An orderly charging scheduling method based on charging load prediction

Abstract

The present invention relates to the field of data processing technology, and specifically to an orderly charging scheduling method based on charging load prediction, including: obtaining charging pile target load data and charging pile historical load data sequence; obtaining the degree of correlation between the charging pile target load data and each charging pile historical load data in the charging pile historical load data sequence; obtaining a number of historical neighborhood data sequences and target neighborhood data sequences, and obtaining the degree of correlation confidence between the target neighborhood data sequence and each historical neighborhood data sequence according to the degree of correlation between each charging pile historical load data and the charging pile target load data in the charging pile historical load data sequence; adjusting the model prediction error related parameters according to the correlation confidence of the best correlation degree difference sequence, and realizing the orderly charging scheduling of the charging pile. The present invention realizes the weakening of the influence of the model on the non-stationary factors in the charging pile load data, obtains accurate charging load prediction, and realizes the orderly charging scheduling of the charging pile.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an orderly charging scheduling method based on charging load prediction. Background Art

[0002] As an important part of energy supply, charging piles can be scheduled for orderly charging at appropriate time periods by predicting charging loads. Charging demand can be reasonably allocated during peak load periods of the power grid, reducing the load pressure on the power grid and avoiding charging speed reduction or charging pile congestion caused by overload during peak periods, thus improving charging efficiency. Therefore, the accuracy of charging load prediction is the basis for orderly charging scheduling, and the use of The model can perform effective time series forecasting, but it is easily affected by the unstable charging load in the short term, which reduces the accuracy of the forecast results.

[0003] because The model is sensitive to the non-stationary influence of charging pile load data, which often leads to inaccurate acquisition of lag order and moving average lag order, thus affecting The accuracy of the model in predicting data causes a large error between the actual data and the predicted data, making subsequent orderly charging scheduling difficult to achieve. Summary of the invention

[0004] In order to solve the above problems, the present invention provides an orderly charging scheduling method based on charging load prediction, the method comprising:

[0005] Obtain charging pile target load data and charging pile historical load data sequence;

[0006] Obtaining prediction data of each charging pile historical load data in the charging pile historical load data sequence; obtaining the degree of correlation between each charging pile historical load data in the charging pile historical load data sequence and the charging pile target load data according to the prediction data of each charging pile historical load data in the charging pile historical load data sequence;

[0007] Obtain several historical neighborhood data sequences and target neighborhood data sequences; obtain a correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence according to the correlation degree between each charging pile historical load data and the charging pile target load data in the charging pile historical load data sequence; obtain the correlation confidence degree of the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence;

[0008] The best correlation degree difference sequence is obtained according to the correlation confidence degree of the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence; the adjusted lag order of the prediction error related parameters and the adjusted moving average lag order of the prediction error related parameters are obtained according to the correlation confidence degree of the best correlation degree difference sequence; the charging piles are orderly charged and scheduled according to the adjusted lag order of the prediction error related parameters and the adjusted moving average lag order of the prediction error related parameters.

[0009] Preferably, the specific method of obtaining the charging pile target load data and the charging pile historical load data sequence includes:

[0010] The load data of the charging pile on the current day and the load data of the charging pile on the past several days are collected through electrical sensors; for the charging pile load data of each day, the empty part of the load data is removed, and then the other discrete load data are regularized to obtain the regularized charging pile load data for each day; similarly, the regularized load data of the charging pile on the current day and the regularized load data of the charging pile on the past several days are obtained; the regularized load data of the charging pile on the current day is recorded as the charging pile target load data; the charging pile load data sequence composed of the regularized load data of the charging pile on the past several days is recorded as the charging pile historical load data sequence.

[0011] Preferably, the specific method of obtaining the prediction data of each charging pile historical load data in the charging pile historical load data sequence includes:

[0012] pass The model predicts the historical load data of each charging pile in the historical load data sequence of the charging pile to obtain the predicted data of the historical load data of each charging pile in the historical load data sequence of the charging pile.

[0013] Preferably, the method of obtaining the correlation degree between each charging pile historical load data in the charging pile historical load data sequence and the charging pile target load data according to the predicted data of each charging pile historical load data in the charging pile historical load data sequence includes the following specific methods:

[0014] Obtain the time and date interval length between each charging pile historical load data and the charging pile target load data in the charging pile historical load data sequence; then The calculation expression of the correlation degree between the historical load data of a charging pile and the target load data of the charging pile is:

[0015]

[0016] In the formula, Indicates the number of the charging pile historical load data sequence The degree of correlation between the historical load data of each charging pile and the target load data of the charging pile; Indicates the number of the charging pile historical load data sequence Prediction data of historical load data of charging piles; Indicates the number of the charging pile historical load data sequence Prediction data of historical load data of charging piles; Indicates the target load data of the charging pile; Indicates the number of the charging pile historical load data sequence The length of the time interval between the historical load data of each charging pile and the target load data of the charging pile; represents the DTW distance; Indicates taking the absolute value; represents the linear normalization function.

[0017] Preferably, the method of obtaining a plurality of historical neighborhood data sequences and a target neighborhood data sequence includes:

[0018] Preset Parameters , record the charging pile historical load data adjacent to the charging pile target load data in the charging pile historical load data sequence as the first historical load data; The historical load data sequence consisting of the historical load data of the charging piles and the first historical load data is recorded as the target neighborhood data sequence; in the charging pile historical load data sequence, any charging pile historical load data except all the charging pile historical load data of the target neighborhood data sequence is recorded as the second historical load data, and the second historical load data is recorded as the left adjacent to the charging pile historical load data sequence. The historical load data sequence consisting of the historical load data of the charging piles and the second historical load data is recorded as a historical neighborhood data sequence. Similarly, several historical neighborhood data sequences are obtained.

[0019] Preferably, the method of obtaining the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence includes the following specific methods:

[0020] For the historical neighborhood data sequence and the target neighborhood data sequence, obtain the degree of correlation between the historical load data of each charging pile and the target load data of the charging pile in each historical neighborhood data sequence, and obtain the degree of correlation between the historical load data of each charging pile and the target load data of the charging pile in the target neighborhood data sequence; for the A historical neighborhood data sequence, The first The degree of correlation between the historical load data of a charging pile and the target load data of the charging pile and the target neighborhood data sequence The degree of correlation between the historical load data of a charging pile and the target load data of the charging pile The difference between The first in the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence elements, and similarly, get the All elements in the difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence are obtained, and then the first The difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence.

[0021] Preferably, the specific method of obtaining the association confidence level of the association degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence includes:

[0022] No. The confidence calculation expression of the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence is:

[0023]

[0024] In the formula, Indicates The confidence level of the difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence; Indicates The first in the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence elements; Represents the first in the sequence of differences in the degree of association between all historical neighborhood data sequences and the target neighborhood data sequence The mean of the elements; Indicates taking the absolute value; are preset parameters.

[0025] Preferably, the method of obtaining the best correlation degree difference sequence according to the correlation confidence level of the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence includes the following specific methods:

[0026] The association confidence levels of the association degree difference sequences between all historical neighborhood data sequences and target neighborhood data sequences are combined into an association confidence level sequence, and the maximum association confidence level of the association degree difference sequence between the historical neighborhood data sequences and the target neighborhood data sequences in the association confidence level sequence is recorded as the optimal association degree difference sequence.

[0027] Preferably, the method of obtaining the adjusted lag order of the prediction error related parameter and the adjusted moving average lag order of the prediction error related parameter according to the correlation confidence level of the best correlation level difference sequence includes the following specific methods:

[0028] The historical neighborhood data sequence corresponding to the best correlation degree difference sequence is recorded as the best historical neighborhood data sequence. The historical load data of charging piles is shown in the autocorrelation diagram. and partial autocorrelation plot Get the Historical load data of charging piles The lag order of the model With the moving average lag order ; Through the autocorrelation plot And partial autocorrelation plot Obtaining charging pile target load data The lag order of the model With the moving average lag order ; The adjusted forecast error related parameter lag order With the moving average lag order The calculation expression is:

[0029]

[0030] In the formula, Indicates the correlation confidence level of the best correlation difference sequence; are preset parameters.

[0031] Preferably, the orderly charging scheduling of the charging piles according to the adjusted hysteresis order of the prediction error related parameters and the adjusted moving average hysteresis order of the prediction error related parameters includes the following specific methods:

[0032] The adjusted lag order of the forecast error related parameters and the adjusted moving average lag order of the forecast error related parameters are taken as The parameters of the model are The model predicts the charging load data of the charging piles, and the charging piles are scheduled for charging in an orderly manner according to the predicted charging load data of the charging piles.

[0033] The beneficial effects of the technical solution of the present invention are: The model is sensitive to the non-stationary influence of charging pile load data, which often leads to inaccurate acquisition of lag order and moving average lag order, thus affecting The accuracy of the model for predicting data causes a large error between the actual data and the predicted data, making it difficult to implement subsequent orderly charging scheduling; the present invention is based on The model processes the segmented load data of the charging pile load data to quantify the correlation degree of the segmented load data, and adjusts the values ​​of the lag order and the moving average lag order based on the similarity of the historical load data, thereby reducing the prediction error and achieving The model weakens the impact of non-stationary factors in the charging pile load data, obtains accurate charging load prediction, and thus realizes orderly charging scheduling of charging piles. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 The present invention is a flowchart of the steps of an orderly charging scheduling method based on charging load prediction. DETAILED DESCRIPTION

[0036] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of an orderly charging scheduling method based on charging load prediction proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0038] The following is a detailed description of a specific scheme of an orderly charging scheduling method based on charging load prediction provided by the present invention in conjunction with the accompanying drawings.

[0039] See also Figure 1 , which shows a flowchart of a method for orderly charging scheduling based on charging load prediction provided by an embodiment of the present invention, the method comprising the following steps:

[0040] Step S001: Obtain charging pile target load data and charging pile historical load data sequence.

[0041] It should be noted that since the charging pile load data of each day corresponds to a certain period, there is no charging load data when the charging pile is in standby state or stops working. Therefore, the charging pile load data of each day is obtained as a data sequence of multiple intervals. Therefore, it is necessary to remove and sort the load data of the charging pile load data of each day and convert the discrete load data into continuous load data in form.

[0042] Specifically, in order to implement the orderly charging scheduling method based on charging load prediction proposed in this embodiment, it is first necessary to collect charging pile target load data and charging pile historical load data sequence. The specific process is as follows:

[0043] The current load data of the charging pile and the load data of the past few days are collected through electrical sensors. The load data of the charging pile for each day includes the timestamp of the charging load of the charging pile. , load value 、Charging pile number For each day's charging pile load data, remove the empty part of the load data, and then regularize the other discrete load data to obtain the regularized charging pile load data for each day; similarly, obtain the regularized charging pile load data for the current day and the regularized charging pile load data for the past few days.

[0044] Among them, data regularization is a prior art and will not be described in detail in this embodiment.

[0045] The regularized load data of the charging pile on the current day is recorded as the charging pile target load data; the charging pile load data sequence composed of the regularized load data of the charging pile in the past several days is recorded as the charging pile historical load data sequence.

[0046] So far, the charging pile target load data and the charging pile historical load data sequence are obtained through the above method.

[0047] Step S002: Obtain the correlation between the charging pile target load data and each charging pile historical load data in the charging pile historical load data sequence.

[0048] It should be noted that in order to adjust The values ​​of the lag order and the moving average lag order of the model can minimize the prediction error of the prediction model. It is necessary to analyze the changing trend characteristics of the charging pile load data. The charging pile load data has a certain periodic structure, and its periodic change characteristics show stability in a large amount of load data. It is common to obtain the user's charging portrait through the charging pile load data of different dates. Specifically, there is a certain correlation between the segmented load data divided according to different dates, which can be obtained through The model obtains different lag orders and moving average lag orders to obtain the error change between the predicted data and the observed data.

[0049] It should be further explained that the degree of correlation between the charging pile target load data and the historical load data of each charging pile in the historical load data sequence can be obtained by analyzing the error between the predicted data of each charging pile historical load data for the time period where the charging pile target load data is located and the actual observed data of the charging pile target load data.

[0050] Specifically, through The model predicts the historical load data of each charging pile in the historical load data sequence of the charging pile, obtains the predicted data of the historical load data of each charging pile in the historical load data sequence of the charging pile, and obtains the time and date interval length between the historical load data of each charging pile in the historical load data sequence of the charging pile and the target load data of the charging pile; then the first The calculation expression of the correlation degree between the historical load data of a charging pile and the target load data of the charging pile is:

[0051]

[0052] In the formula, Indicates the number of the charging pile historical load data sequence The degree of correlation between the historical load data of each charging pile and the target load data of the charging pile; Indicates the number of the charging pile historical load data sequence Prediction data of historical load data of charging piles; Indicates the number of the charging pile historical load data sequence Prediction data of historical load data of charging piles; Indicates the target load data of the charging pile; Indicates the number of the charging pile historical load data sequence The length of the time interval between the historical load data of each charging pile and the target load data of the charging pile; Represents DTW distance, which is used to indicate the similarity between two data sequences; Indicates taking the absolute value; represents the linear normalization function.

[0053] So far, the correlation degree between each charging pile historical load data and the charging pile target load data in the charging pile historical load data sequence is obtained through the above method.

[0054] Step S003: Obtain several historical neighborhood data sequences and target neighborhood data sequences, and obtain the association confidence level between the target neighborhood data sequence and each historical neighborhood data sequence according to the association level between each charging pile historical load data and the charging pile target load data in the charging pile historical load data sequence.

[0055] It should be noted that since the degree of correlation between the target load data of the charging pile and each historical load data of the charging pile in the historical load data sequence of the charging pile is only the individual correlation feature between the target load data of the charging pile and each historical load data of the charging pile in the historical load data sequence of the charging pile, its representation of the periodic structure of the load data is not accurate enough, and there will be fluctuations in the correlation in the historical load data sequence of the charging pile with a larger data base. Therefore, the overall similarity feature of the historical load data sequence of the charging pile can be combined, which is specifically manifested in that the degree of correlation between each historical load data obtained in the historical load data sequence of the charging pile and the target load data of the charging pile has a certain similarity structure and has strong stability. Further, the correlation confidence level can be obtained, which is used to characterize the stability of the correlation of the target load data of the charging pile in the historical load data sequence of the charging pile, so as to further reduce the prediction error and improve The model's predictive accuracy.

[0056] 1. Obtain several historical neighborhood data sequences and target neighborhood data sequences.

[0057] Preset a parameter , wherein this embodiment is based on The present embodiment is not specifically limited. It depends on the specific implementation situation.

[0058] Specifically, the charging pile historical load data adjacent to the charging pile target load data in the charging pile historical load data sequence is recorded as the first historical load data; The historical load data sequence consisting of the historical load data of the charging piles and the first historical load data is recorded as the target neighborhood data sequence; in the charging pile historical load data sequence, any charging pile historical load data except all the charging pile historical load data of the target neighborhood data sequence is recorded as the second historical load data, and the second historical load data is recorded as the left adjacent to the charging pile historical load data sequence. The historical load data sequence consisting of the historical load data of the charging piles and the second historical load data is recorded as a historical neighborhood data sequence. Similarly, several historical neighborhood data sequences are obtained.

[0059] At this point, the target neighborhood data sequence and several historical neighborhood data sequences are obtained.

[0060] 2. Obtain the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence.

[0061] For the historical neighborhood data sequence and the target neighborhood data sequence, obtain the degree of correlation between the historical load data of each charging pile and the target load data of the charging pile in each historical neighborhood data sequence, and obtain the degree of correlation between the historical load data of each charging pile and the target load data of the charging pile in the target neighborhood data sequence; for the A historical neighborhood data sequence, The first The degree of correlation between the historical load data of a charging pile and the target load data of the charging pile and the target neighborhood data sequence The degree of correlation between the historical load data of a charging pile and the target load data of the charging pile The difference between The first in the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence elements, and similarly, get the All elements in the difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence are obtained, and then the first The difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence.

[0062] At this point, a sequence of difference values ​​of the degree of association between each historical neighborhood data sequence and the target neighborhood data sequence is obtained.

[0063] 3. Obtain the correlation confidence level of the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence.

[0064] It should be noted that the fluctuation degree of the difference sequence refers to the cumulative sum of the fluctuation degrees of the value changes of the elements at corresponding positions of multiple difference sequences at each position in the overall difference sequence.

[0065] Specifically, The confidence calculation expression of the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence is:

[0066]

[0067] In the formula, Indicates The confidence level of the difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence; Indicates The first in the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence elements; Represents the first in the sequence of differences in the degree of association between all historical neighborhood data sequences and the target neighborhood data sequence The mean of the elements; Indicates taking the absolute value; is a preset parameter, indicating the total number of elements in the association degree difference sequence.

[0068] At this point, the association confidence level of the association degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence is obtained through the above method.

[0069] Step S004: According to the correlation confidence level of the best correlation degree difference sequence The prediction error related parameters of the model are adjusted to realize the prediction of charging pile load data and thus complete the orderly charging scheduling.

[0070] It should be noted that the prediction error related parameters are adjusted according to the obtained correlation confidence level. Since the correlation confidence level indicates the correlation confidence level between the correlation level sequence and the segmented data correlation level sequence, the correlation confidence level is for The lag order of the model With the moving average lag order A measure of the accuracy of the prediction. This allows for the lag order of the parameter With the moving average lag order adjustments.

[0071] Specifically, the association confidence levels of the association degree difference sequences between all historical neighborhood data sequences and the target neighborhood data sequence are combined into an association confidence degree sequence, and the maximum association confidence level of the association degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence in the association confidence degree sequence is recorded as the best association degree difference sequence; the historical neighborhood data sequence corresponding to the best association degree difference sequence is recorded as the best historical neighborhood data sequence, and for the historical neighborhood data sequence The historical load data of charging piles is shown in the autocorrelation diagram. And partial autocorrelation plot Get the Historical load data of charging piles The lag order of the model With the moving average lag order ; Through the autocorrelation plot And partial autocorrelation plot Obtaining charging pile target load data The lag order of the model With the moving average lag order ; The adjusted forecast error related parameter lag order With the moving average lag order The calculation expression is:

[0072]

[0073] In the formula, represents the lag order of the adjusted forecast error related parameters; represents the moving average lag order of the adjusted forecast error related parameters; Indicates the target load data of the charging pile The number of lags in the model; Indicates the target load data of the charging pile The moving average lag order of the model; Indicates the correlation confidence level of the best correlation difference sequence; is a preset parameter, indicating the total number of elements in the association degree difference sequence.

[0074] At this point, the adjusted lag order of the prediction error related parameters and the adjusted moving average lag order of the prediction error related parameters are obtained.

[0075] It should be noted that according to the adjusted The prediction error related parameters of the model, constructed from this The charging pile charging load data predicted by the model can remove the influence of unstable characteristics. It can be combined with the prediction error redundant parameters set according to the specific implementation environment to further enhance the accommodation of the predicted charging load under the guarantee of prediction accuracy, and ultimately achieve orderly charging scheduling through charging load prediction combined with conventional charging scheduling schemes, so as to achieve the purpose of reasonably allocating charging demand, improving user experience and improving charging utilization efficiency.

[0076] Specifically, the adjusted lag order of the prediction error related parameters and the adjusted moving average lag order of the prediction error related parameters are used as The parameters of the model are The model predicts the charging load data of the charging piles, and the charging piles are scheduled for charging in an orderly manner according to the predicted charging load data of the charging piles.

[0077] At this point, this embodiment is completed.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An orderly charging scheduling method based on charging load prediction, characterized in that: The method comprises the following steps: Obtain charging pile target load data and charging pile historical load data sequence; Obtaining prediction data of each charging pile historical load data in the charging pile historical load data sequence; obtaining the degree of correlation between each charging pile historical load data in the charging pile historical load data sequence and the charging pile target load data according to the prediction data of each charging pile historical load data in the charging pile historical load data sequence; Obtain several historical neighborhood data sequences and target neighborhood data sequences; obtain a correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence according to the correlation degree between each charging pile historical load data and the charging pile target load data in the charging pile historical load data sequence; obtain the correlation confidence degree of the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence; Obtain the best correlation degree difference sequence according to the correlation confidence degree of the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence; obtain the adjusted lag order of the prediction error related parameter and the adjusted moving average lag order of the prediction error related parameter according to the correlation confidence degree of the best correlation degree difference sequence; and perform orderly charging scheduling of the charging piles according to the adjusted lag order of the prediction error related parameter and the adjusted moving average lag order of the prediction error related parameter; The specific method of obtaining the adjusted lag order of the prediction error related parameter and the adjusted moving average lag order of the prediction error related parameter according to the correlation confidence degree of the best correlation degree difference sequence is as follows: The historical neighborhood data sequence corresponding to the best correlation degree difference sequence is recorded as the best historical neighborhood data sequence. The historical load data of charging piles is shown in the autocorrelation diagram. And partial autocorrelation plot Get the Historical load data of charging piles The lag order of the model With the moving average lag order ; Through the autocorrelation plot And partial autocorrelation plot Obtain charging pile target load data The lag order of the model With the moving average lag order ; The adjusted forecast error related parameter lag order With the moving average lag order The calculation expression is: In the formula, Indicates the correlation confidence level of the best correlation difference sequence; are preset parameters.

2. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The specific method of obtaining the charging pile target load data and the charging pile historical load data sequence includes: The load data of the charging pile on the current day and the load data of the charging pile on the past several days are collected through electrical sensors; for the charging pile load data of each day, the empty part of the load data is removed, and then the other discrete load data are regularized to obtain the regularized charging pile load data for each day; similarly, the regularized load data of the charging pile on the current day and the regularized load data of the charging pile on the past several days are obtained; the regularized load data of the charging pile on the current day is recorded as the charging pile target load data; the charging pile load data sequence composed of the regularized load data of the charging pile on the past several days is recorded as the charging pile historical load data sequence.

3. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The specific method of obtaining the prediction data of each charging pile historical load data in the charging pile historical load data sequence includes: pass The model predicts the historical load data of each charging pile in the historical load data sequence of the charging pile to obtain the predicted data of the historical load data of each charging pile in the historical load data sequence of the charging pile.

4. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The method of obtaining the correlation degree between each charging pile historical load data in the charging pile historical load data sequence and the charging pile target load data according to the predicted data of each charging pile historical load data in the charging pile historical load data sequence includes the following specific methods: Obtain the time and date interval length between each charging pile historical load data and the charging pile target load data in the charging pile historical load data sequence; then The calculation expression of the correlation degree between the historical load data of a charging pile and the target load data of the charging pile is: In the formula, Indicates the number of the charging pile historical load data sequence The degree of correlation between the historical load data of each charging pile and the target load data of the charging pile; Indicates the number of the charging pile historical load data sequence Prediction data of historical load data of charging piles; Indicates the number of the charging pile historical load data sequence Prediction data of historical load data of charging piles; Indicates the target load data of the charging pile; Indicates the number of the charging pile historical load data sequence The length of the time interval between the historical load data of each charging pile and the target load data of the charging pile; represents the DTW distance; Indicates taking the absolute value; represents the linear normalization function.

5. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The specific method of obtaining a plurality of historical neighborhood data sequences and a target neighborhood data sequence includes: Preset Parameters , record the charging pile historical load data adjacent to the charging pile target load data in the charging pile historical load data sequence as the first historical load data; The historical load data sequence consisting of the historical load data of the charging piles and the first historical load data is recorded as the target neighborhood data sequence; in the charging pile historical load data sequence, any charging pile historical load data except all the charging pile historical load data of the target neighborhood data sequence is recorded as the second historical load data, and the second historical load data is recorded as the left adjacent to the charging pile historical load data sequence. The historical load data sequence consisting of the historical load data of the charging piles and the second historical load data is recorded as a historical neighborhood data sequence. Similarly, several historical neighborhood data sequences are obtained.

6. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The specific method of obtaining the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence includes: For the historical neighborhood data sequence and the target neighborhood data sequence, obtain the degree of correlation between the historical load data of each charging pile and the target load data of the charging pile in each historical neighborhood data sequence, and obtain the degree of correlation between the historical load data of each charging pile and the target load data of the charging pile in the target neighborhood data sequence; for the A historical neighborhood data sequence, The first The degree of correlation between the historical load data of a charging pile and the target load data of the charging pile and the target neighborhood data sequence The degree of correlation between the historical load data of a charging pile and the target load data of the charging pile The difference between The first in the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence elements, and similarly, get the All elements in the difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence are obtained, and then the first The difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence.

7. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The specific method of obtaining the association confidence level of the association degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence includes: No. The confidence calculation expression of the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence is: In the formula, Indicates The confidence level of the difference sequence of the degree of association between the historical neighborhood data sequence and the target neighborhood data sequence; Indicates The first in the correlation degree difference sequence between the historical neighborhood data sequence and the target neighborhood data sequence elements; Represents the first in the sequence of differences in the degree of association between all historical neighborhood data sequences and the target neighborhood data sequence The mean of the elements; Indicates taking the absolute value; are preset parameters.

8. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The specific method of obtaining the optimal correlation degree difference sequence according to the correlation confidence degree of the correlation degree difference sequence between each historical neighborhood data sequence and the target neighborhood data sequence is as follows: The association confidence levels of the association degree difference sequences between all historical neighborhood data sequences and target neighborhood data sequences are combined into an association confidence level sequence, and the maximum association confidence level of the association degree difference sequence between the historical neighborhood data sequences and the target neighborhood data sequences in the association confidence level sequence is recorded as the optimal association degree difference sequence.

9. The orderly charging scheduling method based on charging load prediction according to claim 1 is characterized in that: The method of orderly charging the charging piles according to the adjusted hysteresis order of the prediction error related parameters and the adjusted moving average hysteresis order of the prediction error related parameters includes the following specific methods: The adjusted lag order of the forecast error related parameters and the adjusted moving average lag order of the forecast error related parameters are taken as The parameters of the model are The model predicts the charging load data of the charging piles, and the charging piles are scheduled for charging in an orderly manner according to the predicted charging load data of the charging piles.

Images