Bus load prediction method and system in travel peak period of holidays and festivals, medium and processor

Abstract

The invention provides a bus load prediction method in the travel peak period of holidays and festivals. The bus load prediction method comprises the steps of collecting historical passenger data entering and leaving a certain region and bus load historical data; the collected data are preprocessed, the passenger variation of each day is calculated, and the bus load variation of each day is calculated; dividing the passenger variable quantity and the corresponding bus load variable quantity into a training set and a verification set; constructing a first neural network model, training the first neural network model by adopting the training set, and adjusting a hidden layer weight parameter of the model; verifying the trained model by adopting a verification set; if the verification is passed, outputting a first standard model for predicting the bus load change in the tourism peak period of the region; if the verification is not passed, returning to the step S3, re-dividing the data set, and then training; and collecting and predicting passenger variation, inputting the passenger variation into the first standard model to calculate and predict the variation of the bus load of the power grid, and further calculating the bus load of the power grid according to the variation.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid load prediction, and in particular to a bus load prediction method, system, medium and processor during holiday travel peak periods. Background Art

[0002] With the changes and development of technology and social life, the power structure, load characteristics, grid form, technical foundation and operation characteristics of the existing power grid have changed significantly. Bus load forecasting is an important basis for formulating grid dispatching operation mode, calculating line transmission capacity, and ensuring safe, reliable and economic operation of the grid. It is also an important guarantee for the fairness and economy of the power spot market and ancillary service market. Its accuracy directly affects the market operation and the quotation behavior of market players.

[0003] The existing technologies for predicting bus loads in power grid systems mostly consider the impact of natural and technical conditions on the power grid, including the impact of weather factors such as temperature, irradiance, wind speed, and renewable energy output on bus load prediction, or the calculation method of bus load prediction is too biased towards the bus load historical data fitting algorithm. These prediction methods have a certain degree of stability in some living and industrial environments, and have better effects in areas where the power grid load is mainly affected by changes in natural environmental factors. However, for some areas with strong personnel mobility and large mobility changes (such as tourist areas), it is impossible to achieve ideal accuracy, that is, the accuracy of bus load prediction is not high.

[0004] In view of this, a bus load prediction method, system, medium and processor for holiday travel peak period are needed. Summary of the invention

[0005] In view of the problem that the power grid bus load prediction cannot achieve ideal accuracy in areas with high mobility and large mobility changes in the prior art, the present invention provides a bus load prediction method, system, medium and processor during the peak holiday tourism period, which can solve the above problem. The specific technical solution is as follows:

[0006] A bus load forecasting method during holiday travel peak period, comprising:

[0007] S1: Collect historical passenger data entering and leaving a certain area, and collect historical bus load data of the power grid on the corresponding date;

[0008] S2: pre-process the collected data, and calculate the daily passenger change based on the passenger data entering and leaving, and calculate the daily bus load change based on the bus load data;

[0009] S3: Divide the passenger change and the corresponding bus load change into a training set and a validation set;

[0010] S4: constructing a first neural network model, and using the training set to train the first neural network model, and adjusting the hidden layer weight parameters of the model;

[0011] S5: Use the validation set to validate the trained model; if the validation passes, output the first qualified model for predicting bus load changes during the peak tourist season in the region; if the validation fails, return to step S3 to re-divide the data set and then perform training;

[0012] S6: Collect and predict the passenger change, and input it into the first standard-reaching model to calculate and predict the change in the power grid bus load, and further calculate the power grid bus load based on the change.

[0013] Furthermore, the preprocessing of the collected data and calculating the daily passenger change according to the passenger data of entry and exit include the following steps:

[0014] Combine the passenger arrivals, passenger departures and bus load data belonging to the same date into one data entry;

[0015] Remove entries with missing data items;

[0016] Passengers are counted by age group and the age group data is incorporated into the data entry; the age groups are classified into teenagers, young people, middle-aged people and the elderly;

[0017] Passengers are counted by gender and the gender data is incorporated into the data entry;

[0018] The daily passenger change is calculated based on the data of incoming and outgoing passengers; the passenger change includes the daily change in the total number of passengers, the change in each age group and the change in gender.

[0019] Furthermore, the collecting and predicting of passenger change amount includes the following steps:

[0020] S61: Since the number of passengers entering a certain area is constantly changing, the number of tickets purchased, weather data, and actual passenger data entering the area on the forecast date are collected;

[0021] S62: Divide the incoming ticket purchase quantity, weather data and actual incoming passenger data into a training set and a validation set;

[0022] S63: construct a second neural network model and train it using the training set, and adjust the hidden layer weight parameters of the model;

[0023] S64: Use the validation set to validate the trained model; if the validation passes, output the second qualified model and use it to predict the passenger data entering the area; if the validation fails, return to step S62 and re-divide the data set before training.

[0024] Furthermore, the collecting and predicting of passenger change amount further includes the following steps:

[0025] S65: Since the number of passengers entering a certain area is constantly changing, the number of tickets purchased, weather data, and data on passengers actually leaving the area on the forecast date are collected;

[0026] S66: Divide the number of departing tickets and the data of passengers who actually depart into a training set and a validation set;

[0027] S67: construct a third neural network model and train it using the training set, and adjust the hidden layer weight parameters of the model;

[0028] S68: Use the validation set to validate the trained model; if the validation passes, output the third qualified model and use it to predict the passenger data leaving the area; if the validation fails, return to step S66 to re-divide the data set and then perform training;

[0029] S69: Calculate the number of passengers changed or the number of population changed in the area based on the predicted incoming passenger data and outgoing passenger data.

[0030] Furthermore, the number of tickets purchased includes the number of tickets purchased by high-speed trains, ferries, and airplane passengers or the number of tickets purchased for park admission; the change in the number of passengers is classified and counted by age group and gender, and the results include the change in the total number of people, the change in each age group, and the change in gender.

[0031] Furthermore, the collection and statistics date of the ticket purchase quantity ends at 24:00 on the day before the prediction day.

[0032] Furthermore, the construction of the first neural network model includes the following steps:

[0033] The calculation formula for the number of hidden layer nodes is:

[0034]

[0035] In the above formula, X1 is the total change of passengers; X3 is the change of young people; X4 is the change of middle-aged people; h is the number of hidden layer nodes; m is the number of input layer nodes; n is the number of output layer nodes; a is an additional parameter; ∑X1 is the total change of this training sample; ∑(X3+X4) is the total change of young and middle-aged people in this training sample.

[0036] Therefore, we can know that the input calculation formula of the hidden layer is:

[0037] z j =∑ i w Xi,j ×X i +b j ;

[0038] In addition, the hidden layer node is neuj, and the function f is selected neuj (x) is used as the activation function of the hidden layer. It can be seen that the input calculation formula of the output layer is:

[0039]

[0040] Therefore, we choose the function f o (x) is used as the activation function of the hidden layer, and the final output calculation formula of the output layer is:

[0041]

[0042] In the above formula, j is the hidden layer node number, ranging from 1 to h; z j is the input of the hidden layer node neuj; z o is the input of the output layer node; w Xi,j is the weight parameter between the input node Xi and the hidden layer node neuj; b j is the bias term of hidden layer node j; b o is the bias term of the output layer node; w neuj is the weight parameter between the hidden layer node neuj and the output layer node; where the activation function f o (x) and f neuj (x) can be the same or different; ∑X1 is the total number of changes in this training sample; ∑(X3+X4) is the total number of changes in young and middle-aged people in this training sample.

[0043] Furthermore, let the actual output be y o , and the minimum mean square error of the output layer output is:

[0044]

[0045] Furthermore, the weight parameter update formula of each layer is:

[0046]

[0047] Furthermore, the update formula of the bias term is:

[0048]

[0049] In the above formula, W is the weight parameter of each layer; η is the learning rate; b is the bias parameter of each layer.

[0050] Furthermore, the method further comprises the following steps:

[0051] Collect date data and meteorological data; the meteorological data includes temperature, ultraviolet intensity and whether there is rain; and use the date data and meteorological data together with the passenger change to train, verify and predict the first neural network model.

[0052] A bus load forecasting system during holiday travel peak periods, applied to the above method, comprises:

[0053] The first collection module is used to collect historical passenger data entering and leaving a certain area, and collects historical bus load data of the power grid on the corresponding date;

[0054] A first calculation module is used to pre-process the collected data and calculate the daily passenger change based on the passenger data entering and leaving, and calculate the daily bus load change based on the bus load data;

[0055] A first division module, which is used to divide the passenger change amount and the corresponding bus load change amount into a training set and a verification set;

[0056] A training module, which is used to construct a first neural network model, train the first neural network model using a training set, and adjust the hidden layer weight parameters of the model;

[0057] A verification module is used to verify the trained model using a verification set; if the verification is passed, the first qualified model is output for predicting bus load changes during the peak tourist season in the region; if the verification is not passed, the module returns to step S3 to re-divide the data set and then conduct training;

[0058] The prediction module is used to collect and predict the change in passenger volume, and input the first standard-reaching model to calculate and predict the change in the power grid bus load, and further calculate the power grid bus load based on the change.

[0059] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned bus load prediction method during the holiday travel peak period.

[0060] A processor is used to run a program, wherein the program executes the above-mentioned bus load prediction method during the holiday travel peak period when running.

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

[0062] 1. For some areas with strong tourism characteristics, the travel behavior of passengers is the main driving force of the region's GDP and the main factor affecting the load of the power grid bus in the region. Therefore, the load of the power grid bus in the region is predicted by the change in the number of passengers in the region, which has strong correlation and high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0064] Figure 1 It is a flow chart of a bus load forecasting method during holiday travel peak period;

[0065] Figure 2 It is a schematic diagram of the first neural network model structure of the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0068] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0069] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0070] Embodiment 1

[0071] like Figure 1The figure shows a flow chart of a bus load forecasting method during the holiday travel peak period, including:

[0072] S1: Collect historical passenger data entering and leaving a certain area, and collect historical bus load data of the power grid on the corresponding date.

[0073] S2: Preprocess the collected data, and calculate the daily passenger change based on the passenger data entering and leaving, and calculate the daily bus load change based on the bus load data.

[0074] In a specific implementation, the collected data is preprocessed, and the daily passenger change is calculated based on the passenger data entering and leaving, including the following steps:

[0075] Combine the passenger arrivals, passenger departures and bus load data belonging to the same date into one data entry;

[0076] Remove entries with missing data items;

[0077] Passengers are counted by age group and the age group data are incorporated into the data entries; the age groups are classified into teenagers, young people, middle-aged people and the elderly; the specific age group corresponding to each classification shall be based on national standards.

[0078] Passengers are counted by gender and the gender data is incorporated into the data entry;

[0079] The daily passenger change is calculated based on the data of incoming and outgoing passengers; the passenger change includes the daily change in the total number of passengers, the change in each age group and the change in gender.

[0080] Furthermore, based on the number of passengers, the passengers can be classified, for example, by age, into teenagers, young people, middle-aged people and the elderly. The number of passengers per day can be classified and counted by age, and then used as a parameter to train and verify the first neural network and make the final prediction. By classifying the passengers by age, we can know the age distribution of passengers entering the area, because passengers of different age groups have different consumption and entertainment habits, and use different tourist facilities and equipment, which will affect the use of electric energy.

[0081] Similarly, passenger data is classified and counted by gender, and gender is used as a parameter in training, validation, and prediction calculations.

[0082] S3: Divide the passenger changes and the corresponding bus load changes into a training set and a validation set.

[0083] S4: construct a first neural network model, train the first neural network model using the training set, and adjust the hidden layer weight parameters of the model.

[0084] In this embodiment, Figure 2 As shown, the first neural network model adopts a three-layer structure (input layer, one hidden layer and output layer), where the input node parameters of the input layer are: total passenger change X1, juvenile change X2, youth change X3, middle-aged change X4, elderly change X5, male change X6 and male change X7;

[0085] The output layer has only one output: load change Y1;

[0086] The calculation formula for the number of hidden layer nodes is:

[0087]

[0088] In the above formula, m is the number of input layer nodes, n is the number of output layer nodes, and a is an additional parameter, which is usually an empirical parameter with a value of 0-10.

[0089] However, in order to make the accuracy of the present application scheme the same when it is applied to different regions, the value of parameter a is not an empirical constant value, but is calculated in association with the characteristics of the regional population; further, the value of parameter a is determined according to the proportion of the sum of the youth change X3 and the middle-aged change X4 in the total passenger change X1, that is:

[0090]

[0091] Therefore, we can know that the input calculation formula of the hidden layer is:

[0092] z j =∑ i w Xi,j ×X i +b j ;

[0093] In addition, the hidden layer node is neuj, and the function f is selected neuj (x) is used as the activation function of the hidden layer. It can be seen that the input calculation formula of the output layer is:

[0094]

[0095] Therefore, we choose the function f o (x) is used as the activation function of the hidden layer, and the final output calculation formula of the output layer is:

[0096]

[0097] In the above formula, j is the hidden layer node number, ranging from 1 to h; z j is the input of the hidden layer node neuj; z ois the input of the output layer node; w Xi,j is the weight parameter between the input node Xi and the hidden layer node neuj; b j is the bias term of hidden layer node j; b o is the bias term of the output layer node; w neuj is the weight parameter between the hidden layer node neuj and the output layer node; where the activation function f o (x) and f neuj (x) can be the same or different; ∑X1 is the total number of changes in this training sample; ∑(X3+X4) is the total number of changes in young and middle-aged people in this training sample.

[0098] Furthermore, let the actual output be y o , and the minimum mean square error of the output layer output is:

[0099]

[0100] Furthermore, the weight parameter update formula of each layer is:

[0101]

[0102] Furthermore, the update formula of the bias term is:

[0103]

[0104] In the above formula, W is the weight parameter of each layer; η is the learning rate; b is the bias parameter of each layer. If the learning rate is set too large, then in each iteration, the model parameters may cross the optimal solution, resulting in oscillation or divergence, which is called "oscillation phenomenon" or "instability". On the contrary, if the learning rate is set too small, the model will converge to the optimal solution very slowly, and may fall into a local minimum point instead of the global optimal solution.

[0105] The loss function is used to verify whether the output of the first neural network model is within an acceptable range. If it is not within the acceptable range, the weight parameters and bias item values ​​are updated through continuous adjustment to eventually reach acceptable model parameters.

[0106] S5: Use the validation set to validate the trained model; if the validation passes, output the first qualified model for predicting bus load changes during the peak tourist season in the region; if the validation fails, return to step S3 to re-divide the data set and then conduct training.

[0107] S6: Collect and predict the passenger change, and input it into the first standard-reaching model to calculate and predict the change in the power grid bus load, and further calculate the power grid bus load based on the change.

[0108] In a specific implementation, the process of collecting and predicting the passenger change includes the following steps:

[0109] S61: Since the number of passengers entering a certain area is constantly changing, the number of tickets purchased, weather data, and actual passenger data entering the area on the forecast date are collected;

[0110] S62: Divide the incoming ticket purchase quantity, weather data and actual incoming passenger data into a training set and a validation set;

[0111] S63: construct a second neural network model and train it using the training set, and adjust the hidden layer weight parameters of the model;

[0112] S64: Use the validation set to validate the trained model; if the validation passes, output the second qualified model and use it to predict the passenger data entering the area; if the validation fails, return to step S62 and re-divide the data set before training.

[0113] In a specific implementation, the collecting and predicting of passenger change volume also includes the following steps:

[0114] S65: Since the number of passengers entering a certain area is constantly changing, the number of tickets purchased, weather data, and data on passengers actually leaving the area on the forecast date are collected;

[0115] S66: Divide the number of departing tickets, weather data, and actual departing passenger data into a training set and a validation set;

[0116] S67: construct a third neural network model and train it using the training set, and adjust the hidden layer weight parameters of the model;

[0117] S68: Use the validation set to validate the trained model; if the validation passes, output the third qualified model and use it to predict the passenger data leaving the area; if the validation fails, return to step S66 to re-divide the data set and then perform training;

[0118] S69: Calculate the number of passengers changed or the number of population changed in the area based on the predicted incoming passenger data and outgoing passenger data.

[0119] Furthermore, the second neural network model and the third neural network model structure may adopt existing model structures.

[0120] Furthermore, the number of tickets purchased includes the number of tickets purchased by high-speed train, ferry, and airplane passengers or the number of tickets purchased for park admission; further, the number of tickets purchased.

[0121] Furthermore, the collection and statistics date of the ticket purchase quantity ends at 24:00 on the day before the prediction day.

[0122] For some areas with a strong tourism nature, the travel behavior of passengers is the main driving force for the region's GDP and the main factor affecting the load of the region's power grid bus. Therefore, predicting the load of the region's power grid bus based on the change in the number of passengers in the region has strong correlation and high accuracy.

[0123] The present application scheme is not only applicable to the load forecasting of the main busbar of the power grid in a larger area, such as a province, but also applicable to the load forecasting of the power grid busbar in a small or medium-sized area in the region, such as a city or a tourist area.

[0124] Furthermore, different tourist facilities and equipment in tourist areas are used differently in different seasons. For example, water-related entertainment equipment and facilities are more commonly used in summer, while ice, snow, and indoor entertainment equipment and facilities are more commonly used in winter. Therefore, training, verification, and prediction based on the age and gender characteristics of tourists according to the season can more effectively predict the load changes of the power grid bus.

[0125] Furthermore, load forecasting can be carried out in combination with the weather, such as the temperature, ultraviolet light, whether there is rain, etc. Among them, the temperature and ultraviolet light have different impacts on people of different genders, especially on women's travel behavior; and whether there is rain has an impact on the overall travel mode of the population; even in coastal areas, beaches are popular tourist destinations, and wave height prediction data can be collected. When the wave height is not suitable for playing on the beach, tourists can easily transfer to other tourist projects, which in turn affects the use of electric energy; the above-mentioned prediction data can be collected from the Meteorological Bureau.

[0126] In a specific implementation, in this embodiment, when calculating the power grid bus load, the first neural network model adopts a BP neural network model, and other neural network models that can achieve the same effect may also be adopted.

[0127] For some areas with a strong tourism nature, the travel behavior of passengers is the main driving force for the region's GDP and the main factor affecting the load of the region's power grid bus. Therefore, predicting the load of the region's power grid bus based on the change in the number of passengers in the region has strong correlation and high accuracy.

[0128] Embodiment 2

[0129] A bus load forecasting system during holiday travel peak periods, applied to the above method, comprises:

[0130] The first collection module is used to collect historical passenger data entering and leaving a certain area, and collects historical bus load data of the power grid on the corresponding date;

[0131] A first calculation module is used to pre-process the collected data and calculate the daily passenger change based on the passenger data entering and leaving, and calculate the daily bus load change based on the bus load data;

[0132] A first division module, which is used to divide the passenger change amount and the corresponding bus load change amount into a training set and a verification set;

[0133] A training module, which is used to construct a first neural network model, train the first neural network model using a training set, and adjust the hidden layer weight parameters of the model;

[0134] A verification module is used to verify the trained model using a verification set; if the verification is passed, the first qualified model is output for predicting bus load changes during the peak tourist season in the region; if the verification is not passed, the module returns to step S3 to re-divide the data set and then conduct training;

[0135] The prediction module is used to collect and predict the change in passenger volume, and input the first standard-reaching model to calculate and predict the change in the power grid bus load, and further calculate the power grid bus load based on the change.

[0136] Embodiment 3

[0137] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned bus load prediction method during the holiday travel peak period.

[0138] Embodiment 4

[0139] A processor is used to run a program, wherein the program executes the above-mentioned bus load prediction method during the holiday travel peak period when running.

[0140] The present application provides a method for predicting bus load during peak tourist seasons during holidays, comprising: S1: collecting historical passenger data entering and leaving a certain area, and collecting historical bus load data of the power grid on the corresponding date; S2: preprocessing the collected data, and calculating the daily passenger change based on the passenger data entering and leaving, and calculating the daily bus load change based on the bus load data; S3: dividing the passenger change and the corresponding bus load change into a training set and a validation set; S4: constructing a first neural network model, and using the training set to train the first neural network model, and adjusting the hidden layer weight parameters of the model; S5: using the validation set to validate the trained model; if the validation passes, outputting a first qualified model for predicting bus load changes during the tourist season in the region; if the validation fails, returning to step S3 to re-divide the data set and then perform training; S6: collecting and predicting passenger changes, and inputting the first qualified model to calculate and predict the change in the power grid bus load, and further calculating the power grid bus load based on the change. For some areas with a strong tourism nature, the travel behavior of passengers is the main driving force for the region's GDP and the main factor affecting the load of the region's power grid bus. Therefore, predicting the load of the region's power grid bus based on the change in the number of passengers in the region has strong correlation and high accuracy.

[0141] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0142] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0143] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0144] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0145] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A bus load forecasting method during holiday travel peak period, characterized in that: include: S1: Collect historical passenger data entering and leaving a certain area, and collect historical bus load data of the power grid on the corresponding date; S2: pre-process the collected data, and calculate the daily passenger change based on the passenger data entering and leaving, and calculate the daily bus load change based on the bus load data; S3: Divide the passenger change and the corresponding bus load change into a training set and a validation set; S4: constructing a first neural network model, and using the training set to train the first neural network model, and adjusting the hidden layer weight parameters of the model; S5: Use the validation set to validate the trained model; if the validation passes, output the first qualified model for predicting bus load changes during the peak tourist season in the region; if the validation fails, return to step S3 to re-divide the data set and then perform training; S6: Collect and predict the passenger change, and input it into the first standard-reaching model to calculate and predict the change in the power grid bus load, and further calculate the power grid bus load based on the change.

2. The bus load forecasting method during holiday travel peak period according to claim 1 is characterized in that: The preprocessing of the collected data and calculating the daily passenger change according to the passenger data of entry and exit include the following steps: Combine the passenger arrivals, passenger departures and bus load data belonging to the same date into one data entry; Remove entries with missing data items; Passengers are counted by age group and the age group data is incorporated into the data entry; the age groups are classified into teenagers, young people, middle-aged people and the elderly; Passengers are counted by gender and the gender data is incorporated into the data entry; The daily passenger change is calculated based on the data of incoming and outgoing passengers; the passenger change includes the daily change in the total number of passengers, the change in each age group and the change in gender.

3. The bus load forecasting method during holiday travel peak period according to claim 1 is characterized in that: The collecting and predicting of passenger change volume comprises the following steps: S61: Collect the number of tickets purchased and weather data for entering a certain area on the forecast date and the data of passengers actually entering the area; S62: Divide the incoming ticket purchase quantity, weather data and actual incoming passenger data into a training set and a validation set; S63: construct a second neural network model and train it using the training set, and adjust the hidden layer weight parameters of the model; S64: Use the validation set to validate the trained model; if the validation passes, output the second qualified model and use it to predict the passenger data entering the area; if the validation fails, return to step S62 and re-divide the data set before training.

4. The bus load forecasting method during holiday travel peak period according to claim 3 is characterized in that: The collecting and predicting of passenger change amount further includes the following steps: S65: Collect the number of tickets purchased and weather data of passengers who will leave a certain area on the forecast date and the data of passengers who actually leave the area; S66: Divide the number of departing tickets and the data of passengers who actually depart into a training set and a validation set; S67: construct a third neural network model and train it using the training set, and adjust the hidden layer weight parameters of the model; S68: Use the validation set to validate the trained model; if the validation passes, output the third qualified model and use it to predict the passenger data leaving the area; if the validation fails, return to step S66 to re-divide the data set and then perform training; S69: Calculate the number of passengers changed or the number of population changed in the area based on the predicted incoming passenger data and outgoing passenger data.

5. The method for predicting bus load during holiday travel peak periods according to claim 4, characterized in that: The collection and statistics date of the ticket purchase quantity is up to 24:00 on the day before the prediction day.

6. The bus load forecasting method during holiday travel peak period according to claim 1 is characterized in that: The construction of the first neural network model comprises the following steps: The calculation formula for the number of hidden layer nodes is: In the above formula, X1 is the total change of passengers; X3 is the change of young people; X4 is the change of middle-aged people; h is the number of hidden layer nodes; m is the number of input layer nodes; n is the number of output layer nodes; a is an additional parameter; ∑X1 is the total change of this training sample; ∑(X3+X4) is the total change of young and middle-aged people in this training sample.

7. The bus load forecasting method during holiday travel peak period according to claim 1 is characterized in that: The following steps are also included: Collect date data and meteorological data; the meteorological data includes temperature, ultraviolet intensity and whether there is rain; and use the date data and meteorological data together with the passenger change to train, verify and predict the first neural network model.

8. A bus load forecasting system for peak holiday travel periods, characterized in that: The method applied to any one of claims 1 to 7, comprising: The first collection module is used to collect historical passenger data entering and leaving a certain area, and collects historical bus load data of the power grid on the corresponding date; A first calculation module is used to pre-process the collected data and calculate the daily passenger change based on the passenger data entering and leaving, and calculate the daily bus load change based on the bus load data; A first division module, which is used to divide the passenger change amount and the corresponding bus load change amount into a training set and a verification set; A training module, which is used to construct a first neural network model, train the first neural network model using a training set, and adjust the hidden layer weight parameters of the model; A verification module is used to verify the trained model using a verification set; if the verification is passed, the first qualified model is output for predicting bus load changes during the peak tourist season in the region; if the verification is not passed, the module returns to step S3 to re-divide the data set and then conduct training; The prediction module is used to collect and predict the change in passenger volume, and input the first standard-reaching model to calculate and predict the change in the power grid bus load, and further calculate the power grid bus load based on the change.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the bus load prediction method during the holiday travel peak period as described in any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the bus load forecasting method during the holiday travel peak period as described in any one of claims 1 to 7.

Images