Regional power prediction system and method based on artificial intelligence

Abstract

The invention relates to the technical field of power prediction, and discloses a regional power prediction system and method based on artificial intelligence, and the method comprises the steps: collecting historical power data in a target region, carrying out the outlier analysis of the historical power data, and constructing a power data prediction sequence; classifying the power data prediction sequence to obtain an operation data prediction sequence and an associated data prediction sequence; calculating a power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence; according to the historical meteorological information of the position of the target area, establishing historical meteorological change curves of different time periods, and determining a meteorological electric power influence coefficient of the target area; and compensating the electric power predicted value based on the meteorological electric power influence coefficient to obtain a target electric power predicted value of the target region, so that the regional electric power can be accurately predicted, the utilization efficiency of electric power resources is improved, electric power waste is avoided, and a scientific basis is provided for electric power resource management and allocation.

Description

Technical Field

[0001] The present invention relates to the technical field of power forecasting, and in particular to a regional power forecasting system and method based on artificial intelligence. Background Art

[0002] With the rapid development of society and the continuous advancement of science and technology, electricity, as one of the most important energy sources in modern society, plays a vital role in all fields. However, the contradiction between the rapid growth of electricity demand and the limited power resources has become increasingly prominent, and how to effectively predict and manage electricity demand has become an urgent problem to be solved.

[0003] Traditional power forecasting methods mainly rely on historical data and manual experience. This method has great limitations when dealing with large amounts of data and complex situations, and the forecasting accuracy and efficiency cannot meet the needs of modern power systems. In addition, due to the volatility and uncertainty of power demand, traditional forecasting methods often cannot accurately predict future power demand, resulting in waste of power resources and unstable power supply. Summary of the invention

[0004] The embodiments of the present invention provide a regional power forecasting system and method based on artificial intelligence, which are used to solve the technical problems in the prior art that power cannot be accurately predicted, power resource utilization efficiency cannot be improved, and power waste is caused.

[0005] In order to achieve the above object, the present invention provides a regional power forecasting method based on artificial intelligence, comprising:

[0006] Determine a target area, collect historical power data in the target area, perform outlier analysis on the historical power data, and construct a power data prediction sequence;

[0007] Classifying the power data prediction sequence to obtain an operation data prediction sequence and a correlation data prediction sequence;

[0008] Calculating the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence;

[0009] According to the historical meteorological information of the location of the target area, establish the historical meteorological change curves of different periods, and determine the meteorological power impact coefficient of the target area;

[0010] The power forecast value is compensated based on the meteorological power impact coefficient to obtain a target power forecast value for the target area, wherein a product value of the meteorological power impact coefficient and the power forecast value is used as the target power forecast value.

[0011] Furthermore, when collecting historical power data in the target area, performing outlier analysis on the historical power data, and constructing a power data prediction tree, it includes:

[0012] Convert all historical power data into coordinate points, and establish a historical power data coordinate system based on all coordinate points;

[0013] Connect all coordinate points in sequence to obtain the historical power data fitting curve;

[0014] Performing smooth mark calibration on the smooth curve on the historical power data fitting curve, and performing to-be-determined mark calibration on the over-fitting point or under-fitting point on the non-smooth curve;

[0015] Determine the overfitting slope at all overfitting points, and determine the underfitting slope at underfitting points;

[0016] Calculating a convertible factor according to all overfitting slopes and underfitting slopes, and judging whether the identification to be determined can be converted into a smooth identification according to the convertible factor;

[0017] The conversion factor is calculated according to the following formula:

[0018]

[0019] Where a is the convertible factor, n is the number of overfitting points, bi is the overfitting slope corresponding to the i-th overfitting point, c is the smoothing slope corresponding to the smooth curve, f1 is the calculation coefficient of the overfitting point, m is the number of underfitting points, ej is the underfitting slope corresponding to the j-th underfitting point, and f2 is the calculation coefficient of the underfitting point;

[0020] When the convertible factor is less than the preset convertible factor, it is determined that the identification to be determined can be converted into a smooth identification;

[0021] When the convertible factor is greater than or equal to the preset convertible factor, it is determined that the identification to be determined cannot be converted into a smooth identification;

[0022] The power data prediction sequence is constructed based on the historical power data corresponding to all smoothing identifiers.

[0023] Furthermore, when all historical power data are converted into coordinate points and a historical power data coordinate system is established based on all the coordinate points, it includes:

[0024] Outputting the data entropy value of each historical power data based on the pre-trained data entropy value model, and using the data entropy value as the X-axis coordinate value of the historical power data;

[0025] Obtaining historical usage behavior of the historical power data and constructing a corresponding historical usage behavior value set;

[0026] Determining a standard value set corresponding to the historical usage behavior value set;

[0027] Comparing the historical usage behavior value set with the standard value set, filtering out all historical usage behavior values ​​greater than the standard value, and obtaining an excess value set;

[0028] Filter out all historical usage behavior values ​​that are less than the standard value to obtain a missing value set;

[0029] Filter out all historical usage behavior values ​​that are equal to the standard values ​​to obtain a reasonable value set;

[0030] Calculate the comprehensive data dimension of the historical power data according to the excess value set, the deficiency value set and the reasonable value set;

[0031] Using the comprehensive data dimension as the Y-axis coordinate value of the historical power data;

[0032] A historical power data coordinate system is established according to the X-axis coordinate value and the Y-axis coordinate value.

[0033] Furthermore, the comprehensive data dimension of the historical power data is calculated according to the following formula:

[0034]

[0035] Wherein, G is the comprehensive data dimension of historical power data, n1 is the number of historical usage behavior values ​​in the excess value set, pk is the kth historical usage behavior value in the excess value set, pmin is the smallest historical usage behavior value in the excess value set, pmax is the largest historical usage behavior value in the excess value set, m1 is the number of historical usage behavior values ​​in the missing value set, rq is the qth historical usage behavior value in the missing value set, rmin is the smallest historical usage behavior value in the missing value set, rmax is the largest historical usage behavior value in the missing value set, and h is the comprehensive data dimension correction coefficient;

[0036] Determine the comprehensive data dimension correction coefficient h according to the following method;

[0037] Counting the number of historical usage behavior values ​​in the reasonable value set;

[0038] Calculating the total number of the excess value set and the deficiency value set;

[0039] When the number is less than the total number, the comprehensive data dimension correction coefficient h is set to y1;

[0040] When the number is equal to the total number, the comprehensive data dimension correction coefficient h is set to y2;

[0041] When the number is greater than the total number, the comprehensive data dimension correction coefficient h is set to y3.

[0042] Further, when calculating the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence, it includes:

[0043] Performing numerical mapping on the operation data prediction sequence to obtain an operation data prediction numerical value set;

[0044] Performing numerical mapping on the associated data prediction sequence to obtain an associated data prediction numerical value set;

[0045] The power prediction value of the target area is calculated according to the operation data prediction value set and the associated data prediction value set.

[0046] Furthermore, the power prediction value of the target area is calculated according to the following formula:

[0047]

[0048] Among them, S is the power prediction value of the target area, m1 is the number of operating data prediction values ​​in the operating data prediction value set, Tmax is the maximum value in the operating data prediction value set, Tx is the x-th operating data prediction value in the operating data prediction value set, Ux is the weight corresponding to the x-th operating data prediction value in the operating data prediction value set, m2 is the number of associated data prediction values ​​in the associated data prediction value set, Vmax is the maximum value in the associated data prediction value set, Vz is the value of the z-th associated data prediction value in the associated data prediction value set, Ly is the weight corresponding to the z-th associated data prediction value in the associated data prediction value set, and α is the integration coefficient for calculating the power prediction value.

[0049] Furthermore, when establishing historical meteorological change curves of different time periods according to the historical meteorological information of the location of the target area, and determining the meteorological power impact coefficient of the target area, it includes:

[0050] Determine the division time and divide the historical meteorological change curve into time divisions;

[0051] Analyze the reference points of the historical meteorological change curve corresponding to the same time, and determine the percentage difference in power between the remaining points and the reference points;

[0052] Based on the corresponding historical meteorological change curve, determine the meteorological difference and meteorological type corresponding to the power difference percentage, and use the power difference percentage and the corresponding meteorological difference as the first data;

[0053] Determine, based on all data corresponding to the meteorological type, a first relationship between the percentage of power difference and the change in meteorological difference in a corresponding time period;

[0054] The meteorological power influence coefficient of the target area is determined based on the first relationship and the first data.

[0055] In order to achieve the above object, the present invention also provides a regional power forecasting system based on artificial intelligence, which is characterized by comprising:

[0056] The first module is used to determine the target area, collect historical power data in the target area, perform outlier analysis on the historical power data, and construct a power data prediction sequence;

[0057] The second module is used to classify the power data prediction sequence to obtain an operation data prediction sequence and a correlation data prediction sequence;

[0058] A third module is used to calculate the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence;

[0059] The fourth module is used to establish historical meteorological change curves of different time periods according to the historical meteorological information of the location of the target area, and determine the meteorological power impact coefficient of the target area;

[0060] The fifth module is used to compensate the power forecast value based on the meteorological power impact coefficient to obtain the target power forecast value of the target area, wherein the product value of the meteorological power impact coefficient and the power forecast value is used as the target power forecast value.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention discloses a regional power forecasting system and method based on artificial intelligence, the method comprising: collecting historical power data in a target area, performing outlier analysis on the historical power data, and constructing a power data forecasting sequence; classifying the power data forecasting sequence to obtain an operation data forecasting sequence and a correlation data forecasting sequence; calculating a power forecasting value of the target area according to the operation data forecasting sequence and the correlation data forecasting sequence; establishing a historical meteorological change curve for different time periods according to the historical meteorological information of the location of the target area, and determining the meteorological power influence coefficient of the target area; compensating the power forecasting value based on the meteorological power influence coefficient to obtain a target power forecasting value of the target area. The present invention can accurately forecast regional power, improve the utilization efficiency of power resources, avoid power waste, and provide a scientific basis for power resource management and allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0064] Figure 1 A schematic diagram of a flow chart of a regional power forecasting method based on artificial intelligence in an embodiment of the present invention is shown;

[0065] Figure 2 A structural schematic diagram of a regional power forecasting system based on artificial intelligence in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0066] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0067] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0068] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0069] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0070] The following is a description of preferred embodiments of the present invention with reference to the accompanying drawings.

[0071] like Figure 1 As shown, an embodiment of the present invention discloses a regional power forecasting method based on artificial intelligence, comprising:

[0072] S110: determining a target area, collecting historical power data in the target area, performing outlier analysis on the historical power data, and constructing a power data prediction sequence;

[0073] In some embodiments of the present application, when collecting historical power data in the target area, performing outlier analysis on the historical power data, and constructing a power data prediction tree, the process includes:

[0074] Convert all historical power data into coordinate points, and establish a historical power data coordinate system based on all coordinate points;

[0075] Connect all coordinate points in sequence to obtain the historical power data fitting curve;

[0076] Performing smooth mark calibration on the smooth curve on the historical power data fitting curve, and performing to-be-determined mark calibration on the over-fitting point or under-fitting point on the non-smooth curve;

[0077] Determine the overfitting slope at all overfitting points, and determine the underfitting slope at underfitting points;

[0078] Calculating a convertible factor according to all overfitting slopes and underfitting slopes, and judging whether the identification to be determined can be converted into a smooth identification according to the convertible factor;

[0079] The conversion factor is calculated according to the following formula:

[0080]

[0081] Where a is the convertible factor, n is the number of overfitting points, bi is the overfitting slope corresponding to the i-th overfitting point, c is the smoothing slope corresponding to the smooth curve, f1 is the calculation coefficient of the overfitting point, m is the number of underfitting points, ej is the underfitting slope corresponding to the j-th underfitting point, and f2 is the calculation coefficient of the underfitting point;

[0082] When the convertible factor is less than the preset convertible factor, it is determined that the identification to be determined can be converted into a smooth identification;

[0083] When the convertible factor is greater than or equal to the preset convertible factor, it is determined that the identification to be determined cannot be converted into a smooth identification;

[0084] The power data prediction sequence is constructed based on the historical power data corresponding to all smoothing identifiers.

[0085] In this embodiment, historical power data includes power grid operation data, equipment monitoring data, power enterprise marketing data, enterprise management data, etc. These data come from various links such as power generation, transmission, transformation, distribution and power consumption.

[0086] In this embodiment, the coordinate system of historical power data is a rectangular coordinate system, that is, an XY axis coordinate system;

[0087] In this embodiment, some data points are located above the smooth curve, which are over-fitting points, and some data points are located below the smooth curve, which are under-fitting points.

[0088] In this embodiment, the preset convertibility factor is 0.8.

[0089] The beneficial effect of the above technical solution is that the present invention calculates the convertible factor, obtains the smoothing mark according to the relationship between the convertible factor and the preset convertible factor, and finally constructs the power data prediction sequence, which can lay the foundation for the power prediction of the target area and provide reliable data support.

[0090] In some embodiments of the present application, when all historical power data are converted into coordinate points and a historical power data coordinate system is established according to all the coordinate points, it includes:

[0091] Outputting the data entropy value of each historical power data based on the pre-trained data entropy value model, and using the data entropy value as the X-axis coordinate value of the historical power data;

[0092] Obtaining historical usage behavior of the historical power data and constructing a corresponding historical usage behavior value set;

[0093] Determining a standard value set corresponding to the historical usage behavior value set;

[0094] Comparing the historical usage behavior value set with the standard value set, filtering out all historical usage behavior values ​​greater than the standard value, and obtaining an excess value set;

[0095] Filter out all historical usage behavior values ​​that are less than the standard value to obtain a missing value set;

[0096] Filter out all historical usage behavior values ​​that are equal to the standard values ​​to obtain a reasonable value set;

[0097] Calculate the comprehensive data dimension of the historical power data according to the excess value set, the deficiency value set and the reasonable value set;

[0098] Using the comprehensive data dimension as the Y-axis coordinate value of the historical power data;

[0099] A historical power data coordinate system is established according to the X-axis coordinate value and the Y-axis coordinate value.

[0100] In this embodiment, the data entropy value model is pre-trained and can output the data entropy value corresponding to each historical power data.

[0101] In this embodiment, the coordinates corresponding to the historical power data are (X, Y).

[0102] In this embodiment, the historical usage behavior includes historical click counts, historical extraction counts, historical reference counts, historical browsing counts, etc.

[0103] The beneficial effect of the above technical solution is that the present invention establishes a historical power data coordinate system according to the X-axis coordinate values ​​and the Y-axis coordinate values, which can provide a basis for determining the removal of outliers, realize data visualization, and improve the overall efficiency and effect of power forecasting.

[0104] In some embodiments of the present application, the comprehensive data dimension of the historical power data is calculated according to the following formula:

[0105]

[0106] Wherein, G is the comprehensive data dimension of historical power data, n1 is the number of historical usage behavior values ​​in the excess value set, pk is the kth historical usage behavior value in the excess value set, pmin is the smallest historical usage behavior value in the excess value set, pmax is the largest historical usage behavior value in the excess value set, m1 is the number of historical usage behavior values ​​in the missing value set, rq is the qth historical usage behavior value in the missing value set, rmin is the smallest historical usage behavior value in the missing value set, rmax is the largest historical usage behavior value in the missing value set, and h is the comprehensive data dimension correction coefficient;

[0107] Determine the comprehensive data dimension correction coefficient h according to the following method;

[0108] Counting the number of historical usage behavior values ​​in the reasonable value set;

[0109] Calculating the total number of the excess value set and the deficiency value set;

[0110] When the number is less than the total number, the comprehensive data dimension correction coefficient h is set to y1;

[0111] When the number is equal to the total number, the comprehensive data dimension correction coefficient h is set to y2;

[0112] When the number is greater than the total number, the comprehensive data dimension correction coefficient h is set to y3.

[0113] In this embodiment, y1=0.8, y2=1, y3=1.2.

[0114] The beneficial effect of the above technical solution is that the present invention realizes the dynamic setting of the comprehensive data dimension and ensures the calculation accuracy.

[0115] S120: Classifying the power data prediction sequence to obtain an operation data prediction sequence and a correlation data prediction sequence;

[0116] S130: Calculating the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence;

[0117] In some embodiments of the present application, when calculating the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence, it includes:

[0118] Performing numerical mapping on the operation data prediction sequence to obtain an operation data prediction numerical value set;

[0119] Performing numerical mapping on the associated data prediction sequence to obtain an associated data prediction numerical value set;

[0120] The power prediction value of the target area is calculated according to the operation data prediction value set and the associated data prediction value set.

[0121] In this embodiment, the operation data prediction sequence involves data that directly reflects the real-time operation status of the power system, such as load demand, power generation, grid frequency, voltage level, etc. These data are usually collected by sensors, smart meters or SCADA systems. The associated data prediction sequence includes those factors that indirectly affect the demand and supply of electricity. This includes but is not limited to environmental factors (such as temperature, humidity, wind speed), time factors (such as seasons, holidays, time of day) and economic indicators (such as industrial growth rate, resident consumption level).

[0122] The beneficial effect of the above technical solution is that the present invention realizes accurate prediction of power demand in the target area, which not only optimizes the allocation of power resources, but also ensures the efficient and stable operation of the power grid, avoids errors caused by human participation, and improves prediction accuracy and efficiency.

[0123] In some embodiments of the present application, the power prediction value of the target area is calculated according to the following formula:

[0124]

[0125] Among them, S is the power prediction value of the target area, m1 is the number of operating data prediction values ​​in the operating data prediction value set, Tmax is the maximum value in the operating data prediction value set, Tx is the x-th operating data prediction value in the operating data prediction value set, Ux is the weight corresponding to the x-th operating data prediction value in the operating data prediction value set, m2 is the number of associated data prediction values ​​in the associated data prediction value set, Vmax is the maximum value in the associated data prediction value set, Vz is the value of the z-th associated data prediction value in the associated data prediction value set, Ly is the weight corresponding to the z-th associated data prediction value in the associated data prediction value set, and α is the integration coefficient for calculating the power prediction value.

[0126] S140: Establishing historical meteorological change curves of different time periods according to historical meteorological information of the target area, and determining the meteorological power impact coefficient of the target area;

[0127] In some embodiments of the present application, when establishing historical meteorological change curves of different time periods according to the historical meteorological information of the location of the target area, and determining the meteorological power impact coefficient of the target area, it includes:

[0128] Determine the division time and divide the historical meteorological change curve into time divisions;

[0129] Analyze the reference points of the historical meteorological change curve corresponding to the same time, and determine the percentage difference in power between the remaining points and the reference points;

[0130] Based on the corresponding historical meteorological change curve, determine the meteorological difference and meteorological type corresponding to the power difference percentage, and use the power difference percentage and the corresponding meteorological difference as the first data;

[0131] Determine, based on all data corresponding to the meteorological type, a first relationship between the percentage of power difference and the change in meteorological difference in a corresponding time period;

[0132] The meteorological power influence coefficient of the target area is determined based on the first relationship and the first data.

[0133] In this embodiment, a suitable time scale is determined to analyze historical meteorological data, including seasons, months, days, hours and other time periods to divide the meteorological change curve, so as to capture the meteorological characteristics and change trends under different time scales.

[0134] In this embodiment, in each weather change curve at the same time, a reference point is analyzed and selected. This reference point can be the lowest temperature moment of the day, the average temperature or other key weather indicators. Then, the power consumption difference between other points on the same time curve and the reference point is calculated, and the percentage of these differences in the power consumption of the reference point is calculated.

[0135] In this embodiment, the meteorological differences and types corresponding to the power differences are determined: based on the historical meteorological change curve, the relationship between the power difference percentage and the meteorological conditions (such as temperature, humidity, wind speed, etc.) is identified. These meteorological differences and their corresponding power difference percentages are used as the first data.

[0136] In this embodiment, the above first data is used to establish a mathematical model or relationship between the percentage change of power difference and the change of meteorological difference, which usually involves statistical analysis methods, such as regression analysis, to determine the change pattern of power demand under different meteorological conditions.

[0137] In this embodiment, a further analysis is performed based on the first relationship and the first data to determine the meteorological power impact coefficient of the specific meteorological type on the power demand.

[0138] The beneficial effect of the above technical solution is: the present invention determines the meteorological power impact coefficient, thereby realizing dynamic adjustment of the power forecast value, avoiding inaccurate prediction, improving the efficiency and reliability of the power system operation, and ensuring the stability of power supply.

[0139] S150: Compensating the power forecast value based on the meteorological power impact coefficient to obtain a target power forecast value for the target area, wherein the product of the meteorological power impact coefficient and the power forecast value is used as the target power forecast value.

[0140] In order to further explain the technical idea of ​​the present invention, the technical solution of the present invention is now described in combination with specific application scenarios.

[0141] Correspondingly, such as Figure 2 As shown, the present application also provides a regional power forecasting system based on artificial intelligence, including:

[0142] The first module is used to determine the target area, collect historical power data in the target area, perform outlier analysis on the historical power data, and construct a power data prediction sequence;

[0143] The second module is used to classify the power data prediction sequence to obtain an operation data prediction sequence and a correlation data prediction sequence;

[0144] A third module is used to calculate the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence;

[0145] The fourth module is used to establish historical meteorological change curves of different time periods according to the historical meteorological information of the location of the target area, and determine the meteorological power impact coefficient of the target area;

[0146] The fifth module is used to compensate the power forecast value based on the meteorological power impact coefficient to obtain the target power forecast value of the target area, wherein the product value of the meteorological power impact coefficient and the power forecast value is used as the target power forecast value.

[0147] In the description of the above embodiments, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0148] Although the present invention has been described above with reference to the embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed by the present invention may be used in combination with each other in any manner, and the fact that these combinations are not fully described in this specification is only for the sake of omitting space and saving resources.

[0149] Those skilled in the art can understand that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A regional power forecasting method based on artificial intelligence, characterized in that: include: Determine a target area, collect historical power data in the target area, perform outlier analysis on the historical power data, and construct a power data prediction sequence; Classifying the power data prediction sequence to obtain an operation data prediction sequence and a correlation data prediction sequence; Calculating the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence; According to the historical meteorological information of the location of the target area, establish the historical meteorological change curves of different periods, and determine the meteorological power impact coefficient of the target area; The power forecast value is compensated based on the meteorological power impact coefficient to obtain a target power forecast value for the target area, wherein a product value of the meteorological power impact coefficient and the power forecast value is used as the target power forecast value.

2. The regional power forecasting method based on artificial intelligence according to claim 1 is characterized in that: When collecting historical power data in the target area, performing outlier analysis on the historical power data, and constructing a power data prediction tree, the method includes: Convert all historical power data into coordinate points, and establish a historical power data coordinate system based on all coordinate points; Connect all coordinate points in sequence to obtain the historical power data fitting curve; Performing smooth mark calibration on the smooth curve on the historical power data fitting curve, and performing to-be-determined mark calibration on the over-fitting point or under-fitting point on the non-smooth curve; Determine the overfitting slope at all overfitting points, and determine the underfitting slope at underfitting points; Calculating a convertible factor according to all overfitting slopes and underfitting slopes, and judging whether the identification to be determined can be converted into a smooth identification according to the convertible factor; The conversion factor is calculated according to the following formula: Where a is the convertible factor, n is the number of overfitting points, bi is the overfitting slope corresponding to the i-th overfitting point, c is the smoothing slope corresponding to the smooth curve, f1 is the calculation coefficient of the overfitting point, m is the number of underfitting points, ej is the underfitting slope corresponding to the j-th underfitting point, and f2 is the calculation coefficient of the underfitting point; When the convertible factor is less than the preset convertible factor, it is determined that the identification to be determined can be converted into a smooth identification; When the convertible factor is greater than or equal to the preset convertible factor, it is determined that the identification to be determined cannot be converted into a smooth identification; The power data prediction sequence is constructed based on the historical power data corresponding to all smoothing identifiers.

3. The regional power forecasting method based on artificial intelligence according to claim 2 is characterized in that: When all historical power data are converted into coordinate points and a historical power data coordinate system is established based on all coordinate points, it includes: Outputting the data entropy value of each historical power data based on the pre-trained data entropy value model, and using the data entropy value as the X-axis coordinate value of the historical power data; Obtaining historical usage behavior of the historical power data and constructing a corresponding historical usage behavior value set; Determining a standard value set corresponding to the historical usage behavior value set; Comparing the historical usage behavior value set with the standard value set, filtering out all historical usage behavior values ​​greater than the standard value, and obtaining an excess value set; Filter out all historical usage behavior values ​​that are less than the standard value to obtain a missing value set; Filter out all historical usage behavior values ​​that are equal to the standard values ​​to obtain a reasonable value set; Calculate the comprehensive data dimension of the historical power data according to the excess value set, the deficiency value set and the reasonable value set; Using the comprehensive data dimension as the Y-axis coordinate value of the historical power data; A historical power data coordinate system is established according to the X-axis coordinate value and the Y-axis coordinate value.

4. The regional power forecasting method based on artificial intelligence according to claim 3 is characterized in that: The comprehensive data dimension of the historical power data is calculated according to the following formula: Wherein, G is the comprehensive data dimension of historical power data, n1 is the number of historical usage behavior values ​​in the excess value set, pk is the kth historical usage behavior value in the excess value set, pmin is the smallest historical usage behavior value in the excess value set, pmax is the largest historical usage behavior value in the excess value set, m1 is the number of historical usage behavior values ​​in the missing value set, rq is the qth historical usage behavior value in the missing value set, rmin is the smallest historical usage behavior value in the missing value set, rmax is the largest historical usage behavior value in the missing value set, and h is the comprehensive data dimension correction coefficient; Determine the comprehensive data dimension correction coefficient h according to the following method; Counting the number of historical usage behavior values ​​in the reasonable value set; Calculating the total number of the excess value set and the deficiency value set; When the number is less than the total number, the comprehensive data dimension correction coefficient h is set to y1; When the number is equal to the total number, the comprehensive data dimension correction coefficient h is set to y2; When the number is greater than the total number, the comprehensive data dimension correction coefficient h is set to y3.

5. The regional power forecasting method based on artificial intelligence according to claim 1 is characterized in that: When calculating the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence, it includes: Performing numerical mapping on the operation data prediction sequence to obtain an operation data prediction numerical value set; Performing numerical mapping on the associated data prediction sequence to obtain an associated data prediction numerical value set; The power prediction value of the target area is calculated according to the operation data prediction value set and the associated data prediction value set.

6. The regional power forecasting method based on artificial intelligence according to claim 5 is characterized in that: The power forecast value of the target area is calculated according to the following formula: Among them, S is the power prediction value of the target area, m1 is the number of operating data prediction values ​​in the operating data prediction value set, Tmax is the maximum value in the operating data prediction value set, Tx is the x-th operating data prediction value in the operating data prediction value set, Ux is the weight corresponding to the x-th operating data prediction value in the operating data prediction value set, m2 is the number of associated data prediction values ​​in the associated data prediction value set, Vmax is the maximum value in the associated data prediction value set, Vz is the value of the z-th associated data prediction value in the associated data prediction value set, Ly is the weight corresponding to the z-th associated data prediction value in the associated data prediction value set, and α is the integration coefficient for calculating the power prediction value.

7. The regional power forecasting method based on artificial intelligence according to claim 1 is characterized in that: When establishing historical meteorological change curves of different time periods according to the historical meteorological information of the location of the target area, and determining the meteorological power impact coefficient of the target area, it includes: Determine the division time and divide the historical meteorological change curve into time divisions; Analyze the reference points of the historical meteorological change curve corresponding to the same time, and determine the percentage difference in power between the remaining points and the reference points; Based on the corresponding historical meteorological change curve, determine the meteorological difference and meteorological type corresponding to the power difference percentage, and use the power difference percentage and the corresponding meteorological difference as the first data; Determine, based on all data corresponding to the meteorological type, a first relationship between the percentage of power difference and the change in meteorological difference in a corresponding time period; The meteorological power influence coefficient of the target area is determined based on the first relationship and the first data.

8. A regional power forecasting system based on artificial intelligence, characterized in that: include: The first module is used to determine the target area, collect historical power data in the target area, perform outlier analysis on the historical power data, and construct a power data prediction sequence; The second module is used to classify the power data prediction sequence to obtain an operation data prediction sequence and a correlation data prediction sequence; A third module is used to calculate the power prediction value of the target area according to the operation data prediction sequence and the associated data prediction sequence; The fourth module is used to establish historical meteorological change curves of different time periods according to the historical meteorological information of the location of the target area, and determine the meteorological power impact coefficient of the target area; The fifth module is used to compensate the power forecast value based on the meteorological power impact coefficient to obtain the target power forecast value of the target area, wherein the product value of the meteorological power impact coefficient and the power forecast value is used as the target power forecast value.

Images