN117093943A (en) | Power consumption monitoring and early warning method and device | |
| CN117494009A (en) | Electrical equipment state evaluation method based on insulating material pyrolysis analysis and cloud platform | |
| CN112365082A (en) | Public energy consumption prediction method based on machine learning | |
| EP3906447A1 (en) | Method, system and computer program product for evaluation of energy consumption in industrial environments | |
| CN116383937A (en) | Digital twin protection evaluation method for villages | |
| CN110455370A (en) | Flood-control and drought relief long-haul telemetry display system | |
| CN101923605B (en) | Wind pre-warning method for railway disaster prevention | |
| CN117458722B (en) | Data monitoring method and system based on electric power energy management system |
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 |