A power consumption analysis system and method for large industrial users based on unsupervised learning

Abstract

This invention relates to the technical field of electricity consumption analysis for large industrial users, and in particular to a system and method for analyzing electricity consumption for large industrial users based on unsupervised learning. It employs a directional transmission method to transmit collected data, reducing the impact of surrounding electrical equipment on data transmission and minimizing the influence of surrounding electrical equipment during data transmission. During data transmission, the system compares the data with randomly extracted data features and compares the data size before and after transmission to determine the integrity and authenticity of the data, ensuring the accuracy of the analysis and prediction results. Furthermore, it uses the prediction results for virtual simulation to rationally plan electricity consumption, thereby improving the system's practicality. The system includes: a data acquisition module, a data aggregation and transmission module, a data preprocessing module, a redundant transmission module, a data analysis module, a data detection module, a simulation module, and a data sharing module.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power consumption analysis of large industrial users, and particularly relates to a large industrial user power consumption analysis system and method based on unsupervised learning. BACKGROUND

[0002] Large industrial users refer to enterprises that use a large amount of raw materials and energy in the industrial field. In order to improve the use efficiency of electric energy and reduce the peak and valley of power consumption, the power management department needs to analyze and predict the power consumption of large industrial users to facilitate reasonable power supply.

[0003] Generally, the power consumption of users is analyzed by using a user power consumption behavior analysis system disclosed in an invention patent with publication number CN112785338A and a user power consumption data analysis integrated system based on power consumption information collection disclosed in a utility model patent with publication number CN207283595U.

[0004] However, due to the large site of large industrial users, a large amount of data needs to be collected, and the signals in part of the detection area are easily interfered, and the data is easily lost and maliciously tampered with. The existing power consumption analysis system is not convenient for detecting and predicting the integrity and authenticity of data in the data transmission process, and it is easy to use incomplete or false data for analysis and prediction, which affects the analysis and prediction results and the planning of power transmission by the power management department, resulting in poor practicability. Therefore, a large industrial user power consumption analysis system and method based on unsupervised learning are needed to improve the above problems. SUMMARY

[0005] To solve the above technical problems, the present application provides a large industrial user power consumption analysis system and method based on unsupervised learning, which uses a directional transmission method to transmit the collected data, reduces the influence of surrounding electrical equipment on data transmission, and also reduces the influence of surrounding electrical equipment in the data transmission process. Then, in the data transmission process, the integrity and authenticity of the data are judged by comparing the data with randomly extracted data features and comparing the size of the data before and after transmission, ensuring the accuracy of the analysis and prediction results, and using the prediction results to perform virtual simulation to reasonably plan power consumption, thereby improving the practicability of the system.

[0006] The large industrial user power consumption analysis system and method based on unsupervised learning provided by the present application comprises:

[0007] Data collection module: collect the power consumption data of large industrial users, adjust the data transmission direction according to the installation environment, and transmit the data to other data collection modules through directional wireless transmission, other data collection modules continue to transmit data to the designated direction through this way until the data is transmitted to the designated data collection module;

[0008] Summary transmission module: summarize the received data and transmit the data to the data preprocessing module;

[0009] Data preprocessing module: compare the received data with the past received data, complete preliminary analysis, and add the data to get the total value of the data, and send the received data, preliminary analysis results and total value of the data to the data analysis module;

[0010] Redundant transmission module: transmit the data of sudden situation separately;

[0011] Data analysis module: analyze the power consumption of large industrial users according to the received data, preliminary analysis results and total value of the data, and predict the power consumption mode of large industrial users in the future period;

[0012] Data detection module: the data collection module randomly extracts part of the characteristics in the data before data transmission, and records the size of the data, the summary transmission module compares the extracted characteristics with the corresponding position in the data after receiving the data, and compares the size of the data with the recorded size, the data preprocessing module, the data analysis module receives the data again, and the above method is used to detect the data to ensure the integrity and authenticity of the data;

[0013] Simulation module: users can virtually adjust their power consumption mode in the future period according to the prediction results;

[0014] Data sharing module: transmit the data and analysis prediction results received by the data analysis module to other management departments, so that other power management departments can use the data and prediction results as training data for the data analysis module, and receive the power consumption data and analysis prediction results from other management departments and combine the data of the management department as training data for the data analysis module.

[0015] Preferably, the data collection module comprises:

[0016] Collection unit: collect the power consumption data of users;

[0017] Directional transmission unit: send data through directional transmission;

[0018] Storage module: backup the collected raw data and store the backup.

[0019] Preferably, the storage module comprises:

[0020] a storage unit: backing up data and storing the backed-up data;

[0021] a deletion unit: deleting data that exceeds the storage period;

[0022] a limited recovery unit: recovering mis-deleted data within a specified period.

[0023] Preferably, the data preprocessing module comprises:

[0024] a data comparison module: comparing the received data with previously received data, and analyzing the difference between the received data and the previous data;

[0025] a data calculation module: adding the received data to obtain the total of the power consumption data of large industrial users.

[0026] Preferably, the redundant transmission module comprises:

[0027] an independent transmission unit: transmitting emergency data in the data acquisition module to the sorting transmission unit;

[0028] a sorting transmission unit: compressing and packaging the received data, and transmitting the compressed and packaged data to the unmanned aerial vehicle transmission unit;

[0029] an independent power supply module: independently supplying power to the independent transmission unit, the sorting transmission unit, and the data acquisition module, so that the independent transmission unit and the sorting transmission unit can normally operate in the case of power failure;

[0030] an unmanned aerial vehicle transmission unit: the unmanned aerial vehicle transmission unit flies on a preset route, receives the compressed and packaged data sent by the sorting transmission unit, returns to the takeoff location, and then transmits the compressed and packaged data to the data analysis module.

[0031] Preferably, the unmanned aerial vehicle transmission unit comprises:

[0032] an unmanned aerial vehicle module: flying in the air along a set route;

[0033] a navigation module: identifying the flight route and identifying obstacles on the route, so that the unmanned aerial vehicle avoids the obstacles;

[0034] a wireless transmission module: receiving data sent by the sorting transmission unit, returning to the takeoff location, and then sending the data to the data analysis module.

[0035] Preferably, the data analysis module uses a deep learning processor to analyze the data and predict the power consumption pattern of large industrial users in the future.

[0036] The deep learning processor adopts a PyTorch framework and a recurrent neural network.

[0037] Preferably, the data detection module comprises:

[0038] a feature extraction unit: randomly extracting a plurality of features in the data before data transmission, and arranging the features into feature data;

[0039] a detection unit: comparing the received data and the feature data, and comparing the size of the data before transmission and the data after reception;

[0040] a warning unit: issuing a warning to remind the staff when the comparison shows that the data is different or the data is missing.

[0041] Preferably, the simulation module comprises:

[0042] a display module: displaying the prediction results and the collected data to enable the user to know his / her power consumption situation in time;

[0043] a setting module: managing the display module, the adjustment module, the transceiver module, and the login management module;

[0044] an adjustment module: enabling the user to virtually adjust and plan his / her power consumption situation in a future period of time according to the prediction results;

[0045] a receiving module: receiving the prediction results and the collected data;

[0046] a login management module: checking the information of the logged-in personnel.

[0047] The analysis method of the large industrial user power consumption analysis system based on unsupervised learning comprises the following steps:

[0048] S1, through a plurality of data acquisition modules: collecting power consumption data at different positions of a large industrial user, transmitting the data to other data acquisition modules through directional wireless transmission, and continuing to transmit the data to the designated direction through this method until the data is transmitted to the designated data collection and transmission module, thereby reducing the mutual interference between the data acquisition modules and the surrounding electrical equipment during data transmission;

[0049] S2, the data acquisition module randomly extracts part of the features in the data before data transmission, records the size of the data, and compares the extracted features with the corresponding positions in the data after the data collection and transmission module receives the data, and compares the size of the data with the recorded size to detect the integrity and authenticity of the data;

[0050] S3, the summary transmission module: the received data is summarized, and the data is transmitted to the data preprocessing module, the received data is compared with the past received data through the data preprocessing module, the difference between the received data and the past data is analyzed, and the received data is added, the sum of the large industrial user electricity data is obtained, then the received data, the preliminary analysis result and the data total value are sent to the data analysis module;

[0051] S4, the data detection module is also used in the transmission process to detect the integrity and authenticity of the data;

[0052] S5, the data analysis module is used to analyze the electricity consumption of the large industrial user according to the received data, the preliminary analysis result and the data total value, and to predict the electricity consumption mode of the large industrial user in the future;

[0053] S6, the user adjusts the electricity consumption mode in the future according to the prediction result through the simulation module;

[0054] S7, the data sharing module is used to transmit the data received by the data analysis module and the analysis prediction result to other management departments, so that other power management departments can use the data and prediction result as training data of the data analysis module, and receive the electricity data and analysis prediction result sent by other management departments and combine the data of the management department as the training data of the data analysis module.

[0055] Compared with the prior art, the beneficial effects of the present application are:

[0056] 1, by extracting part of the features in the data before data transmission, and transmitting the extracted feature data together with the complete data, and after receiving the data, comparing the feature data with the complete data, and comparing the size before data transmission with the size after data transmission, the integrity and authenticity of the data are realized;

[0057] 2, after data collection, by the way of directional transmission, the interference in the data transmission process is reduced, and by the way of short distance data transmission between the data collection modules, the data loss is reduced;

[0058] 3, by sending the prediction result and the electricity consumption data to the user end, the user can adjust the electricity consumption data in the virtual environment, which is convenient for the user to adjust and understand the future electricity consumption mode;

[0059] 4, by the way of redundant line and unmanned aerial vehicle transmission, the emergency data is transmitted, which is convenient for timely processing of special situations. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1It is a structural schematic diagram of the power consumption analysis system for large industrial users based on unsupervised learning of the application;

[0061] Figure 2 It is a structural schematic diagram of the data collection module of the application;

[0062] Figure 3 It is a structural schematic diagram of the storage module of the application;

[0063] Figure 4 It is a structural schematic diagram of the data preprocessing module of the application;

[0064] Figure 5 It is a structural schematic diagram of the redundancy transmission module of the application;

[0065] Figure 6 It is a structural schematic diagram of the unmanned aerial vehicle transmission unit of the application;

[0066] Figure 7 It is a structural schematic diagram of the data analysis module of the application;

[0067] Figure 8 It is a structural schematic diagram of the data detection module of the application;

[0068] Figure 9 It is a structural schematic diagram of the simulation module of the application. DETAILED DESCRIPTION

[0069] In order to facilitate the understanding of the application, the application will be described more fully below with reference to the accompanying drawings. The application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the application more thorough and comprehensive. EMBODIMENT

[0070] As Figures 1 to 9 shown, including:

[0071] Data collection module: collect the power consumption data of large industrial users, adjust the data transmission direction according to the installation environment, and transmit the data to other data collection modules through directional wireless transmission, and other data collection modules continue to transmit data to the designated direction through this way until the data is transmitted to the designated data collection module for aggregation and transmission module;

[0072] Aggregation and transmission module: aggregate the received data and transmit the data to the data preprocessing module;

[0073] Data preprocessing module: compare the received data with the previously received data, complete preliminary analysis, and add the data to obtain the total data value, and send the received data, preliminary analysis result and total data value to the data analysis module;

[0074] Redundant transmission module: separate transmission of data in burst conditions;

[0075] Data analysis module: analyze the power consumption of large industrial users according to the received data, preliminary analysis results and total data value, and predict the power consumption mode of large industrial users in the future period;

[0076] Data detection module: the data acquisition module randomly extracts part of the characteristics in the data before data transmission, and records the size of the data. After the data transmission module receives the data, it compares the extracted characteristics with the corresponding position in the data, and compares the size of the data with the recorded size. The data preprocessing module detects the data before data transmission, and the data analysis module detects the data after data reception again through the above method, to ensure the integrity and authenticity of the data;

[0077] Simulation module: users can virtually adjust their power consumption mode in the future period according to the prediction results;

[0078] Data sharing module: transmit the data and analysis and prediction results received by the data analysis module to other management departments, so that other power management departments can use the data and prediction results as training data for the data analysis module, and receive the power consumption data and analysis and prediction results from other management departments and combine the data of the management department as training data for the data analysis module;

[0079] The data acquisition module comprises:

[0080] Acquisition unit: acquires the power consumption data of users;

[0081] Directional transmission unit: sends data through directional transmission;

[0082] Storage module: backs up the collected raw data and stores the backup;

[0083] The storage module comprises:

[0084] Storage unit: backs up data and stores the backed-up data;

[0085] Deletion unit: deletes data that exceeds the storage period;

[0086] Limited recovery unit: recovers mistakenly deleted data within the specified period;

[0087] The data preprocessing module comprises:

[0088] Data comparison module: compares the received data with the data received in the past, and analyzes the difference between the received data and the past data;

[0089] Data calculation module: add the received data to obtain the total of the power consumption data of large industrial users;

[0090] The redundant transmission module comprises:

[0091] Independent transmission unit: transmit the emergency data in the data acquisition module to the arrangement transmission unit;

[0092] Arrangement transmission unit: compress and package the received data, and transmit the compressed and packaged data to the unmanned aerial vehicle transmission unit;

[0093] Independent power supply module: independently supplies power to the independent transmission unit, the arrangement transmission unit and the data acquisition module, so that the independent transmission unit and the arrangement transmission unit normally operate in the case of power failure;

[0094] Unmanned aerial vehicle transmission unit: the unmanned aerial vehicle transmission unit flies on a preset route, receives the compressed and packaged data sent by the arrangement transmission unit, and returns to the take-off location, and then transmits the compressed and packaged data to the data analysis module;

[0095] The unmanned aerial vehicle transmission unit comprises:

[0096] Unmanned aerial vehicle module: fly in the air along the set route;

[0097] Navigation module: identify the flight route and identify the obstacles on the route, so that the unmanned aerial vehicle avoids the obstacles;

[0098] Wireless transmission module: receive the data sent by the arrangement transmission unit, return to the take-off location, and then send the data to the data analysis module;

[0099] The data analysis module uses a deep learning processor to analyze the data and predict the power consumption mode of large industrial users in the future period of time;

[0100] The deep learning processor uses a PyTorch framework and a recurrent neural network;

[0101] The data detection module comprises:

[0102] Feature extraction unit: randomly extract a plurality of features in the data before data transmission, and arrange the features into feature data;

[0103] Detection unit: compare the received data and the feature data, and compare the sizes of the data before transmission and the data after reception;

[0104] Early warning unit: when the comparison shows that the data is different or the data is missing, an alarm is sent to remind the staff;

[0105] The simulation module comprises:

[0106] Display module: display the prediction results and collected data, so that the user can know his / her power consumption situation in time;

[0107] Setting module: manage the display module, adjustment module, transceiver module and login management module;

[0108] Adjustment module: the user can virtually adjust and plan his / her power consumption situation in the future according to the prediction results;

[0109] Receiving module: receive the prediction results and collected data;

[0110] Login management module: check the information of the login personnel.

[0111] The main functions realized by the present application are: improving the prediction accuracy, reducing data loss and improving the convenience of system use;

[0112] 1. Improving the prediction accuracy: by protecting the integrity and authenticity of the data during transmission, the prediction accuracy is improved;

[0113] 2. Reducing data loss: by short-distance data transmission and directional transmission between the data collection modules, the data loss is reduced;

[0114] 3. Improving the convenience of system use: by using the data of other power management departments and the present power management department, the data analysis module can train itself, improve the prediction accuracy and reduce the labor of the staff.

[0115] The deep learning processor of the large industrial user power consumption analysis system and method based on unsupervised learning of the present application is purchased on the market, and the technical personnel in the industry only need to operate according to the use instruction book attached, without the technical personnel in the field having to pay creative labor.

[0116] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary technical personnel in the technical field, several improvements and modifications can be made without departing from the technical principles of the present application, and these improvements and modifications should also be regarded as the protection range of the present application.

Claims

1. A large industrial consumer power consumption analysis system based on unsupervised learning, characterized by, Comprise: Data acquisition module: the power data of large industrial users are collected, the data transmission direction is adjusted according to the installation environment, the data is transmitted to other data acquisition modules through directional wireless transmission, other data acquisition modules continue to transmit data to the designated direction through this way, until the data is transmitted to the designated data collection and transmission module; Collection and transmission module: the received data are summarized, and the data are transmitted to the data preprocessing module; Data preprocessing module: the received data are compared with the previously received data, preliminary analysis is completed, the data are added, the total data value is obtained, and the received data, preliminary analysis results and total data value are sent to the data analysis module; Redundant transmission module: the data of sudden situation are transmitted separately; Data analysis module: the power consumption of large industrial users is analyzed according to the received data, preliminary analysis results and total data value, and the power consumption mode of large industrial users in the future period of time is predicted; Data detection module: the data detection module extracts part of the characteristics of the data before data transmission, compares the extracted characteristics with the corresponding position in the data, and compares the size of the data with the recorded size, the data preprocessing module detects the data before data transmission, and the data analysis module detects the data after data reception, to ensure the integrity and authenticity of the data; Simulation module: the user virtually adjusts the power consumption mode in the future period of time according to the prediction result; Data sharing module: the data received by the data analysis module and the analysis and prediction results are transmitted to other management departments, so that other power management departments can use the data and prediction results as training data of the data analysis module, and receive the power consumption data and analysis and prediction results sent by other management departments and combine the data of the management department to use as training data of the data analysis module; The redundant transmission module comprises: Independent transmission unit: the emergency data in the data acquisition module are transmitted to the arrangement transmission unit; Arrangement transmission unit: the received data are compressed and packaged, and the compressed and packaged data are transmitted to the unmanned aerial vehicle transmission unit; Independent power supply module: the independent transmission unit, arrangement transmission unit and data acquisition module are independently powered, so that the independent transmission unit and arrangement transmission unit can normally operate in the case of power failure; Unmanned aerial vehicle transmission unit: the unmanned aerial vehicle transmission unit flies on the preset line, receives the compressed and packaged data sent by the arrangement transmission unit, returns to the take-off site, and then transmits the compressed and packaged data to the data analysis module.

2. The system for power consumption analysis of large industrial consumers based on unsupervised learning as claimed in claim 1 wherein, The data acquisition module comprises: Acquisition unit: the power data of users are collected; Directional transmission unit: the data are sent through directional transmission; Storage module: the collected raw data are backed up, and the backup is stored.

3. A large industrial consumer electricity consumption analysis system based on unsupervised learning according to claim 2, characterized in that, The storage module comprises: Storage unit: the data are backed up and stored; Deletion unit: the data beyond the storage period are deleted; Time limit recovery unit: recover the misdeletion data within the specified period.

4. The system for power consumption analysis of large industrial consumers based on unsupervised learning as claimed in claim 1 wherein, The data preprocessing module comprises: Data comparison module: compare the received data with the previously received data, analyze the difference between the received data and the previous data; Data calculation module: add the received data to obtain the total of the large industrial user electricity data.

5. The system for power consumption analysis of large industrial consumers based on unsupervised learning as claimed in claim 1 wherein, The unmanned aerial vehicle transmission unit comprises: Unmanned aerial vehicle module: fly in the air along the set route; Navigation module: identify the flight route and identify the obstacles on the route, so that the unmanned aerial vehicle avoids the obstacles; Wireless transmission module: receive the data sent by the data sorting and transmission unit, return to the take-off location, and then send the data to the data analysis module.

6. A large industrial consumer power consumption analysis system based on unsupervised learning according to claim 1, characterized in that, The data analysis module uses a deep learning processor to analyze the data and predict the electricity usage pattern of the large industrial user in the future period; The deep learning processor uses a PyTorch framework and a recurrent neural network.

7. A large industrial consumer electricity consumption analysis system based on unsupervised learning according to claim 1, characterized in that, The data detection module comprises: Feature extraction unit: randomly extract multiple features in the data before sending, and sort them into feature data; Detection unit: compare the received data and the feature data, and compare the size of the data before and after sending; Early warning unit: when the data is different or missing, an alarm is sent to remind the staff.

8. The system for power consumption analysis of large industrial consumers based on unsupervised learning as claimed in claim 1 wherein, The simulation module comprises: Display module: display the prediction results and collected data to enable users to timely understand their electricity usage; Setting module: manage the display module, adjustment module, transceiver module and login management module; Adjustment module: users can virtually adjust and plan their electricity usage in the future period according to the prediction results; Receiving module: receive the prediction results and collected data; Login management module: check the information of the login personnel.

9. An analysis method for a non-supervised learning-based large industrial user electricity analysis system according to any one of claims 1 to 8, comprising the following steps: S1, through a plurality of data collection modules: collect the electricity data of different positions of the large industrial user, and transmit the data to other data collection modules through directional wireless transmission, other data collection modules continue to transmit data to the designated direction through this way, until the data is transmitted to the designated data collection module, reduce the mutual interference between the data collection module and the surrounding electrical equipment during data transmission; S2, the data collection module randomly extracts part of the features in the data before data transmission, records the size of the data, and the data collection module compares the extracted features with the corresponding position in the data after receiving the data, and compares the size of the data with the recorded size, detects the integrity and authenticity of the data; S3, the summary transmission module: the received data is summarized, and the data is transmitted to the data preprocessing module, the received data is compared with the past received data through the data preprocessing module, the difference between the received data and the past data is analyzed, and the received data is added, the sum of the large industrial user electricity data is obtained, then the received data, the preliminary analysis result and the data total value are sent to the data analysis module; S4, the data detection module is also used to detect the integrity and authenticity of the data during transmission; S5, the data analysis module is used to analyze the electricity consumption of large industrial users according to the received data, the preliminary analysis result and the data total value, and to predict the electricity consumption mode of large industrial users in the future; S6, the user adjusts the electricity consumption mode in the future according to the prediction result through the simulation module; S7, the data analysis module receives the data and analysis prediction result, and transmits the data and analysis prediction result to other management departments through the data sharing module, so that other power management departments can use the data and prediction result as training data of the data analysis module, and receive the electricity consumption data and analysis prediction result sent by other management departments and combine the data of the management department, as the training data of the data analysis module.

Images