Intelligent energy scheduling method and system based on Internet of Things

Abstract

The invention discloses an intelligent energy scheduling method and system based on the Internet of Things, and relates to the technical field of the Internet of Things, and the method comprises the steps: calling a historical electricity utilization record recorded by an intelligent electric meter, building and analyzing a power line graph and a current line graph, and carrying out the judgment and extraction of a feature record in the electricity utilization record; predicting a current value of the power supply area at each moment of the to-be-measured date; setting the minimum window duration of each power supply area, and extracting all time windows corresponding to the power supply areas; obtaining a target value of each time window in each power supply area; and obtaining a minimum monitoring time interval, and adjusting the monitoring time interval of each power supply area to complete power monitoring of the power supply area. According to the invention, by analyzing the power utilization record, based on the actual conditions of different power supply areas, the monitoring time interval is intelligently adjusted, the real-time deviation between power supply and power utilization demands is captured, the dynamic change rhythm of load and energy storage can be effectively matched, and the power energy dispatching reliability is improved.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, specifically to an IoT-based intelligent energy dispatching method and system. Background Technology

[0002] With the improvement of residents' living standards and the popularization of smart homes, residents' electricity consumption has shown a continuous upward trend, and the electricity consumption structure has shifted from basic necessities to comfort and intelligence. At the same time, new characteristics such as more complex load fluctuations and more dispersed electricity consumption periods have emerged. These changes have placed higher demands on the dynamic balance of the power grid and the reliability of power supply. By monitoring the power supply and demand status on the load side in real time to identify imbalances and capture the real-time deviation between power supply capacity and power demand, energy dispatch measures can be triggered as early as possible to avoid the imbalance from expanding and causing voltage abnormalities, line overloads, or even power outages. However, because the power grid's electricity consumption is unbalanced over time, when the time interval for monitoring electricity consumption data is fixed, it will lead to the lag in energy storage dispatch during some periods, making it impossible to match the dynamic changes in load and energy storage, and causing problems such as the expansion of the power supply and demand deviation, thus reducing the reliability of power energy dispatch schemes. Summary of the Invention

[0003] The purpose of this invention is to provide an energy intelligent scheduling method and system based on the Internet of Things to solve the problems raised in the prior art.

[0004] To solve the above-mentioned technical problems, the present invention provides the following technical solution: An IoT-based intelligent energy dispatching method includes the following steps: The power grid load side includes several power supply areas, and each power supply area includes several node lines. Smart meters are deployed on each node line. Historical electricity consumption records recorded by the smart meters are retrieved, the electricity consumption time periods of the electricity consumption records are extracted, and the power and current values ​​of several electricity consumption times are obtained according to the preset data sampling interval. Power line graphs and current line graphs are established and analyzed, and characteristic records in the electricity consumption records are judged and extracted. Extract the feature records corresponding to smart meters in the power supply area, obtain the electricity consumption period of the feature records, and predict the current value of the power supply area at each moment on the date to be measured based on the element values ​​of weather factors affecting electricity consumption during the electricity consumption period. The test date is divided into several sub-time periods of uniform duration, and a minimum window duration is set for each power supply area. Based on the minimum window duration and the current value of the power supply area at each moment on the test date, all time windows corresponding to the power supply area are extracted. Based on the current value at each moment within the corresponding time period of each time window, the entropy value of each time window is obtained, and the entropy value is used as the target value of each time window to obtain the target value of each time window in each power supply area. Obtain the available resource usage value of the server, and determine the minimum monitoring time interval based on the number of power supply areas; obtain the maximum monitoring time interval, and adjust the monitoring time interval of each power supply area according to the minimum monitoring time interval and the target value of each time window to complete the power monitoring of the power supply areas.

[0005] Preferably, the feature records in the electricity consumption records are judged and extracted, including: Extract the electricity usage period D from a certain electricity usage record X. X To obtain the electricity consumption time D of the smart meter X Based on the recorded power and current values ​​and the preset data sampling interval, the electricity consumption period D is obtained. X For several electricity consumption times, establish a power line graph based on the power value at each electricity consumption time, and establish a current line graph based on the current value at each electricity consumption time. The dynamic time warping method is used to obtain the DTW distance between the power line graph and the current line graph. If the DTW distance is less than the preset distance threshold and not every element in the power line graph and the current line graph has a value of 0, then the electricity consumption record X is used as a feature record, and then all feature records are obtained.

[0006] Dynamic Time Warping (DTW) is an algorithm used to measure the similarity between two time series. It is an existing technology, and the specific process will not be elaborated here. DTW distance represents the difference between two line graphs over time. The smaller the DTW distance, the smaller the difference between the two line graphs and the greater the similarity. Since power and current values ​​are time series data that fluctuate continuously over time and have matching trends in the time dimension, they are suitable for using DTW distance to measure the similarity between two line graphs. Therefore, the smaller the DTW distance, the more reliable the electricity consumption record is, and it can be used as a feature record.

[0007] Preferably, predicting the current value of the power supply area at each moment on the date to be measured includes: Extract a feature record Y from a smart meter E in a certain power supply area R0, and obtain the electricity consumption period D of feature record Y. Y The electricity consumption period D is obtained. Y The current value at each moment within the period D will be used for electricity consumption. Y The corresponding date is designated as b; through a weather forecasting platform, the element values ​​N of weather factors affecting electricity consumption on the date to be measured are obtained, and based on the element values ​​N of the weather factors on date b... b The deviation value between date b and the date to be tested is obtained as |NN b | Then, the deviation value between each date corresponding to the smart meter E and the date to be measured is calculated. The date weight of each date is obtained according to the rule that the smaller the deviation value, the greater the date weight. The sum of all weights is 1. Based on the date weight and the current value at each moment, the current value of smart meter E at each moment on the date to be measured is calculated. By combining the current values ​​of each smart meter in the power supply area R0, the current value of the power supply area R0 at each moment on the date to be measured is predicted.

[0008] Preferably, setting the minimum window duration for each power supply area includes: dividing the date to be tested into several sub-time periods of uniform duration; extracting several current values ​​of a power supply area R0 in a certain sub-time period; if the variance obtained based on all current values ​​is less than a preset first variance threshold, then the certain sub-time period is taken as the stable time period of the power supply area R0, thereby obtaining the total number of stable time periods for each power supply area, and setting the minimum window duration for each power supply area according to the rule that the larger the total number of stable time periods, the larger the minimum window duration.

[0009] Preferably, all time windows corresponding to the power supply area are extracted, including: Get the minimum window duration D0 of a certain power supply area R0, get the time T1 with duration D0 after midnight, take the time period between midnight and time T1 as P1, calculate the variance S1 of time period P1 based on the current value of the predicted power supply area in time period P1 on the date to be measured, if the variance S1 is greater than the preset second variance threshold, then time period P1 is taken as a time window. If the variance S1 is not greater than the preset second variance threshold, starting from time T1, the earliest time that satisfies the variance being greater than the preset second variance threshold is obtained. The time interval between zero point and the earliest time is taken as a time window to obtain all time windows of power supply area R0 within the test date, and then all time windows of each power supply area within the test date are obtained.

[0010] Preferably, obtaining the target value for each time window in each power supply area includes: extracting a time window W corresponding to a power supply area R0, obtaining the current values ​​at several moments in the time window W, setting several current ranges, classifying the current values ​​into the corresponding current ranges, collecting the number of elements in each current range, obtaining the probability corresponding to each current range, and then obtaining the entropy value of the time window W. Where Q is the number of current ranges, P i Let be the probability of the i-th current range; use the entropy value as the target value of the time window W to obtain the target value of each time window in each power supply area.

[0011] A higher entropy value indicates greater disorder, which significantly increases the potential risk of accidents for the power grid. Therefore, shorter monitoring intervals should be set for power supply areas with higher entropy values, while longer monitoring intervals should be set for power supply areas with lower entropy values. The specific implementation steps are as follows: Preferably, the monitoring time interval for each power supply area is adjusted, including: Obtain the available resource usage value Z of the server, obtain the number of power supply areas N, and ensure that the monitoring time interval is the same for each power supply area. Calculate the monitoring time interval corresponding to when the server's resource usage value equals Z during the monitoring process, and use this as the minimum monitoring time interval G. min ; Obtain the system's preset maximum monitoring time interval G max ; Obtain the time window corresponding to midnight on the date to be tested for each power supply area, and based on the target value of each time window, select the maximum target value M. max The monitoring time interval for the corresponding power supply area is used as G. min Minimum target value M min The monitoring time interval for the corresponding power supply area is used as G. max Based on the target value M0 within a certain time window, use the formula The monitoring time interval G0 of the corresponding power supply area is obtained, and then the monitoring time interval of each power supply area is obtained. According to the monitoring time interval, if the time window of any power supply area changes in the future, the monitoring time interval of the power supply area is recalculated to adjust the monitoring time interval of each power supply area.

[0012] An Internet of Things-based intelligent energy dispatching system includes a feature recording extraction module, a target value calculation module, and a monitoring time interval adjustment module; Feature record extraction module: Used on the load side of the power grid, which includes several power supply areas and several node lines, with smart meters deployed on each node line; retrieves historical electricity consumption records from the smart meters, extracts the electricity consumption periods from the records, obtains power and current values ​​for several electricity consumption times according to the preset data sampling interval, establishes and analyzes power and current line graphs, and judges and extracts feature records from the electricity consumption records; Target value calculation module: used to extract the feature records corresponding to smart meters in the power supply area, obtain the electricity consumption period of the feature records, and predict the current value of the power supply area at each moment on the date to be measured based on the element values ​​of weather factors affecting electricity consumption during the electricity consumption period. The test date is divided into several sub-time periods of uniform duration, and a minimum window duration is set for each power supply area. Based on the minimum window duration and the current value of the power supply area at each moment on the test date, all time windows corresponding to the power supply area are extracted. Based on the current value at each moment within the corresponding time period of each time window, the entropy value of each time window is obtained, and the entropy value is used as the target value of each time window to obtain the target value of each time window in each power supply area. Monitoring time interval adjustment module: used to obtain the available resource usage value of the server, obtain the minimum monitoring time interval based on the number of power supply areas, obtain the maximum monitoring time interval, and adjust the monitoring time interval of each power supply area according to the minimum monitoring time interval and the target value of each time window, so as to complete the power monitoring of the power supply area.

[0013] Preferably, the feature record extraction module includes a feature record extraction unit; Feature record extraction unit: used to extract the electricity consumption period of the electricity consumption record, obtain the power and current values ​​recorded by the smart meter during the electricity consumption period, and establish power line graphs and current line graphs; using the dynamic time warping method, the DTW distance between the power line graph and the current line graph is obtained, and then all feature records are obtained.

[0014] Preferably, the monitoring time interval adjustment module includes a minimum monitoring time interval obtaining unit and a monitoring time interval adjustment unit; Minimum monitoring time interval obtaining unit: used to obtain the available resource usage value Z of the server, obtain the number of power supply areas, make the monitoring time interval the same for each power supply area, and obtain the monitoring time interval corresponding to when the resource usage value of the server is equal to Z during the monitoring process, which is used as the minimum monitoring time interval; Monitoring time interval adjustment unit: used to obtain the time window corresponding to the zero point of the date to be measured for each power supply area, and obtain the monitoring time interval for each power supply area according to the target value of each time window, so as to adjust the monitoring time interval of each power supply area and complete the power monitoring of the power supply area.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides an intelligent energy dispatching method and system based on the Internet of Things, including: retrieving historical electricity consumption records from smart meters, establishing and analyzing power line graphs and current line graphs, and extracting characteristic records from the electricity consumption records; predicting the current value of the power supply area at each moment on the date to be measured; setting the minimum window duration for each power supply area, and extracting all time windows corresponding to the power supply area; obtaining the target value for each time window in each power supply area; obtaining the minimum monitoring time interval, adjusting the monitoring time interval for each power supply area, and completing the power monitoring of the power supply area. This invention, through analysis of electricity consumption records and based on the actual conditions of different power supply areas, intelligently adjusts the monitoring time interval, captures the real-time deviation between power supply and electricity demand, effectively matches the dynamic changes in load and energy storage, and improves the reliability of power energy dispatching. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an IoT-based intelligent energy scheduling method according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example: Figure 1 As shown, this invention provides a technical solution for an intelligent energy dispatching method based on the Internet of Things, comprising the following steps: (1) The power grid load side includes several power supply areas, and each power supply area includes several node lines. Smart meters are deployed on each node line. Historical electricity consumption records recorded by the smart meters are retrieved, and the electricity consumption time periods of the electricity consumption records are extracted. According to the preset data sampling interval, the power and current values ​​of several electricity consumption times are obtained. Power line graphs and current line graphs are established and analyzed, and the characteristic records in the electricity consumption records are judged and extracted.

[0020] In this scheme, the load side includes several power supply areas, such as industrial areas, residential areas, and commercial areas. The amount of electricity consumption in different power supply areas will vary significantly at different times. For example, industrial areas mainly consume electricity for production during the day (peak hours from 8 am to 6 pm), while residential areas have peak electricity consumption at night, and commercial areas have peak consumption during the day and night for lighting. Therefore, if the power supply areas are uniformly monitored without considering the actual electricity demand of each area and without capturing the real-time deviation between power supply and demand, there will be situations where some areas have an oversupply while others have an undersupply.

[0021] Extract the electricity usage period D from a certain electricity usage record X. X To obtain the electricity consumption time D of the smart meter X Based on the recorded power and current values ​​and the preset data sampling interval, the electricity consumption period D is obtained. X For several electricity consumption times, establish a power line graph based on the power value at each electricity consumption time, and establish a current line graph based on the current value at each electricity consumption time. The dynamic time warping method is used to obtain the DTW distance between the power line graph and the current line graph. If the DTW distance is less than the preset distance threshold and not every element in the power line graph and the current line graph has a value of 0, then the electricity consumption record X is used as a feature record, and then all feature records are obtained.

[0022] (2) Extract the feature records corresponding to the smart meters in the power supply area, obtain the electricity consumption period of the feature records, and predict the current value of the power supply area at each moment on the date to be measured based on the element values ​​of the weather factors affecting the electricity consumption during the electricity consumption period.

[0023] Extract a feature record Y from a smart meter E in a certain power supply area R0, and obtain the electricity consumption period D of feature record Y. Y The electricity consumption period D is obtained. Y The current value at each moment within the period D will be used for electricity consumption. Y The corresponding date is designated as b; through a weather forecasting platform, the element values ​​N of weather factors affecting electricity consumption on the date to be measured are obtained, and based on the element values ​​N of the weather factors on date b... b The deviation value between date b and the date to be tested is obtained as |NN b | Then, the deviation value between each date corresponding to the smart meter E and the date to be measured is calculated. The date weight of each date is obtained according to the rule that the smaller the deviation value, the greater the date weight. The sum of all weights is 1. In this embodiment, the weather element is temperature. Since the electricity consumption varies with different temperatures (e.g., more electricity is used in summer and winter, and less in spring and autumn), the date weight of each date can be obtained based on the temperature value. The date weight follows the rule that the closer the temperature values ​​are, the greater the date weight. Based on the date weight, the current value of the smart meter E at each moment on the date to be measured is calculated. Since smart meters are deployed on each node line in the power supply area, the current value of the power supply area R0 at each moment on the date to be measured can be predicted.

[0024] Based on the date weight and the current value at each moment, the current value of smart meter E at each moment on the date to be measured is calculated. By combining the current values ​​of each smart meter in the power supply area R0, the current value of the power supply area R0 at each moment on the date to be measured is predicted.

[0025] (3) Divide the date to be tested into several sub-time periods of uniform duration and set the minimum window duration for each power supply area.

[0026] The test date is divided into several sub-time periods of uniform duration. Several current values ​​of a certain power supply area R0 are extracted in a certain sub-time period. If the variance obtained based on all current values ​​is less than the preset first variance threshold, then the certain sub-time period is taken as the stable time period of the power supply area R0. Then the total number of stable time periods of each power supply area is obtained. According to the rule that the larger the total number of stable time periods, the larger the minimum window duration, the minimum window duration of each power supply area is set.

[0027] (4) Based on the minimum window duration and the current value of the power supply area at each moment on the date to be tested, extract all time windows corresponding to the power supply area.

[0028] Get the minimum window duration D0 of a certain power supply area R0, get the time T1 with duration D0 after midnight, take the time period between midnight and time T1 as P1, calculate the variance S1 of time period P1 based on the current value of the predicted power supply area in time period P1 on the date to be measured, if the variance S1 is greater than the preset second variance threshold, then time period P1 is taken as a time window. If the variance S1 is not greater than the preset second variance threshold, starting from time T1, the earliest time that satisfies the variance being greater than the preset second variance threshold is obtained. The time interval between zero point and the earliest time is taken as a time window to obtain all time windows of power supply area R0 within the test date, and then all time windows of each power supply area within the test date are obtained.

[0029] (5) Based on the current value at each moment within the corresponding time period of each time window, the entropy value of each time window is obtained, and the entropy value is used as the target value of each time window to obtain the target value of each time window in each power supply area.

[0030] Extract a time window W corresponding to a certain power supply area R0, obtain the current values ​​at several moments within the time window W, set several current ranges, classify the current values ​​into the corresponding current ranges, collect the number of elements in each current range, obtain the probability corresponding to each current range, and then obtain the entropy value of the time window W. Where Q is the number of current ranges, P i Let be the probability of the i-th current range; use the entropy value as the target value of the time window W to obtain the target value of each time window in each power supply area.

[0031] A higher entropy value indicates greater disorder, which significantly increases the potential risk of accidents for the power grid. Therefore, shorter monitoring intervals should be set for power supply areas with higher entropy values, while longer monitoring intervals should be set for power supply areas with lower entropy values. The specific implementation steps are as follows: (6) Obtain the available resource usage value of the server, and obtain the minimum monitoring time interval based on the number of power supply areas; obtain the maximum monitoring time interval, and adjust the monitoring time interval of each power supply area based on the minimum monitoring time interval and the target value of each time window to complete the power monitoring of the power supply area.

[0032] Obtain the available resource usage value Z of the server, obtain the number of power supply areas N, and ensure that the monitoring time interval is the same for each power supply area. Calculate the monitoring time interval corresponding to when the server's resource usage value equals Z during the monitoring process, and use this as the minimum monitoring time interval G. min ; Obtain the system's preset maximum monitoring time interval G max ; Obtain the time window corresponding to midnight on the date to be tested for each power supply area, and based on the target value of each time window, select the maximum target value M. max The monitoring time interval for the corresponding power supply area is used as G. min Minimum target value M min The monitoring time interval for the corresponding power supply area is used as G. max Based on the target value M0 within a certain time window, use the formula The monitoring time interval G0 of the corresponding power supply area is obtained, and then the monitoring time interval of each power supply area is obtained. According to the monitoring time interval, if the time window of any power supply area changes in the future, the monitoring time interval of the power supply area is recalculated to adjust the monitoring time interval of each power supply area.

[0033] Here is an example: Set the maximum target value M max The minimum monitoring interval is 5. min The minimum objective value is 3. min The maximum monitoring time interval is 2. max The monitoring interval is set to 3 if the target value of a power supply area is 5, 10 if the target value is 2, 5.33 if the target value is 4, and 7.67 if the target value is 3. If the time window of any power supply area changes, the monitoring interval is recalculated. This process is repeated to effectively match the dynamic changes in load and energy storage, enabling intelligent monitoring and scheduling, and improving the reliability of power energy dispatch.

[0034] This embodiment also provides an IoT-based intelligent energy dispatching system, including a feature record extraction module, a target value calculation module, and a monitoring time interval adjustment module. The feature record extraction module includes a feature record extraction unit, and the monitoring time interval adjustment module includes a minimum monitoring time interval acquisition unit and a monitoring time interval adjustment unit. When the system executes the computer program, it implements the above-mentioned IoT-based intelligent energy dispatching method. Since the IoT-based intelligent energy dispatching method has been described in detail above, it will not be repeated here.

[0035] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0036] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0037] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An energy intelligent dispatching method based on the Internet of Things, characterized in that, Includes the following steps: The power grid load side includes several power supply areas, and each power supply area includes several node lines. Smart meters are deployed on each node line. Historical electricity consumption records recorded by the smart meters are retrieved, the electricity consumption time periods of the electricity consumption records are extracted, and the power and current values ​​of several electricity consumption times are obtained according to the preset data sampling interval. Power line graphs and current line graphs are established and analyzed, and characteristic records in the electricity consumption records are judged and extracted. Extract the feature records corresponding to smart meters in the power supply area, obtain the electricity consumption period of the feature records, and predict the current value of the power supply area at each moment on the date to be measured based on the element values ​​of weather factors affecting electricity consumption during the electricity consumption period. The test date is divided into several sub-time periods of uniform duration, and a minimum window duration is set for each power supply area. Based on the minimum window duration and the current value of the power supply area at each moment on the test date, all time windows corresponding to the power supply area are extracted. Based on the current value at each moment within the corresponding time period of each time window, the entropy value of each time window is obtained, and the entropy value is used as the target value of each time window to obtain the target value of each time window in each power supply area. Obtain the available resource usage of the server and determine the minimum monitoring time interval based on the number of power supply areas; The maximum monitoring time interval is obtained, and the monitoring time interval for each power supply area is adjusted according to the minimum monitoring time interval and the target value for each time window, thus completing the power monitoring of the power supply area.

2. The energy intelligent scheduling method based on the Internet of Things according to claim 1, characterized in that, The system extracts and identifies key features from electricity usage records, including: Extract the electricity usage period D from a certain electricity usage record X. X To obtain the electricity consumption time D of the smart meter X Based on the recorded power and current values ​​and the preset data sampling interval, the electricity consumption period D is obtained. X For several electricity consumption times, establish a power line graph based on the power value at each electricity consumption time, and establish a current line graph based on the current value at each electricity consumption time. Using the dynamic time warping method, the DTW distance between the power line graph and the current line graph is obtained. If the DTW distance is less than a preset distance threshold and not every element in the power line graph and the current line graph has a value of 0, then the electricity consumption record X is used as a feature record, and all feature records are obtained.

3. The energy intelligent scheduling method based on the Internet of Things according to claim 1, characterized in that, Predict the current value of the power supply area at each moment on the date to be measured, including: Extract a feature record Y from a smart meter E in a certain power supply area R0, and obtain the electricity consumption period D of feature record Y. Y The electricity consumption period D is obtained. Y The current value at each moment within the period D will be used for electricity consumption. Y The corresponding date is designated as b; through a weather forecasting platform, the element values ​​N of weather factors affecting electricity consumption on the date to be measured are obtained, and based on the element values ​​N of the weather factors on date b... b The deviation value between date b and the date to be tested is obtained as |NN b | Then, the deviation value between each date corresponding to the smart meter E and the date to be measured is calculated. The date weight of each date is obtained according to the rule that the smaller the deviation value, the greater the date weight. The sum of all weights is 1. Based on the date weight and the current value at each moment, the current value of smart meter E at each moment on the date to be measured is calculated. By combining the current values ​​of each smart meter in the power supply area R0, the current value of the power supply area R0 at each moment on the date to be measured is predicted.

4. The energy intelligent dispatching method based on the Internet of Things according to claim 1, characterized in that, Setting the minimum window duration for each power supply area includes: dividing the date to be tested into several sub-time periods of uniform duration; extracting several current values ​​of a power supply area R0 in a certain sub-time period; if the variance obtained based on all current values ​​is less than a preset first variance threshold, then the certain sub-time period is taken as the stable time period of the power supply area R0, thereby obtaining the total number of stable time periods for each power supply area, and setting the minimum window duration for each power supply area according to the rule that the larger the total number of stable time periods, the larger the minimum window duration.

5. The energy intelligent scheduling method based on the Internet of Things according to claim 4, characterized in that, Extract all time windows corresponding to the power supply area, including: Get the minimum window duration D0 of a certain power supply area R0, get the time T1 with duration D0 after midnight, take the time period between midnight and time T1 as P1, calculate the variance S1 of time period P1 based on the current value of the predicted power supply area in time period P1 on the date to be measured, if the variance S1 is greater than the preset second variance threshold, then time period P1 is taken as a time window. If the variance S1 is not greater than the preset second variance threshold, starting from time T1, the earliest time that satisfies the variance being greater than the preset second variance threshold is obtained, and the time period between zero point and the earliest time is taken as a time window to obtain all time windows of power supply area R0 within the date to be tested, and then all time windows of each power supply area within the date to be tested are obtained.

6. The energy intelligent scheduling method based on the Internet of Things according to claim 1, characterized in that, Obtaining the target value for each time window in each power supply region includes: extracting a time window W corresponding to a power supply region R0; acquiring the current values ​​at several moments within the time window W; setting several current ranges; assigning the current values ​​to the corresponding current ranges; collecting the number of elements within each current range; obtaining the probability corresponding to each current range; and finally obtaining the entropy value of the time window W. Where Q is the number of current ranges, P i Let be the probability of the i-th current range; use the entropy value as the target value of the time window W to obtain the target value of each time window in each power supply area.

7. The energy intelligent dispatching method based on the Internet of Things according to claim 1, characterized in that, The monitoring time intervals for each power supply area were adjusted, including: Obtain the available resource usage value Z of the server, obtain the number of power supply areas N, and ensure that the monitoring time interval is the same for each power supply area. Calculate the monitoring time interval corresponding to when the server's resource usage value equals Z during the monitoring process, and use this as the minimum monitoring time interval G. min ; Obtain the system's preset maximum monitoring time interval G max ; Obtain the time window corresponding to midnight on the date to be tested for each power supply area, and based on the target value of each time window, select the maximum target value M. max The monitoring time interval for the corresponding power supply area is used as G. min Minimum target value M min The monitoring time interval for the corresponding power supply area is used as G. max Based on the target value M0 within a certain time window, use the formula The monitoring time interval G0 of the corresponding power supply area is obtained, and then the monitoring time interval of each power supply area is obtained. According to the monitoring time interval, if the time window of any power supply area changes in the future, the monitoring time interval of the power supply area is recalculated to adjust the monitoring time interval of each power supply area.

8. An intelligent energy dispatching system, used to execute the Internet of Things-based intelligent energy dispatching method according to any one of claims 1-7, characterized in that, The system includes a feature record extraction module, a target value calculation module, and a monitoring time interval adjustment module; Feature record extraction module: Used on the load side of the power grid, which includes several power supply areas and several node lines, with smart meters deployed on each node line; retrieves historical electricity consumption records from the smart meters, extracts the electricity consumption periods from the records, obtains power and current values ​​for several electricity consumption times according to the preset data sampling interval, establishes and analyzes power and current line graphs, and judges and extracts feature records from the electricity consumption records; Target value calculation module: used to extract the feature records corresponding to smart meters in the power supply area, obtain the electricity consumption period of the feature records, and predict the current value of the power supply area at each moment on the date to be measured based on the element values ​​of weather factors affecting electricity consumption during the electricity consumption period. The test date is divided into several sub-time periods of uniform duration, and a minimum window duration is set for each power supply area. Based on the minimum window duration and the current value of the power supply area at each moment on the test date, all time windows corresponding to the power supply area are extracted. Based on the current value at each moment within the corresponding time period of each time window, the entropy value of each time window is obtained, and the entropy value is used as the target value of each time window to obtain the target value of each time window in each power supply area. Monitoring time interval adjustment module: used to obtain the available resource usage value of the server, obtain the minimum monitoring time interval based on the number of power supply areas, obtain the maximum monitoring time interval, and adjust the monitoring time interval of each power supply area according to the minimum monitoring time interval and the target value of each time window, so as to complete the power monitoring of the power supply area.

9. The intelligent energy dispatching system according to claim 8, characterized in that, The feature record extraction module includes a feature record extraction unit; Feature record extraction unit: used to extract the electricity consumption period of the electricity consumption record, obtain the power and current values ​​recorded by the smart meter during the electricity consumption period, and establish power line graphs and current line graphs; using the dynamic time warping method, the DTW distance between the power line graph and the current line graph is obtained, and then all feature records are obtained.

10. An intelligent energy dispatching system according to claim 8, characterized in that, The monitoring time interval adjustment module includes a minimum monitoring time interval acquisition unit and a monitoring time interval adjustment unit; Minimum monitoring time interval obtaining unit: used to obtain the available resource usage value Z of the server, obtain the number of power supply areas, make the monitoring time interval the same for each power supply area, and obtain the monitoring time interval corresponding to when the resource usage value of the server is equal to Z during the monitoring process, which is used as the minimum monitoring time interval; Monitoring time interval adjustment unit: used to obtain the time window corresponding to the zero point of the date to be measured for each power supply area, and obtain the monitoring time interval for each power supply area according to the target value of each time window, so as to adjust the monitoring time interval of each power supply area and complete the power monitoring of the power supply area.

Images