A distribution network load forecasting method and system based on multi-source information fusion

Abstract

The present application discloses a distribution network load prediction method and system based on multi-source information fusion, which obtains the human environment information and power regulation information of a future set time period in a region to be predicted, and divides the future set time period into a number of time units according to a preset minimum time unit, and then determines the human environment information and power regulation information corresponding to each time unit, and inputs the human environment information and power regulation information corresponding to each time unit into a prediction model created in advance based on historical load data, historical human environment information and historical power regulation information, so as to obtain the predicted load data corresponding to each time unit, and use the predicted load data corresponding to each time unit to merge and obtain the predicted load data in the future time period of the region to be predicted, thereby realizing the prediction of the power grid load in the region to be predicted.

Description

Technical Field

[0001] The present application relates to the field of electric power operation and maintenance technology, and in particular to a distribution network load prediction method and system based on multi-source information fusion. Background Art

[0002] With the continuous development of the power system, the impact of power supply on society is increasing, and the allocation of power is also crucial. At present, the load forecasting of the power grid is an important part of the allocation of power. The load forecasting of the power grid mainly uses the data collected by the power grid system to predict the future power load. Through the prediction results, the start and stop of the power generating units in the power grid can be arranged more reasonably, and a reasonable power supply construction plan can be formulated to reduce unnecessary reserve capacity. At the same time, the maintenance of the units can be arranged reasonably to ensure the normal life and production of the society and effectively reduce the cost of power generation. Therefore, how to realize the load forecasting of the power grid has always been a research direction. Summary of the invention

[0003] In view of this, the present application provides a distribution network load prediction method and system based on multi-source information fusion to facilitate the prediction of power grid load.

[0004] In order to achieve the above objectives, the proposed solution is as follows:

[0005] A distribution network load forecasting method based on multi-source information fusion, comprising:

[0006] Acquire humanistic environment information and power regulation information within a set time period in the future of the predicted area, wherein the humanistic environment information includes geographical information, meteorological information, and humanistic information that affects power use, and the power regulation information includes electricity prices and regulation information that affects power use;

[0007] Divide the future set time period into several time units according to the preset minimum time unit;

[0008] Determine the human environment information and power regulation information corresponding to each time unit;

[0009] Inputting the human environment information and power regulation information corresponding to each time unit into a pre-created prediction model to obtain predicted load data corresponding to each time unit, wherein the prediction model is created based on historical load data, historical human environment information, and historical power regulation information;

[0010] The forecast load data corresponding to each time unit is combined to obtain the forecast load data for the future set time period in the area to be forecasted.

[0011] Optionally, also include:

[0012] Obtain the actual load data corresponding to each preset time unit within the historical set time period of the area to be predicted;

[0013] Determine whether the prediction model needs to be updated based on the actual load data and predicted load data corresponding to each preset time unit within the historical set time period of the area to be predicted;

[0014] If the prediction model needs to be updated, the parameters of the prediction model are updated with the goal of making the predicted load data close to the actual load data.

[0015] Optionally, determining whether the prediction model needs to be updated according to actual load data and predicted load data corresponding to each preset time unit within a historical set time period of the area to be predicted includes:

[0016] Adding the actual load data corresponding to each preset time unit in the historical set time period of the area to be predicted to obtain the sum of the actual load data;

[0017] Add the forecast load data corresponding to each preset time unit in the historical set time period of the area to be forecasted to obtain the sum of the forecast load data;

[0018] Subtracting the sum of the actual load data from the sum of the predicted load data to obtain a load data difference;

[0019] Determining whether the load data difference is within a preset range;

[0020] If not, it is determined that the prediction model needs to be updated.

[0021] Optionally, the process of creating the prediction model includes:

[0022] Obtain load data, human environment information and power control information corresponding to each preset time unit within the historical set time period;

[0023] The load data, human environment information and power control information corresponding to each preset time unit within the historical set time period are integrated to obtain integrated data;

[0024] A prediction model is established using the fused data.

[0025] Optionally, before obtaining the load data, cultural environment information and power regulation information corresponding to each preset time unit within the historical set time period, it also includes:

[0026] Obtain load data corresponding to each preset time unit within a historical set time period transmitted by multiple data sources;

[0027] The load data corresponding to each preset time unit within the historical set time period transmitted by multiple data sources are integrated and processed to obtain the processed load data corresponding to each preset time unit within the historical set time period.

[0028] Optionally, also include:

[0029] The processed load data corresponding to each preset time unit in the historical set time period is transmitted back to the multiple data sources, so that each data source can correct the historical load data according to the processed load data corresponding to each preset time unit in the historical set time period.

[0030] Optionally, after obtaining the load data corresponding to each preset time unit within the historical set time period, the method further includes:

[0031] Determine whether the load data corresponding to each preset time unit within the historical set time period meets the preset range;

[0032] If there is load data corresponding to a preset time unit that does not conform to the preset interval, for each load data that does not conform to the preset interval, a corrected load data is obtained based on the load data corresponding to other preset time units within the historical set time period to replace the load data that does not conform to the preset interval.

[0033] Optionally, for each load data that does not conform to the preset interval, the corrected load data is obtained according to the load data corresponding to other preset time units within the historical set time period, including:

[0034] For each load data that does not conform to the preset interval, searching, from the preset time units corresponding to the load data that conform to the preset interval, two preset time units that are closest in time to the preset time units corresponding to the load data that does not conform to the preset interval;

[0035] The load data corresponding to the two preset time units are added together to obtain an average value to obtain the corrected load data.

[0036] Optionally, also include:

[0037] Displays the forecast load data for the area to be forecasted within the set time period in the future.

[0038] A distribution network load forecasting system based on multi-source information fusion, comprising:

[0039] An information acquisition module, used to acquire humanistic environment information and power regulation information within a set time period in the future of the predicted area, wherein the humanistic environment information includes geographical information, meteorological information, and humanistic information that affects power use, and the power regulation information includes electricity prices and regulation information that affects power use;

[0040] The time division module is used to divide the future set time period into several time units according to the preset minimum time unit;

[0041] An information determination module, used to determine the human environment information and power regulation information corresponding to each time unit;

[0042] A load forecasting module, which is used to input the human environment information and power regulation information corresponding to each time unit into a pre-created forecasting model to obtain forecast load data corresponding to each time unit, wherein the forecasting model is created based on historical load data, historical human environment information and historical power regulation information;

[0043] The data merging module is used to use the forecast load data corresponding to each time unit to merge and obtain the forecast load data for the future time period of the area to be forecasted.

[0044] It can be seen from the above technical scheme that the embodiment of the present application provides a distribution network load forecasting method based on multi-source information fusion, which obtains the cultural environment information and power regulation information of the future set time period in the area to be predicted, and divides the future set time period into several time units according to the preset minimum time unit, and then determines the cultural environment information and power regulation information corresponding to each time unit, and inputs the cultural environment information and power regulation information corresponding to each time unit into the prediction model created in advance based on historical load data, historical cultural environment information and historical power regulation information, so as to obtain the predicted load data corresponding to each time unit, and use the predicted load data corresponding to each time unit to merge and obtain the predicted load data in the future time period of the area to be predicted, thereby realizing the prediction of the power grid load in the area to be predicted. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0046] Figure 1 A flow chart of a distribution network load forecasting method based on multi-source information fusion provided in an embodiment of the present application;

[0047] Figure 2 A flowchart of an optional method for creating a prediction model provided in an embodiment of the present application;

[0048] Figure 3 A schematic diagram of the structure of a distribution network load forecasting system based on multi-source information fusion provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] Figure 1 A flow chart of a distribution network load forecasting method based on multi-source information fusion provided in an embodiment of the present application, the method may include the following steps:

[0051] Step S100: Obtaining cultural environment information and power control information of the area to be predicted within a set time period in the future.

[0052] The above-mentioned human environment information includes geographical information, meteorological information and human information that affects the use of electricity, and the electricity regulation information includes electricity prices and regulation information that affects the use of electricity.

[0053] Specifically, the geographic information may be the geographical location of the area to be predicted, such as a coastal area or an inland area; or the development status of the area to be predicted. Meteorological information may include temperature, humidity, precipitation probability, wind force level, etc. Humanistic information that affects electricity use may include sports events or concerts held in the area to be predicted, and may also include weekdays, weekends, or holidays. Electricity regulation information may include electricity prices and regulation information that affects electricity use, wherein the regulation information that affects electricity use may include relevant information such as plans to promote the development of new energy electric vehicles in the area to be predicted.

[0054] Step S101: Divide a future set time period into a number of time units according to a preset minimum time unit.

[0055] Specifically, in the future set time period, a number of time units are divided by a preset minimum time unit. For example, when the future set time period is a week in the future, the minimum time unit can be preset as one day, and then the week in the future is divided into 7 time units according to the minimum time unit.

[0056] Step S102: Determine the cultural environment information and power regulation information corresponding to each time unit.

[0057] Specifically, through the several time units determined in the above steps, according to each time unit, the corresponding human environment information and power regulation information in each time unit is determined.

[0058] For example, based on the minimum time unit of one day, the next week is divided into seven time units. The first time unit may include cultural environment information: "coastal cities", "comprehensive index of regional development", "heavy rain", "concerts held at the sports center" and "Saturday", as well as power regulation information: "electricity price 0.65 yuan / kWh" and "the region vigorously promotes the development of new energy vehicles".

[0059] Step S103: inputting the human environment information and power regulation information corresponding to each time unit into a pre-created prediction model to obtain the predicted load data corresponding to each time unit.

[0060] Among them, the above prediction model is created based on historical load data, historical cultural environment information and historical power regulation information.

[0061] Specifically, after determining the human environment information and power regulation information corresponding to each time unit, a pre-created prediction model may be input to obtain the predicted load data corresponding to each time unit.

[0062] Step S104: using the forecast load data corresponding to each time unit, the forecast load data for the area to be forecasted within a future set time period is combined to obtain forecast load data.

[0063] Specifically, through the above steps, each time unit is taken as the minimum prediction unit to obtain the predicted load data corresponding to each time unit. The predicted load data corresponding to each time unit are merged to obtain the predicted load data for the future set time period of the area to be predicted.

[0064] It can be seen from the above embodiments that the embodiment of the present application provides a distribution network load forecasting method based on multi-source information fusion, which obtains the cultural environment information and power regulation information of the future set time period in the area to be predicted, and divides the future set time period into several time units according to the preset minimum time unit, and then determines the cultural environment information and power regulation information corresponding to each time unit, and inputs the cultural environment information and power regulation information corresponding to each time unit into the prediction model created in advance based on historical load data, historical cultural environment information and historical power regulation information, so as to obtain the predicted load data corresponding to each time unit, and use the predicted load data corresponding to each time unit to merge and obtain the predicted load data in the future time period of the area to be predicted, thereby realizing the prediction of the power grid load in the area to be predicted.

[0065] In some embodiments of the present application, Figure 2 As shown, a creation process of an optional prediction model provided in an embodiment of the present application is described, and the creation process may include the following steps:

[0066] Step S200: Obtain load data, cultural environment information and power regulation information corresponding to each preset time unit within a historical set time period.

[0067] Specifically, human environment information can be obtained from the corresponding geographic information system, meteorological information system and public sector data.

[0068] Step S201: Fusing the load data, cultural environment information and power control information corresponding to each preset time unit within a historical set time period to obtain fused data.

[0069] Specifically, after obtaining the load data, cultural environment information and power regulation information corresponding to each preset time unit in the historical set time period through the above steps, a data set can be generated for each preset time unit, and each data set will contain the load data, cultural environment information and power regulation information corresponding to the preset time unit. Then all data sets are fused to obtain the final fused data.

[0070] Step S202: Establish a prediction model using the fused data.

[0071] Specifically, after obtaining the fused data, the fused data can be used to create a prediction model. The created prediction model can be a mathematical model, which inputs human environment information and power regulation information and can output a predicted load data. The prediction model can be a linear model, a nonlinear model, or a hybrid model.

[0072] In the above embodiment, the prediction model is created by combining data from multiple aspects, so that when the prediction model is used to predict the load, the impact of human environment information and power regulation information on the load data can be comprehensively considered, thereby predicting the load data more accurately.

[0073] In some embodiments of the present application, the load data corresponding to each preset time unit in the historical set time period can be obtained from multiple data sources. The following is an introduction to a method for obtaining the load data corresponding to each preset time unit in the historical set time period, which may include:

[0074] S11, obtaining load data corresponding to each preset time unit within a historical set time period transmitted by multiple data sources.

[0075] Specifically, multiple data sources may include: a distribution automation system, a power consumption information collection system, and a load monitoring system. Among them, each data source can record the load data corresponding to each preset time unit within the historical set time period from different perspectives. For example, the distribution automation system can record the power distribution conditions of each preset time unit within the historical set time period in the area to be predicted from the perspective of power grid distribution; the power consumption information collection system can record the power consumption conditions of each preset time unit within the historical set time period in the area to be predicted from the perspective of user power consumption; and the load monitoring system directly monitors the load in the area to be predicted. Relevant load data can be obtained through power distribution conditions, power consumption conditions, and load monitoring data.

[0076] S12, integrating and processing the load data corresponding to each preset time unit within the historical set time period transmitted by multiple data sources to obtain processed load data corresponding to each preset time unit within the historical set time period.

[0077] Specifically, multiple load data corresponding to each preset time unit within a historical set time period can be obtained through multiple data sources, and the obtained load data can be integrated and processed to obtain the processed load data corresponding to each preset time unit within the historical set time period.

[0078] When each data source collects data, the collected data may be abnormal due to interference or failure of the device. Therefore, the integration process can be to first determine whether the load data corresponding to each preset time unit in the historical set time period transmitted by each data source meets the preset range. If it exceeds the preset range, the load data can be considered as abnormal data and needs to be excluded.

[0079] After eliminating all abnormal data, for each preset time unit, a load data is selected from the corresponding load data by random selection as the final processed load data. The corresponding load data can also be averaged by adding up all the corresponding load data to obtain an average load data as the final processed load data.

[0080] For example, the historical setting time is last week, the preset time unit is one day, and the load data corresponding to Monday of last week obtained through three data sources are distribution automation system: 100000, power consumption information collection system: 600 and load monitoring system: 620. At this time, the preset range is 400-750. By comparing the load data of the three data sources on Monday of last week with the preset range, it can be determined that the data 100000 collected by the distribution automation system exceeds the preset range of 400-750, so the data collected by the distribution automation system needs to be excluded, so that the load data corresponding to Monday of last week also includes power consumption information collection system: 600 and load monitoring system: 620. At this time, one of 600 and 620 can be randomly selected as the load data corresponding to Monday of last week, or 600 and 620 can be added to get the average value, and 610 can be used as the load data corresponding to Monday of last week.

[0081] In the above embodiment, load data collected from multiple data sources are used to obtain load data corresponding to each preset time unit within a historical set time period, which can effectively avoid inaccurate data when a device fails. In addition, the load data obtained from multiple data sources are integrated. Since each data source obtains load data from a different angle, the accuracy of the data can be further improved.

[0082] Furthermore, after obtaining the load data corresponding to each preset time unit within the processed historical set time period, the load data corresponding to each preset time unit within the processed historical set time period can be transmitted back to multiple data sources, so that each data source can correct the historical load data according to the load data corresponding to each preset time unit within the processed historical set time period.

[0083] Specifically, when each data source collects data, the collected data may be abnormal due to interference or failure of the equipment. For abnormal data, the load data corresponding to each preset time unit within the historical set time period after processing can be used to correct the abnormal data. After the abnormal data is corrected, it can be avoided that the abnormal data needs to be processed again when the load data of the data source is needed for other needs in the future, resulting in repeated work.

[0084] Furthermore, the load data corresponding to each preset time unit in the historical set time period obtained through the above embodiment can also be used in other fields, such as judging the economic strength of a region through power load data.

[0085] In some embodiments of the present application, the load data corresponding to each preset time unit within the historical set time period used to create the prediction model can be screened and corrected to avoid excessive deviation in the prediction results after the prediction model is created due to data anomalies. The following is an introduction to the method for processing the load data corresponding to each preset time unit within the historical set time period, which may include:

[0086] S21 , determining whether the load data corresponding to each preset time unit within the historical set time period conforms to the preset range.

[0087] Specifically, a corresponding preset interval can be set in advance according to the load data of the area to be predicted, and the load data corresponding to each preset time unit in the historical set time period is compared with the preset interval. If the load data corresponding to the preset time unit does not conform to the preset interval, execute S22.

[0088] S22. For each load data that does not conform to the preset interval, a corrected load data is obtained according to the load data corresponding to other preset time units within the historical set time period, and is used to replace the load data that does not conform to the preset interval.

[0089] Specifically, for each load data that does not meet the preset interval, two preset time units that are closest in time to the preset time units corresponding to the load data that does not meet the preset interval can be found from the preset time units corresponding to the load data that meets the preset interval, and the load data corresponding to the two preset time units found can be added and averaged to obtain the corrected load data; the minimum time period can also be set in advance, and from the preset time units corresponding to the load data that meets the preset interval, two preset time units that are at the same time position in the minimum time period and closest in time to the preset time unit corresponding to the load data that does not meet the preset interval can be found, and the load data corresponding to the two preset time units found can be added and averaged to obtain the corrected load data.

[0090] For example, the historical setting time is from March 1, 2021 to March 21, 2021, and the preset time unit is one day. At this time, 21 load data between March 1 and March 21 can be obtained. Assuming that the preset interval is used to judge the 21 load data, it is found that the data on March 12 is abnormal load data. For the abnormal load data on March 12, the preset time units March 1 to March 11 and March 13 to March 21 corresponding to the load data that does not meet the preset interval can be used to find the two preset time units March 11 and March 13 that are closest to the preset time unit March 12 corresponding to the load data that does not meet the preset interval, and the load data corresponding to March 11 and March 13 are added to take the average value to obtain the corrected load data; the minimum time can also be preset The time cycle is one week. At this time, for the three time periods of March 1 to March 7, March 8 to March 14, and March 15 to March 21, the preset time units March 1 to March 11 and March 13 to March 21 corresponding to the load data that meets the preset interval can be used to find the two preset time units that are at the same time position in the minimum time cycle and are closest to the preset time unit March 12 corresponding to the load data that does not meet the preset interval. Since March 12 is on the fifth day of the time cycle from March 8 to March 14, the corresponding March 5 and March 19 can be found from the two similar time periods March 1 to March 7 and March 15 to March 21, and the load data corresponding to March 5 and March 19 can be added and averaged to obtain the corrected load data.

[0091] In the above embodiment, since the load data used to create the prediction model may become abnormal due to the failure of the collection equipment, the load data corresponding to other preset time units within the historical set time period is used to obtain corrected load data to replace the load data that does not conform to the preset interval. This can avoid the impact of abnormal load data on the prediction model to a certain extent.

[0092] In some embodiments of the present application, since the generation of actual load data has a certain degree of randomness, the prediction of load data cannot be absolutely accurate, and it is almost inevitable that there will be deviations between the predicted load data and the actual load data each time. It is unreasonable to judge whether the prediction model is accurate only by comparing a single predicted load data with the corresponding actual load data. Therefore, it is necessary to combine the predicted load data and the actual load data over a period of time to make a more reasonable and accurate judgment on the prediction model. At this time, if it is judged that the prediction model needs to be updated, the predicted load data and the actual load data over a period of time can be combined to update the relevant parameters of the prediction model. The update process of the prediction model is introduced below, and the process may include:

[0093] S31. Obtain actual load data corresponding to each preset time unit within a historical set time period of the area to be predicted.

[0094] Specifically, the acquisition of actual load data in this step can be implemented by adopting the acquisition method of S11-S12 in the above embodiment. Please refer to the above introduction for details, which will not be repeated here.

[0095] S32. Determine whether the prediction model needs to be updated based on the actual load data and the predicted load data corresponding to each preset time unit within the historical set time period of the area to be predicted.

[0096] Specifically, after obtaining the actual load data and predicted load data corresponding to each preset time unit in the historical set time period of the area to be predicted in this step, there are many ways to determine whether the prediction model needs to be updated. The specifics can be referred to the following embodiments, which will not be expanded here. If it is determined that the prediction model needs to be updated, S33 is executed.

[0097] S33. With the goal of making the predicted load data close to the actual load data, update the parameters of the prediction model.

[0098] Specifically, the prediction model may be updated with the goal of making the predicted load data approach the actual load data, so that each predicted load data gradually approaches the corresponding actual load data.

[0099] In the above embodiment, the prediction result of the prediction model is judged by using the actual load data and predicted load data corresponding to each preset time unit within the historical set time period of the area to be predicted. When it is judged that the prediction model needs to be updated, the actual load data and predicted load data corresponding to each preset time unit within the historical set time period can be combined, thereby improving the accuracy of the judgment to a certain extent and avoiding unnecessary updates of the prediction model.

[0100] In the above embodiment, there are multiple ways to determine whether the prediction model needs to be updated according to the actual load data and the predicted load data corresponding to each preset time unit in the historical set time period of the area to be predicted in S32. This embodiment introduces several optional implementation methods, as follows:

[0101] The first method is to determine whether the prediction model needs to be updated based on the accumulated difference between the actual load data and the predicted load data corresponding to each preset time unit within the historical set time period.

[0102] Specifically, the actual load data corresponding to each preset time unit in the historical set time period of the area to be predicted can be added to obtain the sum of the actual load data, and then the predicted load data corresponding to each preset time unit in the historical set time period of the area to be predicted can be added to obtain the sum of the predicted load data, and finally the sum of the actual load data can be subtracted from the sum of the predicted load data to obtain the load data difference, and whether it is within the preset range can be determined according to the difference. If the difference is not within the preset range, it can be considered that the difference is too large, and at this time, the prediction model needs to be updated.

[0103] The second method is to determine whether the prediction model needs to be updated based on the difference trend between the actual load data and the predicted load data corresponding to each preset time unit within the historical set time period.

[0104] Specifically, the actual load data and the predicted load data corresponding to each preset time unit in the historical set time period of the area to be predicted can be subtracted to obtain the difference corresponding to each preset time unit. At this time, it is judged whether the difference corresponding to the preset time unit in the historical set time period will show a uniform deviation trend. For example, in the historical set time period, it is found that the predicted load data corresponding to several consecutive preset time units are higher than the corresponding actual load data. At this time, it can be judged that the prediction model needs to be updated.

[0105] In some embodiments of the present application, in order to display the predicted load data more intuitively, the predicted load data for the area to be predicted within a set time period in the future can be displayed, so that the predicted load data is presented in a visual form.

[0106] Specifically, the displayed content may be the load forecast results for different areas and different time periods in a region. The load forecast results for different time periods may be the total forecast load data for a future set time period; or the forecast load data corresponding to each time unit in a future set time period; or the forecast load data corresponding to each time period in a future set time period, and the time period may be set according to the user's settings.

[0107] A distribution network load forecasting system based on multi-source information fusion provided in an embodiment of the present application is described below. The distribution network load forecasting system based on multi-source information fusion described below and the distribution network load forecasting method based on multi-source information fusion described above can refer to each other.

[0108] refer to Figure 3 As shown, Figure 3 A schematic diagram of the structure of a distribution network load forecasting system based on multi-source information fusion provided in an embodiment of the present application, the system may include:

[0109] The information acquisition module 10 is used to acquire the human environment information and power regulation information of the area to be predicted within a set time period in the future, wherein the human environment information includes geographical information, meteorological information, and human information that affects power use, and the power regulation information includes electricity prices and regulation information that affects power use;

[0110] The time division module 20 is used to divide the future set time period into a number of time units according to a preset minimum time unit;

[0111] An information determination module 30, for determining the human environment information and power regulation information corresponding to each time unit;

[0112] The load forecasting module 40 is used to input the human environment information and power regulation information corresponding to each time unit into a pre-created forecasting model to obtain forecast load data corresponding to each time unit, wherein the forecasting model is created based on historical load data, historical human environment information and historical power regulation information;

[0113] The data merging module 50 is used to use the forecast load data corresponding to each time unit to merge and obtain the forecast load data for the future time period of the area to be forecasted.

[0114] It can be seen from the above technical solution that the embodiment of the present application provides a distribution network load forecasting device based on multi-source information fusion. The information acquisition module 10 obtains the cultural environment information and power regulation information of the future set time period of the area to be predicted. The time division module 20 divides the future set time period into several time units according to the preset minimum time unit. The information determination module 30 determines the cultural environment information and power regulation information corresponding to each time unit. The load forecasting module 40 inputs the cultural environment information and power regulation information corresponding to each time unit into the prediction model created in advance based on historical load data, historical cultural environment information and historical power regulation information, and can obtain the predicted load data corresponding to each time unit. The data merging module 50 uses the predicted load data corresponding to each time unit to merge and obtain the predicted load data in the future time period of the area to be predicted, thereby realizing the prediction of the power grid load in the area to be predicted.

[0115] Optionally, the distribution network load forecasting system based on multi-source information fusion may also include:

[0116] The actual load acquisition module is used to obtain the actual load data corresponding to each preset time unit within the historical set time period of the area to be predicted;

[0117] The prediction model judgment module is used to determine whether the prediction model needs to be updated based on the actual load data and the predicted load data corresponding to each preset time unit within the historical set time period of the area to be predicted;

[0118] The model updating module is used to update the parameters of the prediction model with the goal of making the predicted load data close to the actual load data if the prediction model needs to be updated.

[0119] Optionally, the prediction model judgment module may include:

[0120] The actual load calculation module is used to add the actual load data corresponding to each preset time unit within the historical set time period of the area to be predicted to obtain the sum of the actual load data;

[0121] A forecast load calculation module is used to add the forecast load data corresponding to each preset time unit within the historical set time period of the area to be forecasted to obtain the sum of the forecast load data;

[0122] A load difference calculation module, used for subtracting the sum of the actual load data from the sum of the predicted load data to obtain a load data difference;

[0123] The difference judgment model is used to judge whether the load data difference is within a preset range. If not, it is determined that the prediction model needs to be updated.

[0124] Optionally, the distribution network load forecasting system based on multi-source information fusion may also include:

[0125] A prediction model creation module, used to create the prediction model;

[0126] The process of creating a predictive model may include:

[0127] Obtain load data, human environment information and power control information corresponding to each preset time unit within the historical set time period;

[0128] The load data, human environment information and power control information corresponding to each preset time unit within the historical set time period are integrated to obtain integrated data;

[0129] A prediction model is established using the fused data.

[0130] Optionally, before obtaining the load data, cultural environment information and power regulation information corresponding to each preset time unit within the historical set time period, the model creation module may also include:

[0131] Obtain load data corresponding to each preset time unit within a historical set time period transmitted by multiple data sources;

[0132] The load data corresponding to each preset time unit within the historical set time period transmitted by multiple data sources are integrated and processed to obtain the processed load data corresponding to each preset time unit within the historical set time period.

[0133] Optionally, after obtaining the processed load data corresponding to each preset time unit within the historical set time period, the model creation module may further include:

[0134] The processed load data corresponding to each preset time unit in the historical set time period is transmitted back to the multiple data sources, so that each data source can correct the historical load data according to the processed load data corresponding to each preset time unit in the historical set time period.

[0135] Optionally, after acquiring the load data, cultural environment information and power regulation information corresponding to each preset time unit within the historical set time period, the model creation module may further include:

[0136] Determine whether the load data corresponding to each preset time unit within the historical set time period meets the preset range;

[0137] If there is load data corresponding to a preset time unit that does not conform to the preset interval, for each load data that does not conform to the preset interval, a corrected load data is obtained based on the load data corresponding to other preset time units within the historical set time period to replace the load data that does not conform to the preset interval.

[0138] Optionally, the process of the model creation module obtaining the corrected load data for each load data that does not conform to the preset interval according to the load data corresponding to other preset time units within the historical set time period may include:

[0139] For each load data that does not conform to the preset interval, searching, from the preset time units corresponding to the load data that conform to the preset interval, two preset time units that are closest in time to the preset time units corresponding to the load data that does not conform to the preset interval;

[0140] The load data corresponding to the two preset time units are added together to obtain an average value to obtain the corrected load data.

[0141] Optionally, the distribution network load forecasting system based on multi-source information fusion may also include:

[0142] The display module is used to display the predicted load data of the area to be predicted within a set time period in the future.

[0143] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0144] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined with each other, and the same or similar parts can be referenced to each other.

[0145] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A distribution network load forecasting method based on multi-source information fusion, characterized in that: include: Acquire humanistic environment information and power regulation information within a set time period in the future of the predicted area, wherein the humanistic environment information includes geographical information, meteorological information, and humanistic information that affects power use, and the power regulation information includes electricity prices and regulation information that affects power use; Divide the future set time period into several time units according to the preset minimum time unit; Determine the human environment information and power regulation information corresponding to each time unit; Inputting the human environment information and power regulation information corresponding to each time unit into a pre-created prediction model to obtain predicted load data corresponding to each time unit, wherein the prediction model is created based on historical load data, historical human environment information, and historical power regulation information; The forecast load data corresponding to each time unit is combined to obtain the forecast load data for the future set time period of the area to be forecasted; The process of creating the prediction model includes: Obtain load data, human environment information and power control information corresponding to each preset time unit within the historical set time period; The load data, human environment information and power control information corresponding to each preset time unit within the historical set time period are integrated to obtain integrated data; Establishing a prediction model using the fused data; Before obtaining the load data, cultural environment information and power control information corresponding to each preset time unit within the historical set time period, it also includes: Obtain load data corresponding to each preset time unit within a historical set time period transmitted by multiple data sources; Integrate and process the load data corresponding to each preset time unit within the historical set time period transmitted by multiple data sources to obtain processed load data corresponding to each preset time unit within the historical set time period; The processed load data corresponding to each preset time unit in the historical set time period is transmitted back to the multiple data sources, so that each data source can correct the historical load data according to the processed load data corresponding to each preset time unit in the historical set time period.

2. The method according to claim 1, characterized in that: Also includes: Obtain the actual load data corresponding to each preset time unit within the historical set time period of the area to be predicted; Determine whether the prediction model needs to be updated based on the actual load data and predicted load data corresponding to each preset time unit within the historical set time period of the area to be predicted; If the prediction model needs to be updated, the parameters of the prediction model are updated with the goal of making the predicted load data close to the actual load data.

3. The method according to claim 2, characterized in that The step of determining whether the prediction model needs to be updated based on the actual load data and the predicted load data corresponding to each preset time unit within the historical set time period of the area to be predicted includes: Adding the actual load data corresponding to each preset time unit in the historical set time period of the area to be predicted to obtain the sum of the actual load data; Add the forecast load data corresponding to each preset time unit in the historical set time period of the area to be forecasted to obtain the sum of the forecast load data; Subtracting the sum of the actual load data from the sum of the predicted load data to obtain a load data difference; Determining whether the load data difference is within a preset range; If not, it is determined that the prediction model needs to be updated.

4. The method according to claim 1, characterized in that: After obtaining the load data corresponding to each preset time unit within the historical set time period, it also includes: Determine whether the load data corresponding to each preset time unit within the historical set time period meets the preset range; If there is load data corresponding to a preset time unit that does not conform to the preset interval, for each load data that does not conform to the preset interval, a corrected load data is obtained based on the load data corresponding to other preset time units within the historical set time period to replace the load data that does not conform to the preset interval.

5. The method according to claim 4, characterized in that For each load data that does not meet the preset interval, the corrected load data is obtained according to the load data corresponding to other preset time units within the historical set time period, including: For each load data that does not conform to the preset interval, searching, from the preset time units corresponding to the load data that conform to the preset interval, two preset time units that are closest in time to the preset time units corresponding to the load data that does not conform to the preset interval; The load data corresponding to the two preset time units are added together to obtain an average value to obtain the corrected load data.

6. The method according to claim 1, characterized in that Also includes: Displays the forecast load data for the area to be forecasted within the set time period in the future.

7. A distribution network load forecasting system based on multi-source information fusion, characterized in that: include: An information acquisition module, used to acquire humanistic environment information and power regulation information within a set time period in the future of the predicted area, wherein the humanistic environment information includes geographical information, meteorological information, and humanistic information that affects power use, and the power regulation information includes electricity prices and regulation information that affects power use; The time division module is used to divide the future set time period into several time units according to the preset minimum time unit; An information determination module, used to determine the human environment information and power regulation information corresponding to each time unit; A load forecasting module, which is used to input the human environment information and power regulation information corresponding to each time unit into a pre-created forecasting model to obtain forecast load data corresponding to each time unit, wherein the forecasting model is created based on historical load data, historical human environment information and historical power regulation information; A data merging module is used to use the forecast load data corresponding to each time unit to merge and obtain the forecast load data for the future time period of the area to be forecasted; A prediction model creation module, used to create the prediction model; The process of creating the prediction model includes: Obtain load data, human environment information and power control information corresponding to each preset time unit within the historical set time period; The load data, human environment information and power control information corresponding to each preset time unit within the historical set time period are integrated to obtain integrated data; Establishing a prediction model using the fused data; Before obtaining the load data, cultural environment information and power control information corresponding to each preset time unit within the historical set time period, it also includes: Obtain load data corresponding to each preset time unit within a historical set time period transmitted by multiple data sources; Integrate and process the load data corresponding to each preset time unit within the historical set time period transmitted by multiple data sources to obtain processed load data corresponding to each preset time unit within the historical set time period; The processed load data corresponding to each preset time unit in the historical set time period is transmitted back to the multiple data sources, so that each data source can correct the historical load data according to the processed load data corresponding to each preset time unit in the historical set time period.

Images