Charging pile parking space management method and system

Abstract

The present invention relates to the field of charging pile parking space management, and specifically to a charging pile parking space management method and system. Based on basic data, departure volume, a traffic flow causal model of several roads within a certain range around a parking lot that affect its charging pile parking space, and a weather causal model, a neural network model for predicting the occupancy of charging pile parking spaces at any time is finally constructed. Then, based on the parking request information of fuel vehicle users to enter the parking lot obtained in real time, if it is predicted that the charging pile parking space is full within the departure time, the current vehicle income when the current fuel vehicle user is allowed to enter and the charging vehicle income when the current fuel vehicle user is not allowed to enter are calculated respectively, and whether the current fuel vehicle is allowed to enter the parking lot is determined in a manner with the highest income. The problem that it is difficult to determine whether it is necessary to open the charging pile parking space to fuel vehicles in the prior art, thereby ensuring the maximum income of the parking lot, is solved.

Description

Technical Field

[0001] The present invention relates to the field of charging pile parking space management, and in particular to a charging pile parking space management method and system. Background Art

[0002] In some parking lots, there are often ordinary parking spaces for fuel vehicles and charging pile parking spaces for electric vehicles. However, in actual use, some fuel vehicle owners will park their fuel vehicles in the charging pile parking spaces after the ordinary parking spaces in the parking lot are full. This method will not only cause difficulties in charging electric vehicles in the future, but also affect the revenue of the parking lot.

[0003] Therefore, some parking lots will set parking locks on charging pile parking spaces to prevent fuel vehicles from entering. At this time, the owner can scan the code to enter the vehicle information or take a photo of the license plate number to inform the parking lot whether the vehicle is an electric car. If it is an electric car, the parking lock will be opened. However, this method may cause a large number of charging pile parking spaces to be idle after the ordinary parking spaces are full. If the number of electric cars in the parking lot is small, a large number of charging pile parking spaces may be idle. If no parking locks are set, a large number of charging pile parking spaces may be occupied by fuel vehicles, making electric vehicles unusable, which will also affect the parking lot revenue. Therefore, the existing technology makes it difficult to determine whether it is necessary to open charging pile parking spaces to fuel vehicles, thereby making it difficult to ensure the maximum revenue of the parking lot. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides a charging pile parking space management method and system, which solves the problem that it is difficult to determine whether to open the charging pile parking space to fuel vehicles, thereby making it difficult to ensure the maximum profit of the parking lot.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A charging pile parking space management method, the management method comprising the following steps:

[0007] S1. Obtain basic data including the maximum number of charging pile parking spaces in the parking lot, the number of charging pile parking spaces in use at the current moment, the number of charging pile parking spaces in use within a period of time, the parking unit price of the charging pile parking spaces, the charging unit price, the vehicle license plate number, the parking time and the parking duration;

[0008] S2. Predicting the parking time, departure time and departure volume of vehicles at the charging pile parking space at any time relative to the current time based on the basic data;

[0009] S3. Based on the basic data, a traffic flow causal model and a weather causal model are constructed for several roads within a certain range around the parking lot that affect the charging pile parking spaces, and a neural network model is constructed in combination with the departure volume to predict the occupancy of the charging pile parking spaces at any time;

[0010] S4. Obtain parking request information of the fuel vehicle user who is currently waiting to enter the parking lot, and determine whether it is necessary to occupy the charging pile parking space for parking;

[0011] If yes, proceed to step S5;

[0012] If not, the current fuel vehicle user is allowed to enter, and then the process ends;

[0013] S5. Calculate the current vehicle revenue when the current fuel vehicle user is allowed to enter and the charging vehicle revenue when the current fuel vehicle user is not allowed to enter respectively;

[0014] S6. Determine whether the current vehicle benefit is greater than the charging vehicle benefit;

[0015] If yes, then the current fuel vehicle user is allowed to enter, and then the process ends;

[0016] If not, the current fuel vehicle user is prevented from entering and the process ends.

[0017] Preferably, in step S2, the following steps are specifically included:

[0018] S21, obtaining the vehicle license plate number, parking time and parking duration of the vehicle located in the charging pile parking space in the parking lot, and collecting historical vehicle information of the parking duration of the vehicle located in the charging pile parking space at any time;

[0019] S22. Calculate the first probability of the vehicle parking for any length of time at any time based on the historical vehicle information; the calculation formula of the first probability is:

[0020] ;

[0021] In the above formula, Indicates that at the i-th moment, the parking time of the vehicle in the charging pile parking space is The first probability when Indicates that at the i-th moment, the parking time of the vehicle in the charging pile parking space is The number of vehicles, represents the total number of vehicles located in the charging pile parking space at the i-th moment;

[0022] S23, calculating the second probability of the parking time of a vehicle corresponding to any license plate number at any time according to the historical vehicle information; the calculation formula of the second probability is:

[0023] ;

[0024] In the above formula, Any license plate number The parking time of the corresponding vehicle at the i-th moment is The second probability, Indicates that at the i-th moment, the license plate number is The parking time of vehicles is The number of times Indicates that at the i-th moment, the license plate number is The total number of stops of vehicles;

[0025] S24, predicting the departure time of the vehicle at the current charging pile parking space according to the first probability and the second probability of the corresponding vehicle;

[0026] S25. Calculate the number of vehicles leaving the charging pile parking space at any time relative to the number of vehicles leaving the charging pile parking space at the current time according to the vehicle leaving time. The calculation formula for the leaving amount is:

[0027] ;

[0028] In the above formula, Indicates the number of vehicles on the charging pile parking space at any time relative to the current time When the number of vehicles leaving the charging pile parking space is Indicates the current moment, i.e. At this moment, the vehicle in the charging pile parking space Indicates the departure time is The moment and The number of vehicles between moments.

[0029] Preferably, in step S24, the following steps are specifically included:

[0030] S241, determining in sequence whether the license plate number of the vehicle currently parked in the charging pile parking space already exists in the historical vehicle information at the current moment;

[0031] If yes, proceed to step S242;

[0032] If not, proceed to step S243;

[0033] S242, selecting the parking time corresponding to the second probability with the largest value, and calculating the departure time of the vehicle according to the time when the vehicle enters the parking lot, and then proceeding to step S25;

[0034] S243: Select the parking time corresponding to the first probability with the largest value, and calculate the departure time of the vehicle according to the time when the vehicle enters the parking lot.

[0035] Preferably, in step S3, the following steps are specifically included:

[0036] S31, taking the parking lot as the center, establishing a number of initial statistical circles with gradient radii;

[0037] S32, obtaining the traffic volume and weather on several roads within the initial statistical circle that affect the parking spaces for charging piles in the parking lot within a period of time;

[0038] S33, establishing a traffic flow causal model based on the traffic flow on several roads that affect the parking spaces of charging piles in the parking lot within the initial statistical circle over a period of time;

[0039] S34. A weather causal model is established based on several weather factors that affect the parking spaces of charging piles in the initial statistical circle over a period of time. The expression of the weather causal model is:

[0040] ;

[0041] In the above formula, It represents the predicted value of the occupancy of charging pile parking spaces at time m and with a statistical circle radius of r in the weather causal model. is a time series variable, represents the predicted value of the occupancy of charging pile parking spaces at time mt and when the radius of the statistical circle is r, d represents the number of weather factors, the total number of which is D, represents the vector of weather factor d at time mt and with a statistical circle radius of r, and Respectively represent the time weight and weather factor weight;

[0042] S35, constructing a neural network model for predicting the occupancy of charging pile parking spaces based on the traffic flow causal model, the weather causal model, and the departure volume;

[0043] S36. Calculate the error of the predicted value of the number of parking spaces occupied by charging piles under different statistical circle radii. The calculation formula for the error of the predicted value of the number of parking spaces occupied by charging piles is:

[0044] ;

[0045] In the above formula, It represents the error of the predicted value of the occupancy of charging pile parking spaces when the radius of the statistical circle is r, N represents the total number of samples, It represents the predicted value of the charging pile parking space occupancy of the nth sample at the mth moment and when the radius of the statistical circle is r. is the predicted value of the occupancy of charging pile parking spaces at the mth moment;

[0046] S37, selecting the statistical circle radius corresponding to the minimum error of the predicted value of the charging pile parking space occupancy, and predicting the charging pile parking space occupancy at any time of the day based on the neural network model of the predicted value of the charging pile parking space occupancy under this statistical circle radius.

[0047] Preferably, in step S32, the following steps are specifically included:

[0048] S321, obtaining several roads within the initial statistical circle that affect the parking spaces for charging piles in the parking lot, as well as the average speed and weather of vehicles on the roads;

[0049] S322, respectively setting a signal transmitting device and a signal receiving device for collecting CSI signals on both sides of the motor vehicle lane of the road;

[0050] S323, calculating the travel time range between the vehicle passing the signal transmitting device and the signal receiving device according to the driving speed and vehicle length of the vehicle on the road;

[0051] S324, setting an amplitude threshold, identifying a peak in the CSI signal that is greater than the amplitude threshold, and extracting a CSI signal corresponding to a travel time range with the peak as the center to obtain a primary signal;

[0052] S325, using a low-pass filter to reduce noise on the primary signal to obtain a noise-reduced signal;

[0053] S326, performing dimensionality reduction processing on the noise reduction signal to obtain a dimensionality reduction signal;

[0054] S327, performing short-time inverse Fourier transform on the dimension-reduced signal to obtain a frequency spectrum;

[0055] S328, identifying the spectrum graph by a convolutional neural network model to obtain the number of vehicles in the spectrum graph;

[0056] S329. Calculate the traffic volume based on the number of vehicles on the road within a period of time; the calculation formula for the traffic volume is:

[0057] ;

[0058] In the above formula, It means that when the radius of the statistical circle is r, at the time mt Traffic volume on the road, Indicated in The number of vehicles in the time period corresponding to the time, Indicated in The length of the time period corresponding to the moment.

[0059] Preferably, in step S323, the following steps are specifically included:

[0060] S3231, obtaining the maximum speed, minimum speed, maximum vehicle length and minimum vehicle length of vehicles on the road to obtain road data;

[0061] S3232. Calculate the travel time range of vehicles passing the signal transmitting device and the signal receiving device on the road according to the road data; the expression of the travel time range is: ,in, , , in the above formula, represents the minimum travel time, represents the maximum travel time, and Respectively represent the maximum and minimum lengths of vehicles on the road. and They represent the maximum and minimum speeds of vehicles on the road respectively.

[0062] Preferably, in step S326, the following steps are specifically included:

[0063] S3261. Construct a sample matrix according to the noise reduction signal. The expression of the sample matrix is:

[0064] ;

[0065] In the above formula, represents the sample matrix, represents the CSI value at the jth time point in the i-th denoised signal, and the sample matrix has f rows and F columns;

[0066] S3262. Construct a column vector of the sample matrix; the expression of the column vector is:

[0067] ;

[0068] in, represents the column vector of the jth column of the sample matrix, is the average value of all data in the jth column of the sample matrix, Represents the data in the fth row and jth column of the sample matrix;

[0069] S3263. Calculate the covariance matrix of the sample matrix; the expression of the covariance matrix is:

[0070] ;

[0071] in, , is a column vector and The covariance of

[0072] S3264, performing singular value decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors, and constructing an eigenvector matrix using the eigenvectors corresponding to the eigenvalues ​​with the largest values; the expression of the eigenvector matrix is: ,in, is the eigenvector matrix, represents the hth eigenvector in the eigenvector matrix;

[0073] S3265. Obtain a characteristic signal according to the characteristic vector matrix and the sample matrix; the expression of the characteristic signal is:

[0074] ;

[0075] In the above formula, is the characteristic signal, is the eigenvector matrix, is the sample matrix;

[0076] S3266. Calculate the similarity between each signal waveform in the characteristic signal and the noise reduction signal, and mark the signal waveform in the characteristic signal with the highest similarity as the dimension reduction signal; the calculation formula of the similarity is:

[0077] ;

[0078] In the above formula, Indicates similarity, and Respectively represent any signal waveform and noise reduction signal in the characteristic signal, Indicates the maximum value of the x-axis in the feature signal and the noise reduction signal.

[0079] Preferably, in step S5, the following steps are specifically included:

[0080] S51, predicting the departure time of the current fuel vehicle user;

[0081] S52, obtaining a predicted value of the number of parking spaces occupied by charging piles at any time before the departure time;

[0082] S53, determining in sequence whether the predicted value of the charging pile parking space occupancy at each moment is greater than the maximum number of charging pile parking spaces;

[0083] If yes, proceed to step S54;

[0084] If not, the current fuel vehicle user is allowed to enter, and then the process ends;

[0085] S54, obtaining the time when the predicted value of the charging pile parking space occupancy is equal to the maximum number of charging pile parking spaces;

[0086] S55. Calculate the current vehicle revenue from allowing the current fuel vehicle user to enter until the current vehicle leaves; the calculation formula for the current vehicle revenue is:

[0087] ;

[0088] In the above formula, Indicates the license plate number is The current vehicle revenue of vehicles entering the charging pile parking space, Indicates the parking unit price of the charging pile parking space, Indicates the license plate number is The predicted value of the parking time of the vehicle;

[0089] S56. Calculate the benefit of the charging vehicle when the current fuel vehicle user is not allowed to enter and the current vehicle leaves; the calculation formula for the charging vehicle benefit is:

[0090] ;

[0091] represents the revenue of charging vehicles, It represents the sum of the parking price and charging price of the charging pile parking space. Indicates the license plate number is The predicted parking time of the vehicle.

[0092] Preferably, in step S35, the expression of the neural network model of the predicted value of the charging pile parking space occupancy is:

[0093] ;

[0094] In the above formula, It represents the predicted value of the occupancy of charging pile parking spaces at time m in the neural network model when the statistical semicircle is r. represents the bth neuron in the hidden layer of the neural network model, and the total number is B. represents the activation function of the b-th neuron, represents the weight matrix of the b-th neuron, represents the bias of the b-th neuron, It represents the number of vehicles leaving the charging pile at time m relative to the current time. Indicates the total bias.

[0095] The technical solution also provides a management system for implementing the charging pile parking space management method, including a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, any one of the charging pile parking space management methods is implemented.

[0096] Compared with the prior art, the present invention provides a charging pile parking space management method and system, which has the following beneficial effects:

[0097] 1. The present invention is based on basic data, departure volume, traffic flow causal models of several roads within a certain range around the parking lot that affect the charging pile parking spaces, and weather causal models, and finally constructs a neural network model to predict the occupancy of charging pile parking spaces at any time. Then, based on the parking request information of fuel vehicle users to enter the parking lot obtained in real time, if the charging pile parking spaces are predicted to be full within the departure time, the current vehicle revenue when the current fuel vehicle user is allowed to enter and the charging vehicle revenue when the current fuel vehicle user is not allowed to enter are calculated respectively, and whether to allow the current fuel vehicle to enter the parking lot is determined in a way that maximizes the revenue.

[0098] 2. The present invention calculates the first probability and the second probability of the vehicle at any parking duration according to the historical vehicle information in the parking lot and whether the corresponding vehicle has been parked in the parking lot at the corresponding time, thereby more accurately calculating the departure time of the vehicle in the current charging pile parking space in two ways, thereby calculating the departure amount of the vehicles in the charging pile parking space at any time relative to the vehicles in the charging pile parking space at the current time.

[0099] 3. The present invention counts the traffic volume on the road based on CSI signals, thereby avoiding infringement of the privacy of the car owner during data collection, thereby having higher universality and ensuring the legitimacy of the data source.

[0100] 4. The present invention processes the noise reduction signal to finally obtain a dimensionality reduction signal that can more clearly characterize the characteristics of the CSI signal, thereby further extracting the characteristics of the CSI signal fluctuation caused by the vehicle, so that the vehicle flow can be counted based on a clearer spectrum diagram.

[0101] 5. The present invention obtains the parking request information of the fuel vehicle user who is about to enter the parking lot in real time. If there is an ordinary parking space, the vehicle will be allowed to enter the parking lot directly. If there is no ordinary parking space, the fuel vehicle will be allowed to enter the parking lot to obtain revenue based on the predicted departure time of the fuel vehicle. If there are always extra charging pile parking spaces within the departure time, the fuel vehicle will be allowed to enter the charging pile parking space to obtain revenue. If the charging pile parking space is predicted to be full within the departure time, whether to allow the current fuel vehicle to enter the parking lot will be determined in a way that maximizes revenue. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0103] Figure 1 This is a flow chart of the charging pile parking space management method of the present invention;

[0104] Figure 2It is a schematic diagram of the dimensionality reduction signal image of the present invention. DETAILED DESCRIPTION

[0105] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0106] Those skilled in the art can understand that all or part of the steps in the following embodiments can be completed by instructing the relevant hardware through a program, so the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0107] In order to solve the problem that it is difficult to determine whether to open the charging pile parking space to fuel vehicles in the prior art, and thus it is difficult to ensure the maximum profit of the parking lot, the present invention provides a charging pile parking space management method, by predicting the occupancy of the charging pile parking space in the parking lot at any time in the future, and when the current vehicle to be entered is a fuel vehicle, calculating the profit of allowing it to enter the parking lot and not allowing it to enter the parking lot, so as to determine whether to open the charging pile parking space for the fuel vehicle to use based on the principle of maximizing profit, such as Figure 1 As shown, the management method includes the following steps:

[0108] S1. Obtain basic data including the maximum number of charging pile parking spaces in the parking lot, the number of charging pile parking spaces in use at the current moment, the number of charging pile parking spaces in use within a period of time, the parking unit price of the charging pile parking spaces, the charging unit price, the vehicle license plate number, the parking time and the parking duration;

[0109] S2. Predict the parking time, departure time and departure volume of vehicles in the charging pile parking space at any time relative to the current time based on the basic data; parking lots are often set up to meet the parking needs of specific groups, such as near shopping malls, near residential areas, etc. Therefore, car owners often need to complete specific purposes nearby after parking, so their parking time is often related to the surrounding environment of the parking lot, and their parking time has a certain regularity, and affects the occupancy of the charging pile parking space in the parking lot. In order to accurately predict the parking time, departure time and departure volume of vehicles in the charging pile parking space at any time relative to the current time, in step S2, the following steps are specifically included:

[0110] S21, obtaining the vehicle license plate number, parking time and parking duration of the vehicle located in the charging pile parking space in the parking lot, and collecting historical vehicle information of the parking duration of the vehicle located in the charging pile parking space at any time; when different car owners park in the same parking lot and at the same time, their behaviors are very similar, so their parking durations are also very similar and regular;

[0111] S22. Calculate the first probability of the vehicle parking for any length of time at any time based on the historical vehicle information; the calculation formula of the first probability is:

[0112] ;

[0113] In the above formula, Indicates that at the i-th moment, the parking time of the vehicle in the charging pile parking space is The first probability when Indicates that at the i-th moment, the parking time of the vehicle in the charging pile parking space is The number of vehicles, represents the total number of vehicles located in the charging pile parking space at the i-th moment;

[0114] S23, calculating the second probability of the parking time of a vehicle corresponding to any license plate number at any time according to the historical vehicle information; the calculation formula of the second probability is:

[0115] ;

[0116] In the above formula, Any license plate number The parking time of the corresponding vehicle at the i-th moment is The second probability, Indicates that at the i-th moment, the license plate number is The parking time of vehicles is The number of times Indicates that at the i-th moment, the license plate number is The total number of stops of vehicles;

[0117] S24, predicting the departure time of the vehicle in the current charging pile parking space according to the first probability and the second probability of the corresponding vehicle; when the license plate number of the corresponding vehicle has not been parked in the parking lot at any time, the parking time and the departure time are calculated by the more universal first probability; if the license plate number of the corresponding vehicle has been parked in the parking lot at any time, the parking time and the departure time are calculated more accurately by the second probability according to its historical parking rules. In step S24, the following steps are specifically included:

[0118] S241, determining in sequence whether the license plate number of the vehicle currently parked in the charging pile parking space already exists in the historical vehicle information at the current moment;

[0119] If yes, proceed to step S242;

[0120] If not, proceed to step S243;

[0121] S242, selecting the parking time corresponding to the second probability with the largest value, and calculating the departure time of the vehicle according to the time when the vehicle enters the parking lot;

[0122] S243, selecting the parking time corresponding to the first probability with the largest value, and calculating the departure time of the vehicle according to the time when the vehicle enters the parking lot;

[0123] In step S242 and step S243, the calculation formula of the departure time is:

[0124] ;

[0125] In the above formula, Indicates the departure time when the vehicle leaves the parking lot. Indicates the time when the vehicle enters the parking lot. Indicates the parking duration corresponding to the second probability or the first probability with the largest value;

[0126] S25. Calculate the number of vehicles leaving the charging pile parking space at any time relative to the number of vehicles leaving the charging pile parking space at the current time according to the vehicle leaving time. The calculation formula for the leaving amount is:

[0127] ;

[0128] In the above formula, Indicates the number of vehicles on the charging pile parking space at any time relative to the current time When the number of vehicles leaving the charging pile parking space is Indicates the current moment, i.e. At this moment, the vehicle in the charging pile parking space Indicates the departure time is The moment and The number of vehicles between moments.

[0129] S3. Based on the basic data, a traffic flow causal model and a weather causal model are constructed for several roads within a certain range around the parking lot that affect the parking spaces of the charging piles, and a neural network model is constructed in combination with the departure volume to predict the occupancy of the parking spaces of the charging piles at any time. After a large number of experimental verifications, it is concluded that the factors affecting the occupancy of the parking spaces of the charging piles are not only correlated with the departure volume of the vehicles on the parking spaces of the charging piles, but also have a large correlation with the traffic flow and weather of the roads within a certain range around the parking lot. The roads within a certain range around the parking lot that affect the parking spaces of the charging piles refer to the roads on which there are vehicles coming directly to the parking lot or vehicles going directly from the parking lot to the roads. If there is no intersection within a certain range around the parking lot and the road, it means that the traffic flow on the road has no effect on the occupancy of the parking spaces of the charging piles in the parking lot, and the traffic flow causal model and the weather causal model are not constructed based on the road. In step S3, the following steps are specifically included:

[0130] S31, taking the parking lot as the center, establishing a number of initial statistical circles with gradient radii;

[0131] S32, obtaining the traffic flow and weather on several roads that affect the parking spaces of charging piles in the initial statistical circle within a period of time; in order to ensure the legitimacy of the data source and avoid infringing on user privacy, a new method is adopted to obtain the traffic flow on several roads. In step S32, the following steps are specifically included:

[0132] S321, obtaining several roads within the initial statistical circle that affect the parking spaces for charging piles in the parking lot, as well as the average speed and weather of vehicles on the roads;

[0133] S322. A signal transmitter and a signal receiving device for collecting CSI signals are respectively arranged on both sides of the motor vehicle lane of the road; the signal transmitter and the signal receiving device for collecting CSI signals are generally arranged at a section of the road where the vehicle speed is relatively stable, and are generally not arranged at an intersection, because the flow of people and vehicles at the intersection is relatively complex, which affects the selection of the subsequent travel time range and is more likely to cause errors in the statistics of the subsequent vehicle flow. When a vehicle passes between the signal transmitter and the signal receiving device, the signal will change significantly, wherein the signal transmitter may be a router, and the signal receiving device only needs to receive and output the CSI signal sent by the router;

[0134] S323, calculating the travel time range of the vehicle passing between the signal transmitting device and the signal receiving device according to the driving speed and the length of the vehicle on the road; the travel time range is related to the length and the speed of the vehicle. Therefore, in order to further illustrate the calculation method of the travel time range, in step S323, the following steps are specifically included:

[0135] S3231, obtaining the maximum speed, minimum speed, maximum vehicle length and minimum vehicle length of vehicles on the road to obtain road data;

[0136] S3232. Calculate the travel time range of vehicles passing the signal transmitting device and the signal receiving device on the road according to the road data; the expression of the travel time range is: ,in, , , in the above formula, represents the minimum travel time, represents the maximum travel time, and Respectively represent the maximum and minimum lengths of vehicles on the road. and They represent the maximum and minimum speeds of vehicles on the road respectively.

[0137] S324, setting an amplitude threshold, identifying a peak in the CSI signal that is greater than the amplitude threshold, and extracting a CSI signal corresponding to a travel time range with the peak as the center to obtain a primary signal;

[0138] S325, using a low-pass filter to reduce noise on the primary signal to obtain a noise-reduced signal;

[0139] S326, performing dimensionality reduction processing on the noise reduction signal to obtain a dimensionality reduction signal; in order to perform dimensionality reduction processing on the noise reduction signal to obtain a dimensionality reduction signal that can fully characterize the change characteristics of the CSI signal, step S326 specifically includes the following steps:

[0140] S3261. Construct a sample matrix according to the noise reduction signal. The expression of the sample matrix is:

[0141] ;

[0142] In the above formula, represents the sample matrix, represents the CSI value at the jth time point in the i-th denoised signal, and the sample matrix has f rows and F columns;

[0143] S3262. Construct a column vector of the sample matrix; the expression of the column vector is:

[0144] ;

[0145] in, represents the column vector of the jth column of the sample matrix, is the average value of all data in the jth column of the sample matrix, Represents the data in the fth row and jth column of the sample matrix;

[0146] S3263. Calculate the covariance matrix of the sample matrix; the expression of the covariance matrix is:

[0147] ;

[0148] in, , is a column vector and The covariance of

[0149] S3264, performing singular value decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors, and constructing an eigenvector matrix using the eigenvectors corresponding to the eigenvalues ​​with the largest values; the expression of the eigenvector matrix is: ,in, is the eigenvector matrix, represents the hth eigenvector in the eigenvector matrix;

[0150] S3265. Obtain a characteristic signal according to the characteristic vector matrix and the sample matrix; the expression of the characteristic signal is:

[0151] ;

[0152] In the above formula, is the characteristic signal, is the eigenvector matrix, is the sample matrix;

[0153] S3266. Calculate the similarity between each signal waveform in the characteristic signal and the noise reduction signal, and mark the signal waveform in the characteristic signal with the highest similarity as the dimension reduction signal; the calculation formula of the similarity is:

[0154] ;

[0155] In the above formula, Indicates similarity, and Respectively represent any signal waveform and noise reduction signal in the characteristic signal, Indicates the maximum value of the x-axis in the characteristic signal and the noise reduction signal. In the matrix represented by the characteristic signal z, there are several rows, each row represents a principal component. Figure 2 It shows that there are 3 rows in the matrix represented by the final characteristic signal z, that is, there are 3 groups such as Figure 2 The main component signal shown in the figure, the final dimensionality reduction signal is the green signal represented by the main component 1, which is most similar to the initial CSI signal image.

[0156] S327, performing short-time inverse Fourier transform on the dimension-reduced signal to obtain a frequency spectrum;

[0157] S328, identifying the spectrum graph by a convolutional neural network model to obtain the number of vehicles in the spectrum graph;

[0158] S329. Calculate the traffic volume based on the number of vehicles on the road within a period of time; the calculation formula for the traffic volume is:

[0159] ;

[0160] In the above formula, It means that when the radius of the statistical circle is r, at the time mt Traffic volume on the road, Indicated in The number of vehicles in the time period corresponding to the time, Indicated in The length of the time period corresponding to the moment.

[0161] S33. A traffic flow causal model is established based on the traffic flow on several roads that affect the parking spaces of charging piles in the initial statistical circle over a period of time. The expression of the traffic flow causal model is:

[0162] ;

[0163] In the above formula, It represents the predicted value of the occupancy of charging pile parking spaces at time m and when the statistical semicircle is r in the traffic flow causal model. It means to round up the elements. For example, when the element in the rounding symbol is 4.2, the result after the rounding symbol operation is 5. is a time series variable, represents the predicted value of the occupancy of charging pile parking spaces at time mt and when the radius of the statistical circle is r, Granger cause The space-time relationship variable, T is the maximum time before time m, and They are time weight and traffic flow weight, It means that when the radius of the statistical circle is r, at the time mt The volume of traffic on the road;

[0164] S34. A weather causal model is established based on several weather factors that affect the parking spaces of charging piles in the initial statistical circle over a period of time. The expression of the weather causal model is:

[0165] ;

[0166] In the above formula, It represents the predicted value of the occupancy of charging pile parking spaces at time m and with a statistical circle radius of r in the weather causal model. is a time series variable, represents the predicted value of the occupancy of charging pile parking spaces at time mt and when the radius of the statistical circle is r, d represents the number of weather factors, the total number of which is D, represents the vector of weather factor d at time mt and with a statistical circle radius of r, and Respectively represent the time weight and weather factor weight;

[0167] S35. Construct a neural network model for predicting the occupancy of charging pile parking spaces based on the traffic flow causal model, the weather causal model, and the amount of departure. The expression of the neural network model for predicting the occupancy of charging pile parking spaces is:

[0168] ;

[0169] In the above formula, It represents the predicted value of the occupancy of charging pile parking spaces at time m and when the radius of the statistical circle is r in the neural network model. represents the bth neuron in the hidden layer of the neural network model, and the total number is B. represents the activation function of the b-th neuron, represents the weight matrix of the b-th neuron, represents the bias of the b-th neuron, It represents the number of vehicles leaving the charging pile at time m relative to the current time. represents the total bias;

[0170] S36. Calculate the error of the predicted value of the number of parking spaces occupied by charging piles under different statistical circle radii. The calculation formula for the error of the predicted value of the number of parking spaces occupied by charging piles is:

[0171] ;

[0172] In the above formula, It represents the error of the predicted value of the occupancy of charging pile parking spaces when the radius of the statistical circle is r, N represents the total number of samples, It represents the predicted value of the charging pile parking space occupancy of the nth sample at the mth moment and when the radius of the statistical circle is r. is the predicted value of the occupancy of charging pile parking spaces at the mth moment;

[0173] S37, selecting the statistical circle radius corresponding to the minimum error of the predicted value of the charging pile parking space occupancy, and predicting the charging pile parking space occupancy at any time of the day based on the neural network model of the predicted value of the charging pile parking space occupancy under this statistical circle radius.

[0174] S4. Obtain parking request information of the fuel vehicle user who is currently waiting to enter the parking lot, and determine whether it is necessary to occupy the charging pile parking space for parking; the determination method is whether the ordinary parking spaces for fuel vehicles in the parking lot are full. If the ordinary parking spaces in the parking lot are full, it is determined that the current fuel vehicle needs to occupy the charging pile parking space for parking;

[0175] If yes, proceed to step S5;

[0176] If not, the current fuel vehicle user is allowed to enter, and then the process ends. That is, if there is an ordinary parking space for fuel vehicles in the parking lot at this time, the current fuel vehicle user is directly allowed to enter;

[0177] S5. Calculate the current vehicle revenue when the current fuel vehicle user is allowed to enter and the charging vehicle revenue when the current fuel vehicle user is not allowed to enter respectively; the current vehicle revenue and the charging vehicle revenue are related to the departure time of the current fuel vehicle user. Therefore, in order to further explain the calculation method of the current vehicle revenue and the charging vehicle revenue, in step S5, the following steps are specifically included:

[0178] S51, predicting the departure time of the current fuel vehicle user, the departure time is predicted by the steps mentioned in step S2;

[0179] S52, obtaining a predicted value of the number of parking spaces occupied by charging piles at any time before the departure time; the predicted value of the number of parking spaces occupied by charging piles is predicted by the steps mentioned in step S3;

[0180] S53, determining in sequence whether the predicted value of the charging pile parking space occupancy at each moment is greater than the maximum number of charging pile parking spaces;

[0181] If yes, proceed to step S54;

[0182] If not, the current fuel vehicle user is allowed to enter, and then the process ends. That is, before the current fuel vehicle leaves the parking lot, there are a large number of vacant charging pile parking spaces in the parking lot. Therefore, in order to increase revenue, the excess charging pile parking spaces are provided to fuel vehicles for parking. The parking space locks on the charging pile parking spaces that are in a locked state can be opened manually or in a predetermined opening sequence, which will not be described in detail here.

[0183] S54, obtaining the time when the predicted value of the occupancy of charging pile parking spaces is equal to the maximum number of charging pile parking spaces; at this time, it means that there are no extra charging pile parking spaces for fuel vehicles to use;

[0184] S55. Calculate the current vehicle revenue from allowing the current fuel vehicle user to enter until the current vehicle leaves; the calculation formula for the current vehicle revenue is:

[0185] ;

[0186] In the above formula, Indicates the license plate number is The current vehicle revenue of vehicles entering the charging pile parking space, Indicates the parking unit price of the charging pile parking space, Indicates the license plate number is The predicted value of the parking time of the vehicle;

[0187] S56. Calculate the benefit of the charging vehicle when the current fuel vehicle user is not allowed to enter and the current vehicle leaves; the calculation formula for the charging vehicle benefit is:

[0188] ;

[0189] represents the revenue of charging vehicles, It represents the sum of the parking price and charging price of the charging pile parking space. Indicates the license plate number is The predicted parking time of the vehicle.

[0190] S6. Determine whether the current vehicle benefit is greater than the charging vehicle benefit;

[0191] If yes, then the current fuel vehicle user is allowed to enter, and then the process ends;

[0192] If not, the current fuel vehicle user is prevented from entering and the process ends.

[0193] When the present invention is implemented, various relevant basic data are first obtained, and then based on the departure volume, the traffic flow causal model of several roads within a certain range around the parking lot that affect the charging pile parking spaces, and the weather causal model, a neural network model for predicting the occupancy of charging pile parking spaces at any time is finally constructed. After that, according to the parking request information of the fuel vehicle user to be entered into the parking lot obtained in real time, if there is an ordinary parking space, it is directly released; if there is no ordinary parking space, according to the predicted departure time of the fuel vehicle, if there are always extra charging pile parking spaces within the departure time, the fuel vehicle is allowed to enter the charging pile parking space to obtain revenue; if the charging pile parking space is predicted to be full within the departure time, the revenue of the current vehicle when the current fuel vehicle user is allowed to enter and the revenue of the charging vehicle when the current fuel vehicle user is not allowed to enter are calculated respectively, and whether to allow the current fuel vehicle to enter the parking lot is determined in a way that maximizes the revenue.

[0194] Corresponding to the charging pile parking space management method provided in the above embodiment, the present embodiment also provides a system for implementing the charging pile parking space management method. Since the charging pile parking space management system provided in this embodiment corresponds to the charging pile parking space management method provided in the above embodiment, the implementation method of the aforementioned charging pile parking space management method is also applicable to the charging pile parking space management system provided in this embodiment, and will not be described in detail in this embodiment.

[0195] The system includes a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, a charging pile parking space management method is implemented.

[0196] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A charging pile parking space management method, characterized in that: The management method includes the following steps: S1. Obtain basic data including the maximum number of charging pile parking spaces in the parking lot, the number of charging pile parking spaces in use at the current moment, the number of charging pile parking spaces in use within a period of time, the parking unit price of the charging pile parking spaces, the charging unit price, the vehicle license plate number, the parking time and the parking duration; S2. Predicting the parking time, departure time and departure volume of vehicles at the charging pile parking space at any time relative to the current time based on the basic data; S3. Based on the basic data, a traffic flow causal model and a weather causal model are constructed for several roads within a certain range around the parking lot that affect the charging pile parking spaces, and a neural network model is constructed in combination with the departure volume to predict the occupancy of the charging pile parking spaces at any time; In step S3, the following steps are specifically included: S31, taking the parking lot as the center, establishing a number of initial statistical circles with gradient radii; S32, obtaining the traffic volume and weather on several roads within the initial statistical circle that affect the parking spaces for charging piles in the parking lot within a period of time; S33, establishing a traffic flow causal model based on the traffic flow on several roads that affect the parking spaces of charging piles in the parking lot within the initial statistical circle over a period of time; S34. A weather causal model is established based on several weather factors that affect the parking spaces of charging piles in the initial statistical circle over a period of time. The expression of the weather causal model is: ； In the above formula, It represents the predicted value of the occupancy of charging pile parking spaces at time m and with a statistical circle radius of r in the weather causal model. Indicates rounding up. is a time series variable, represents the predicted value of the occupancy of charging pile parking spaces at time mt and when the radius of the statistical circle is r, d represents the number of weather factors, the total number of which is D, represents the vector of weather factor d at time mt and with a statistical circle radius of r, and Respectively represent the time weight and weather factor weight; S35, constructing a neural network model for predicting the occupancy of charging pile parking spaces based on the traffic flow causal model, the weather causal model, and the departure volume; S36. Calculate the error of the predicted value of the number of parking spaces occupied by charging piles under different statistical circle radii. The calculation formula for the error of the predicted value of the number of parking spaces occupied by charging piles is: ； In the above formula, It represents the error of the predicted value of the occupancy of charging pile parking spaces when the radius of the statistical circle is r, N represents the total number of samples, It represents the predicted value of the charging pile parking space occupancy of the nth sample at the mth moment and when the radius of the statistical circle is r. is the predicted value of the occupancy of charging pile parking spaces at the mth moment; S37, selecting the statistical circle radius corresponding to the minimum error of the predicted value of the charging pile parking space occupancy, and predicting the charging pile parking space occupancy at any time of the day based on the neural network model of the predicted value of the charging pile parking space occupancy under this statistical circle radius; S4. Obtain parking request information of the fuel vehicle user who is currently waiting to enter the parking lot, and determine whether it is necessary to occupy the charging pile parking space for parking; If yes, proceed to step S5; If not, the current fuel vehicle user is allowed to enter, and then the process ends; S5. Calculate the current vehicle revenue when the current fuel vehicle user is allowed to enter and the charging vehicle revenue when the current fuel vehicle user is not allowed to enter respectively; S6. Determine whether the current vehicle benefit is greater than the charging vehicle benefit; If yes, then the current fuel vehicle user is allowed to enter, and then the process ends; If not, the current fuel vehicle user is prevented from entering and the process ends.

2. The management method according to claim 1, characterized in that: In step S2, the following steps are specifically included: S21, obtaining the vehicle license plate number, parking time and parking duration of the vehicle located in the charging pile parking space in the parking lot, and collecting historical vehicle information of the parking duration of the vehicle located in the charging pile parking space at any time; S22, calculating a first probability of a vehicle parking for any length of time at any time according to historical vehicle information; The calculation formula for the first probability is: ； In the above formula, Indicates that at the i-th moment, the parking time of the vehicle in the charging pile parking space is The first probability when Indicates that at the i-th moment, the parking time of the vehicle in the charging pile parking space is The number of vehicles, represents the total number of vehicles located in the charging pile parking space at the i-th moment; S23, calculating a second probability of the parking time of a vehicle corresponding to any license plate number at any time according to historical vehicle information; The calculation formula for the second probability is: ； In the above formula, Any license plate number The parking time of the corresponding vehicle at the i-th moment is The second probability, Indicates that at the i-th moment, the license plate number is The parking time of vehicles is The number of times Indicates that at the i-th moment, the license plate number is The total number of stops of vehicles; S24, predicting the departure time of the vehicle at the current charging pile parking space according to the first probability and the second probability of the corresponding vehicle; S25. Calculate the number of vehicles leaving the charging pile parking space at any time relative to the number of vehicles leaving the charging pile parking space at the current time according to the vehicle leaving time. The calculation formula for the leaving amount is: ； In the above formula, Indicates the number of vehicles on the charging pile parking space at any time relative to the current time When the number of vehicles leaving the charging pile parking space is Indicates the current moment, i.e. At this moment, the vehicle in the charging pile parking space Indicates the departure time is The moment and The number of vehicles between moments.

3. The management method according to claim 2, characterized in that: In step S24, the following steps are specifically included: S241, determining in sequence whether the license plate number of the vehicle currently parked in the charging pile parking space already exists in the historical vehicle information at the current moment; If yes, proceed to step S242; If not, proceed to step S243; S242, selecting the parking time corresponding to the second probability with the largest value, and calculating the departure time of the vehicle according to the time when the vehicle enters the parking lot, and then proceeding to step S25; S243: Select the parking time corresponding to the first probability with the largest value, and calculate the departure time of the vehicle according to the time when the vehicle enters the parking lot.

4. The management method according to claim 1, characterized in that: In step S32, the following steps are specifically included: S321, obtaining several roads within the initial statistical circle that affect the parking spaces for charging piles in the parking lot, as well as the average speed and weather of vehicles on the roads; S322, respectively setting a signal transmitting device and a signal receiving device for collecting CSI signals on both sides of the motor vehicle lane of the road; S323, calculating the travel time range between the vehicle passing the signal transmitting device and the signal receiving device according to the driving speed and vehicle length of the vehicle on the road; S324, setting an amplitude threshold, identifying a peak in the CSI signal that is greater than the amplitude threshold, and extracting a CSI signal corresponding to a travel time range with the peak as the center to obtain a primary signal; S325, using a low-pass filter to reduce noise on the primary signal to obtain a noise-reduced signal; S326, performing dimensionality reduction processing on the noise reduction signal to obtain a dimensionality reduction signal; S327, performing short-time inverse Fourier transform on the dimension-reduced signal to obtain a frequency spectrum; S328, identifying the spectrum graph by a convolutional neural network model to obtain the number of vehicles in the spectrum graph; S329, calculating the traffic volume on the road according to the number of vehicles on the road within a period of time; The formula for calculating traffic volume is: ； In the above formula, It means that when the radius of the statistical circle is r, at the time mt Traffic volume on the road, Indicated in The number of vehicles in the time period corresponding to the time, Indicated in The length of the time period corresponding to the moment.

5. The management method according to claim 4, characterized in that: In step S323, the following steps are specifically included: S3231, obtaining the maximum speed, minimum speed, maximum vehicle length and minimum vehicle length of vehicles on the road to obtain road data; S3232. Calculate the travel time range of vehicles passing the signal transmitting device and the signal receiving device on the road according to the road data; the expression of the travel time range is: ,in, , , in the above formula, represents the minimum travel time, represents the maximum travel time, and Respectively represent the maximum and minimum lengths of vehicles on the road. and They represent the maximum and minimum speeds of vehicles on the road respectively.

6. The management method according to claim 4, characterized in that: In step S326, the following steps are specifically included: S3261. Construct a sample matrix according to the noise reduction signal. The expression of the sample matrix is: ； In the above formula, represents the sample matrix, represents the CSI value at the jth time point in the i-th denoised signal, and the sample matrix has f rows and F columns; S3262. Construct a column vector of the sample matrix; the expression of the column vector is: ； in, represents the column vector of the jth column of the sample matrix, is the average value of all data in the jth column of the sample matrix, Represents the data in the fth row and jth column of the sample matrix; S3263. Calculate the covariance matrix of the sample matrix; the expression of the covariance matrix is: ； in, , i, j = 1, 2, ..., f, is a column vector and The covariance of S3264, performing singular value decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors, and constructing an eigenvector matrix using the eigenvectors corresponding to the eigenvalues ​​with the largest values; the expression of the eigenvector matrix is: ,in, is the eigenvector matrix, represents the hth eigenvector in the eigenvector matrix; S3265. Obtain a characteristic signal according to the characteristic vector matrix and the sample matrix; the expression of the characteristic signal is: ； In the above formula, is the characteristic signal, is the eigenvector matrix, is the sample matrix; S3266. Calculate the similarity between each signal waveform in the characteristic signal and the noise reduction signal, and mark the signal waveform in the characteristic signal with the highest similarity as the dimension reduction signal; the calculation formula of the similarity is: ； In the above formula, Indicates similarity, and Respectively represent any signal waveform and noise reduction signal in the characteristic signal, Indicates the maximum value of the x-axis in the feature signal and the noise reduction signal.

7. The management method according to claim 1, characterized in that: In step S5, the following steps are specifically included: S51, predicting the departure time of the current fuel vehicle user; S52, obtaining a predicted value of the number of parking spaces occupied by charging piles at any time before the departure time; S53, determining in sequence whether the predicted value of the charging pile parking space occupancy at each moment is greater than the maximum number of charging pile parking spaces; If yes, proceed to step S54; If not, the current fuel vehicle user is allowed to enter, and then the process ends; S54, obtaining the time when the predicted value of the charging pile parking space occupancy is equal to the maximum number of charging pile parking spaces; S55. Calculate the current vehicle revenue from allowing the current fuel vehicle user to enter until the current vehicle leaves; the calculation formula for the current vehicle revenue is: ； In the above formula, Indicates the license plate number is The current vehicle revenue of vehicles entering the charging pile parking space, Indicates the parking unit price of the charging pile parking space, Indicates the license plate number is The predicted value of the parking time of the vehicle; S56. Calculate the benefit of the charging vehicle when the current fuel vehicle user is not allowed to enter and the current vehicle leaves; the calculation formula for the charging vehicle benefit is: ； represents the revenue of charging vehicles, It represents the sum of the parking price and charging price of the charging pile parking space. Indicates the license plate number is The predicted parking time of the vehicle.

8. The management method according to claim 4, characterized in that: In step S35, the expression of the neural network model of the predicted value of the charging pile parking space occupancy is: ； In the above formula, It represents the predicted value of the occupancy of charging pile parking spaces at time m and when the radius of the statistical circle is r in the neural network model. represents the bth neuron in the hidden layer of the neural network model, and the total number is B. represents the activation function of the b-th neuron, represents the weight matrix of the b-th neuron, represents the bias of the b-th neuron, It represents the number of vehicles leaving the charging pile at time m relative to the current time. Indicates the total bias.

9. A management system for implementing the charging pile parking space management method according to any one of claims 1 to 8, characterized in that: It includes a processor and a memory, the memory is used to store a computer program, and when the computer program is executed by the processor, the charging pile parking space management method according to any one of claims 1 to 8 is implemented.

Images