Method, system, processing device, and medium for profiling user electricity consumption habits and formulating maintenance plans

Abstract

The present invention discloses a system for profiling users' electricity consumption habits and formulating maintenance plans. Using distribution network side data as verification, users' electricity consumption behaviors are obtained, such as regular power outages, network traffic usage periods, etc., so as to statistically analyze tags such as users' electricity consumption demand patterns and time periods. The data with tags is used for model training to obtain a target profiling model. The target profiling model is used to calculate users' electricity consumption profiles. The distribution network side statistically analyzes the impact of maintenance operations based on users' electricity consumption profiles, and then formulates a maintenance plan with a small impact range and low economic losses, thereby improving the service quality of the distribution network. In addition, using the electricity consumption habit data of multiple users in the same area, it is possible to judge users' living behaviors or travel analysis. By analyzing the activity level and patterns of users in this area, users' tolerance levels are analyzed, and the regional electricity consumption profile can be depicted. The present invention utilizes the power restoration and outage data of users' network devices, with high data accuracy, to improve the accuracy of users' electricity consumption profile characterization.

Description

Technical Field

[0001] The present invention relates to the field of power maintenance technology, and specifically to a method, system, processing equipment and medium for profiling users' power usage habits and formulating maintenance plans based on big data of power failure and power restoration of network terminals. Background Art

[0002] The low-voltage distribution network is the end-user power supply network directly facing the user, and is an important part of the power grid. In order to ensure the quality and safety of power consumption by users, daily maintenance work is essential. However, maintenance work may cause temporary power outages for users, directly affecting their power demand. If it is not made reasonably, it will directly affect the user's power experience and even bring related economic losses.

[0003] At present, the data collected by power grid companies at the user end is only the amount of electricity collected by electricity meters, and most of the electricity meter collection values ​​are only the daily frozen electricity. A single data source cannot achieve accurate analysis of the electricity consumption behavior on the user side. The maintenance plan of the low-voltage distribution network is mainly based on the current status of the power grid and is formulated. It does not analyze the actual electricity demand of users very well. Therefore, when the maintenance work is carried out, it is bound to cause certain troubles to users' electricity consumption and affect users' electricity consumption experience.

[0004] Application number CN202010131723.0 discloses a residential user electricity consumption profiling method based on multi-dimensional fine-grained behavior data. The method is based on a residential user electricity consumption profiling method based on multi-dimensional fine-grained behavior data. The method includes four parts: 1. Multi-dimensional fine-grained behavior data collection, including fine-grained electricity consumption behavior measurement data collected based on non-household terminals, user electricity bill data collected based on the marketing system, and user network behavior statistics obtained based on the online business hall and 95598; 2. Feature label model construction, establishing a user multi-source feature label system from the three dimensions of user behavior, electricity consumption characteristics, and consumption habits, and giving the calculation method or estimation method of each feature; 3. Calculating comprehensive indicators of various characteristic indicators by season and time period, proposing an improved k-means clustering algorithm, and using the improved k-means clustering algorithm to divide different electricity customers into clusters with different attributes; 4. Visual presentation of user electricity consumption profiling results as a basis for precise positioning of target users for regulation. Although this method is used to profile users' electricity usage behavior, it is only used to characterize users' electricity usage characteristics and consumption habits, classify users, and regulate the precise positioning of target users. It cannot be directly used on the distribution network side to formulate reasonable maintenance plans. Summary of the invention

[0005] The technical problem to be solved by the present invention is how to use the power failure and restoration data of user network equipment to characterize the user's power usage habits and help the distribution network side to reasonably formulate a maintenance plan.

[0006] The present invention solves the above technical problems through the following technical means:

[0007] The method for profiling the user's electricity usage habits and formulating a maintenance plan is characterized by comprising the following steps:

[0008] Step 1: construct a data set, which at least includes power failure and power restoration data of user network equipment, user network traffic time period data, distribution network side emergency repair data, and user complaint data received by the distribution network side;

[0009] Step 2: Based on users, the power failure and power restoration data of user network equipment is associated with the emergency repair data on the distribution network side and the user complaint data received on the distribution network side to obtain the user power failure habit data and regional power consumption habit data.

[0010] Step 3: Define user power usage habit labels based on user network traffic period data and user power outage habit data, and define regional power usage habit labels based on regional power usage habit data, and use the labeled data as training samples;

[0011] Step 4: Use the training samples to train the random forest model to obtain the target model;

[0012] Step 5: Use the target model to construct user power consumption profiles and regional power consumption profiles;

[0013] Step 6: The distribution network side formulates a maintenance plan based on the user power consumption profile and the regional power consumption profile.

[0014] Furthermore, the specific process of step 2 is as follows:

[0015] According to the SN number and device fingerprint of the user's power-off network device, the city, region, community, building, floor, and household number information of the user side is obtained. According to the user's power outage time and power outage area, the corresponding planned maintenance, early morning switch reclosing, and reported power outage line information are found. Based on the user as the association basis, find out whether there is corresponding planned maintenance, early morning switch reclosing, and reported power outage line information. If so, the power outage is deemed to be passive. If not found, the power outage is deemed to be an autonomous power outage.

[0016] Furthermore, the step 3 is specifically to define the user's power demand and time period label according to the user's network traffic period data, and to define the user's active power-off habit label according to the user's power usage habit data.

[0017] Furthermore, the step 6 is specifically that the distribution network side generates a maintenance plan with a small impact range based on the user's power consumption profile and the current shift schedule of the maintenance team.

[0018] The present invention also provides a system for profiling the user's electricity usage habits and formulating a maintenance plan, comprising:

[0019] A data set construction module is used to obtain historical data, wherein the data set at least includes power failure and power restoration data of user network equipment, user network traffic time period data, distribution network side emergency repair data, and user complaint data received by the distribution network side;

[0020] The user power outage habit data construction module is used to determine the user's power usage habits. Based on the user, the power outage and power restoration data of the user's network equipment are associated with the emergency repair data on the distribution network side and the user complaint data received on the distribution network side to obtain the user's power outage habit data and regional power usage habit data.

[0021] A sample data construction module is used to define user power usage habit labels based on user network traffic period data and user power outage habit data, and to define regional power usage habit labels based on regional power usage habit data, and to use the labeled data as training samples;

[0022] The training module uses the training samples to train the random forest model and obtain the target model;

[0023] The user electricity consumption profile construction module uses the target model to construct user electricity consumption profiles and regional electricity consumption profiles;

[0024] The maintenance plan formulation module is used by the distribution network side to formulate maintenance plans based on user power consumption profiles and regional power consumption profiles.

[0025] The present invention obtains user electricity consumption behaviors, such as regular power outages and network traffic usage time periods, based on user electricity consumption data and using distribution network data as verification, so as to count user electricity demand patterns and time periods and other labels, and uses labeled data for model training to obtain a target portrait model. The target portrait model is used to calculate the user's electricity consumption portrait. The distribution network side statistically analyzes the impact of maintenance operations based on the user's electricity consumption portrait, and then formulates a maintenance plan with a small impact range and low economic losses, thereby improving the service quality of the distribution network.

[0026] Furthermore, the specific process of the user power-off habit data construction module is as follows:

[0027] According to the SN number and device fingerprint of the user's power-off network device, the city, region, community, building, floor, and household number information of the user side is obtained. According to the user's power outage time and power outage area, the corresponding planned maintenance, early morning switch reclosing, and reported power outage line information are found. Based on the user as the association basis, find out whether there is corresponding planned maintenance, early morning switch reclosing, and reported power outage line information. If so, the power outage is deemed to be passive. If not found, the power outage is deemed to be an autonomous power outage.

[0028] Furthermore, the sample data construction module specifically defines the user's power demand and time period label according to the user's network traffic period data, and defines the user's active power-off habit label according to the user's power usage habit data.

[0029] Furthermore, the maintenance plan formulation module is specifically that the distribution network side generates a maintenance plan with a small impact range based on the user's power consumption profile and the current schedule of the maintenance team.

[0030] The present invention also provides a processing device, comprising at least one processor and at least one memory communicatively connected to the processor, wherein: the memory stores program instructions executable by the processor, and the processor can execute the above method by calling the program instructions.

[0031] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the above method.

[0032] The advantages of the present invention are:

[0033] The present invention obtains user electricity consumption behaviors, such as regular power outages and network traffic usage time periods, based on user electricity consumption data and using distribution network data as verification, so as to count user electricity demand patterns and time periods and other labels, and uses labeled data for model training to obtain a target portrait model. The target portrait model is used to calculate the user's electricity consumption portrait. The distribution network side statistically analyzes the impact of maintenance operations based on the user's electricity consumption portrait, and then formulates a maintenance plan with a small impact range and low economic losses, thereby improving the service quality of the distribution network.

[0034] In addition, the electricity usage habit data of multiple users in the same area can be used to determine the user's residential behavior or itinerary analysis. By analyzing the activity level and regularity of users in this area and the user's tolerance, a regional electricity usage profile can be portrayed. The present invention uses the power failure and restoration data of user network devices, which has high data accuracy and improves the accuracy of user electricity usage profile portrayal. Importantly, the present method does not require additional equipment and is low in cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of a method for using network equipment power failure and power restoration data to profile a user's power usage habits and formulate a maintenance plan in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] This embodiment provides a method for using the power failure and power recovery data of network equipment to profile the user's power usage habits and formulate a maintenance plan, such as Figure 1 As shown, the following steps are included:

[0038] Step 1: construct a data set, which includes at least user-side power consumption data and distribution network-side maintenance data. The user-side power consumption data includes power failure and power restoration data of user network equipment and user network traffic time period data. The distribution network-side data includes distribution network-side emergency repair data and user complaint data received by the distribution network;

[0039] Step 2: Based on users, the power failure and power restoration data of user network equipment is associated with the emergency repair data on the distribution network side and the user complaint data received on the distribution network side to obtain the user power failure habit data and regional power consumption habit data.

[0040] Specifically: According to the SN number and device fingerprint of the user's power-off network device, the city, region, community, building, floor, and household number information of the user side are obtained; according to the user's power outage time and power outage area, the corresponding planned maintenance, early morning switch reclosing, and reported power outage line information are searched; based on the user as the association basis, find out whether there is corresponding planned maintenance, early morning switch reclosing, and reported power outage line information; if so, the power outage is deemed to be a passive power outage; if not found, the power outage is deemed to be an autonomous power outage, and the corresponding power-on behavior is also an autonomous power-on behavior.

[0041] Step 3: Define the user's power demand and time period label based on the user's network traffic period data, and define the user's active power-off habit label based on the user's power usage habit data.

[0042] Step 4: Use the training samples to train the random forest model. The height of the decision tree of the random forest algorithm is h, and the number of decision trees is N. The setting of the height h and the number N depends on the data scale and task. In this task, it is set to h = 4 and N = 100.

[0043] Get the target model;

[0044] Step 5: Use the target model to construct user power consumption profiles and regional power consumption profiles;

[0045] Step 6: The distribution network generates a maintenance plan with a small impact range based on the user's power consumption profile and the current schedule of the maintenance team.

[0046] This embodiment obtains user electricity usage behaviors, such as regular power outages and network traffic usage time periods, based on user electricity usage data and using distribution network data as verification, so as to count user electricity demand patterns and time periods, and use labeled data for model training to obtain a target portrait model. The target portrait model is used to calculate the user electricity usage portrait. The distribution network side statistically analyzes the impact of maintenance operations based on the user electricity usage portrait, and then formulates a maintenance plan with a small impact range and low economic losses, thereby improving the service quality of the distribution network.

[0047] Corresponding to the above method, the present invention also provides a system for profiling the user's power usage habits and formulating a maintenance plan by using the power failure and power restoration data of the network device, including:

[0048] A data set construction module is used to obtain historical data, the data set includes at least user-side power consumption data and distribution network-side maintenance data, the user-side power consumption data includes power failure and power restoration data of user network equipment, user network traffic time period data, and the distribution network-side data includes distribution network-side emergency repair data and user complaint data received by the distribution network;

[0049] The user power outage habit data construction module is used to determine the user's power usage habits. Based on the user, the power outage and power restoration data of the user's network equipment are associated with the emergency repair data on the distribution network side and the user complaint data received on the distribution network side to obtain the user's power outage habit data and regional power usage habit data.

[0050] Specifically: According to the SN number and device fingerprint of the user's power-off network device, the city, region, community, building, floor, and household number information of the user side are obtained; according to the user's power outage time and power outage area, the corresponding planned maintenance, early morning switch reclosing, and reported power outage line information are searched; based on the user as the association basis, find out whether there is corresponding planned maintenance, early morning switch reclosing, and reported power outage line information; if so, the power outage is deemed to be a passive power outage; if not found, the power outage is deemed to be an autonomous power outage, and the corresponding power-on behavior is also an autonomous power-on behavior.

[0051] The sample data construction module is used to define the user's electricity demand and time period label according to the user's network traffic period data, and to define the user's active power-off habit label according to the user's electricity usage habit data.

[0052] Training module, use training samples to train the random forest model. The height of the decision tree of the random forest algorithm is h, and the number of decision trees is N. The setting of height h and number N depends on the data scale and task. In this task, it is set to h = 4, N = 100, and the target model is obtained;

[0053] The user electricity consumption profile construction module uses the target model to construct user electricity consumption profiles and regional electricity consumption profiles;

[0054] The maintenance plan formulation module generates a maintenance plan with a small impact range based on the user's power consumption profile and the current schedule of the maintenance team on the distribution network side.

[0055] This embodiment also provides a processing device, including at least one processor and at least one memory communicatively connected to the processor, wherein: the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the method as described above.

[0056] This embodiment also provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the above method.

[0057] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. Method for profiling user electricity consumption habits and formulating maintenance plans, characterized in that, it includes the following steps: Step 1, construct a data set, which at least includes user network device power failure and restoration data, user network traffic period data, distribution network side repair data, and user complaint data received by the distribution network side; Step 2, based on the user, associate the user network device power failure and restoration data with the distribution network side repair data and the user complaint data received by the distribution network side to obtain user power failure habit data and regional electricity consumption habit data, and then determine whether this power failure is a passive power failure or an autonomous power-off behavior; Step 3, define user electricity consumption habit labels according to the user network traffic period data and user power failure habit data, and define regional electricity consumption habit labels for the regional electricity consumption habit data, and use the labeled data as training samples; Step 4, train a random forest model with the training samples to obtain a target model; Step 5, use the target model to construct user electricity consumption profiles and regional electricity consumption profiles; Step 6, the distribution network side formulates a maintenance plan according to the user electricity consumption profile and the regional electricity consumption profile.

2. The method for profiling user electricity consumption habits and formulating maintenance plans according to claim 1, characterized in that, the specific process of step 2 is: According to the sn number and device fingerprint of the user's power failure network device, obtain the city, region, community, building, floor, and household number information where the user is located. According to the user's power outage time and outage area, search for the corresponding planned maintenance, early morning switch reclosing, and reported power failure line information. Based on the user as the association basis, search for whether there is corresponding planned maintenance, early morning switch reclosing, and reported power failure line information. If there is, it is determined that this power failure is a passive power failure. If not, it is determined that this power failure is an autonomous power-off behavior.

3. The method for profiling user electricity consumption habits and formulating maintenance plans according to claim 1, characterized in that, step 3 is specifically to define electricity consumption demand and time period labels for users according to the user network traffic period data, and define active power failure habit labels for users according to the user power failure habit data.

4. The method for profiling user electricity consumption habits and formulating maintenance plans according to claim 1, characterized in that, the specific process of step 6 is that the distribution network side generates a maintenance plan with a small impact range according to the user electricity consumption profile and the current shift schedule of the maintenance team.

5. System for profiling user electricity consumption habits and formulating maintenance plans, characterized in that, it includes: A data set construction module for obtaining historical data, which at least includes user network device power failure and restoration data, user network traffic period data, distribution network side repair data, and user complaint data received by the distribution network side; A user power failure habit data construction module for determining user electricity consumption habits. Based on the user, associate the user network device power failure and restoration data with the distribution network side repair data and the user complaint data received by the distribution network side to obtain user power failure habit data and regional electricity consumption habit data A sample data construction module, which is used to define user electricity consumption habit tags according to user network traffic period data and user power-off habit data, and define regional electricity consumption habit tags for regional electricity consumption habit data, and use the data with tags as training samples; A training module, which trains a random forest model with the training samples to obtain a target model; A user electricity consumption portrait construction module, which constructs user electricity consumption portraits and regional electricity consumption portraits by using the target model; An overhaul plan formulation module, which is used for the distribution network side to formulate an overhaul plan according to the user electricity consumption portrait and the regional electricity consumption portrait.

6. The system for constructing a user electricity consumption habit portrait and formulating an overhaul plan according to claim 5, wherein, the specific process of the user power-off habit data construction module is: According to the sn number and device fingerprint of the user's power-off network device, obtain the information of the city, region, community, building, floor, and household number where the user is located. According to the user's power-off time and power-off area, search for the corresponding power-off line information for planned maintenance, early morning switch reclosing, and repair. Based on the user as the association basis, search for whether there is corresponding power-off line information for planned maintenance, early morning switch reclosing, and repair. If there is, it is determined that this power-off is a passive power-off. If not, it is determined that this power-off is an autonomous power-off behavior.

7. The system for constructing a user electricity consumption habit portrait and formulating an overhaul plan according to claim 5, wherein, the sample data construction module is specifically: defining electricity consumption demand and period tags for users according to user network traffic period data, and defining active power-off habit tags for users according to user electricity consumption habit data.

8. The system for constructing a user electricity consumption habit portrait and formulating an overhaul plan according to claim 5, wherein, the overhaul plan formulation module is specifically: the distribution network side generates an overhaul plan with a small influence range according to the user electricity consumption portrait and in combination with the current shift arrangement of the overhaul team.

9. A processing device, wherein, it includes at least one processor and at least one memory communicatively connected to the processor, wherein: the memory stores program instructions executable by the processor, and the processor can execute the method according to any one of claims 1 to 4 by calling the program instructions.

10. A computer-readable storage medium, wherein, the computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the method according to any one of claims 1 to 4.

Images