Deep learning-based taxi travel demand prediction method and system

Abstract

The invention discloses a taxi travel demand prediction method and system based on deep learning, and the method comprises the steps: decomposing the taxi demand number data of a high-speed rail station through employing a VMD technology, and obtaining a modal component and a variable; a double-layer LSTM prediction model is constructed; searching an optimal hyper-parameter in the LSTM neural network model by adopting an SSA algorithm, and optimizing the double-layer LSTM prediction model to obtain an SSA-LSTM model; and taking the modal components and variables processed by the VMD technology as input data, inputting the input data into an SSA-LSTM model, outputting predicted values corresponding to the modal components through a full connection layer, and superposing the predicted values corresponding to the modal components to obtain a final prediction result. According to the method, the sensitivity of the demand change of the number of taxies in the high-speed railway station is improved, the prediction accuracy is further improved, and the contradiction caused by mismatching between the taxi resource supply of the high-speed railway station and the ever-increasing passenger demand can be relieved.

Description

Background Art

[0002] As the scale of high-speed rail transportation continues to expand and the proportion of taxis in travel modes gradually increases, the contradiction between the taxi dispatching system in high-speed rail stations and the growing travel needs of passengers continues to intensify. Due to the lack of control over the real-time passenger volume, the relevant taxi management system is difficult to provide an effective dispatching plan, and taxi drivers are therefore difficult to obtain effective information. In this case, on the one hand, it may cause vehicles to "wait in vain", causing traffic congestion near the high-speed rail station, while causing unnecessary consumption of resources and pollution to the environment. On the other hand, due to insufficient vehicle supply, a large number of passengers may be stranded in the high-speed rail station, affecting the travel experience of passengers and even forming a safety hazard in the high-speed rail station. Therefore, accurately predicting the demand for taxis in high-speed rail stations can achieve the purpose of improving the dispatching effect of the taxi dispatching system, improving the traffic environment around the high-speed rail station, creating a harmonious and safe atmosphere in the high-speed rail station, and improving the passenger experience. It is an important issue that needs to be urgently solved in the operation of high-speed rail stations.

[0003] The problem of taxi demand prediction is one of the traffic flow prediction problems, which is essentially a prediction problem for time series data. The common prediction method in the early days was based on statistical models. This method is designed to infer the relationship between variables, so the accuracy of the prediction results is poor. With the development of machine learning, a variety of machine learning models have been applied to short-term traffic prediction problems. When faced with noisy data, the prediction results of traditional machine learning methods are generally poor. In addition, with the rise of deep learning, neural network models have also been applied to this field, but most of the existing models are built in common urban public areas, such as business circles and office-intensive areas. The demand for taxis in these places has a strong temporal regularity. However, in addition to considering the time series changes within a day, the demand for taxis at high-speed rail stations also needs to consider complex factors such as arrival and departure times, real-time weather conditions, and whether it is a holiday. Therefore, the existing short-term prediction model cannot meet the characteristics of the taxi demand model of high-speed rail stations. Therefore, if the existing model is used directly to predict the taxi demand at high-speed rail stations, the prediction results will be inaccurate or even have a large deviation. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a taxi travel demand prediction method and system based on deep learning in view of the deficiencies in the above-mentioned prior art, so as to solve the technical problem of insufficient learning of taxi demand characteristics at high-speed railway stations.

[0005] The present invention adopts the following technical solutions: A taxi travel demand prediction method based on deep learning includes the following steps: The VMD technology is used to decompose the taxi demand data at the high-speed railway station to obtain the modal components and variables; Construct a two-layer LSTM prediction model; The SSA algorithm is used to find the optimal hyperparameters in the LSTM neural network model, and the double-layer LSTM prediction model is optimized to obtain the SSA-LSTM model; The modal components and variables processed by VMD technology are used as input data and input into the SSA-LSTM model. The predicted values ​​corresponding to each modal component are output through the fully connected layer. The predicted values ​​corresponding to each modal component are superimposed to obtain the final prediction result.

[0006] Preferably, the data on the number of taxis required at high-speed railway stations is collected as follows: Collect the historical number of taxis and their actual number of passengers, the number of new queued passengers and the number of people currently queuing in the time interval, the number of new queued vehicles and the number of vehicles currently queuing in the time interval, and the weather and holiday conditions, calculate the collected data, and fit the actual demand L for taxis at the railway station; The actual demand L is rounded up; Process the collected weather and holiday text content into data; Constructing a data set based on the collected historical information , dataset Each data in ( ), and They represent the weather and taxi demand at the current time period on the current date. Indicates whether the current date is a holiday, and divides the data set into 8:2 Divide into training set With the test set .

[0007] Preferably, abnormal data in the actual demand L is captured and deleted, and the average value of the data in the previous and next time periods is used as supplementary data.

[0008] Preferably, VMD technology is used to decompose the taxi demand data at the high-speed railway station as follows: Taxi demand data Decompose into The intrinsic mode function components of the finite bandwidth ,right The center frequency corresponding to the component , unit pulse function Initialize and determine the hyperparameter values ​​of the algorithm, including the number of modes, the initial center constraint strength, and the gradient descent learning rate of the dual term; Each modal component is transformed using the Hilbert transform to obtain the analytical function of each mode, the single-sided spectrum of each modal component is obtained, and the adjustment term is added to move the spectrum of the modal component to the corresponding baseband. With quadratic penalty factor Reconstruct the original problem into an unconstrained problem; Based on Fourier isometric transform, the multiplication operator alternating direction method is used to , , renew, express The Wiener filter, The corresponding Fourier transform is , , ; Iterate until .

[0009] Preferably, the Lagrange multiplication operator and penalty coefficient are introduced to transform the original problem into an unconstrained problem, and the augmented Lagrange is obtained as follows:

[0010] in, is the unit pulse function, represents the partial derivative function of time t, * represents convolution, Indicates The center frequency corresponding to each component is Is an imaginary unit.

[0011] Preferably, the termination condition for:

[0012] in, For the The first mode The value after iterations, For the The first mode The value after iterations.

[0013] Preferably, the double-layer LSTM prediction model is specifically:

[0014]

[0015]

[0016]

[0017]

[0018] in, It's the Gate of Forgetfulness. It is an external input between 0 and 1 controlled by the sigmoid activation function. is the sigmoid activation function, is the current input vector, is the current hidden layer vector, , , are the bias, input weight, and loop weight of the forget gate, respectively. is the memory unit of this unit, is an initial output obtained by the activation function sigmoid, is the hidden layer of the previous unit.

[0019] Preferably, the SSA algorithm is used to find the optimal hyperparameters in the LSTM neural network model, and the two-layer LSTM prediction model is optimized as follows: Initialize the sparrow population; take the degree of fit between the model's initial prediction results and the true value as the initial value of fitness; According to the fitness value, the sparrows are divided into discoverers and followers, and the discoverer position is updated. Safety value When , it means the current environment is safe and the sparrow can continue to search the current area. When it is on, it means the sparrow has detected dangerous information and issued an alarm. All sparrows in the current area will be moved to other safe areas. The identity of the follower and the discoverer changes, and the follower position is updated. When , it indicates that the adaptability of the ith follower is poor and needs to move to another area; Assume that the number of sparrows aware of danger in the population is 10% to 20%, and their initial positions are randomly generated. The warning sparrows discover the danger. When the fitness of the i-th sparrow The global best fitness under the current situation When the sparrow is at the edge of the group, When , it indicates that the sparrow is aware of the danger and moves with K as the movement parameter to get closer to other sparrows; When the number of iterations is exhausted, the parameters stop updating and the optimal hyperparameters found by the SSA algorithm are output, namely the optimal learning rate, number of iterations, number of nodes in the first hidden layer, and number of nodes in the second hidden layer of the LSTM network model.

[0020] Preferably, the specific process after the early warning sparrow discovers danger is:

[0021] in, for, is the global optimal position, is the global optimal fitness in the current situation, is a random number that follows a normal distribution, is a random number representing the step size control, is a constant.

[0022] In a second aspect, an embodiment of the present invention provides a taxi travel demand prediction system based on deep learning, comprising: Decomposition module, which uses VMD technology to decompose the taxi demand data at the high-speed railway station to obtain modal components and variables; Building modules and constructing a two-layer LSTM prediction model; The training module uses the SSA algorithm to find the optimal hyperparameters in the LSTM neural network model, trains the double-layer LSTM prediction model, and obtains the SSA-LSTM model; The prediction module uses the modal components and variables processed by VMD technology as input data and inputs them into the SSA-LSTM model. The prediction values ​​corresponding to each modal component are output through the fully connected layer, and the prediction values ​​corresponding to each modal component are superimposed to obtain the final prediction result.

[0023] In a third aspect, a computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned deep learning-based taxi travel demand prediction method when executing the computer program.

[0024] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned taxi travel demand prediction method based on deep learning.

[0025] Compared with the prior art, the present invention has at least the following beneficial effects: A taxi travel demand forecasting method based on deep learning makes full use of the combined advantages of time series decomposition, deep learning and optimization algorithms. First, the VMD technology is used to decompose the taxi demand data at the high-speed railway station to obtain modal components and variables. The complex original data is decomposed to obtain modal components with different frequencies and characteristics, which reduces the complexity of the data and helps to improve the accuracy of the forecasting model. Secondly, a two-layer LSTM forecasting model is constructed to better capture the multi-level information in the complex time series signal and improve the forecasting effect. The SSA algorithm is used to find the optimal hyperparameters in the LSTM neural network model, and the two-layer LSTM forecasting model is optimized to avoid the limitations of manual parameter adjustment and improve the forecasting performance of the model.

[0026] Furthermore, it is necessary to collect the historical number of taxis and their actual number of passengers, the number of new queued passengers and the number of people queuing in the time interval, and the number of new queued vehicles and the number of vehicles queuing in the time interval. The purpose is to collect dynamic supply and demand information around the high-speed railway station, capture the relationship between taxi supply (number of taxis, queued vehicles) and actual demand (number of passengers, queued passengers), so as to better support the subsequent prediction model and improve the model's ability to perceive demand fluctuations. It is also necessary to collect weather conditions and holiday conditions, and calculate the collected data to fit the actual demand for taxis at the railway station, because weather and holidays have a significant impact on passenger travel behavior. Adding these factors helps the model identify external driving factors of demand changes. Finally, it is necessary to digitize the collected weather and holiday text content, and convert non-numerical text information (such as weather descriptions and holiday signs) into a numerical form acceptable to the model, so that the deep learning model can better understand the impact of weather and holidays on taxi demand, thereby improving the accuracy of the prediction.

[0027] Furthermore, the purpose of using VMD technology to decompose the taxi demand data at high-speed railway stations is to decompose complex time series data into several modal components with different frequencies and characteristics, so as to better identify the features in the data and improve the accuracy of subsequent prediction models. VMD is a variational method that can decompose signals into multiple essential mode functions (IMFs). These modal functions have limited bandwidth and clear physical meanings in the time domain. By initializing the center frequency and unit pulse function and setting appropriate hyperparameters, VMD can adaptively separate components of different frequencies, so that the different features of complex signals can be more clearly displayed. This method can effectively extract different frequency components in the signal, making it easier for subsequent deep learning models to capture the regularity and characteristics therein and improve prediction accuracy.

[0028] Furthermore, the purpose of setting up the double-layer LSTM prediction model is to capture the complex time series features in the taxi demand data of high-speed railway stations through the long short-term memory (LSTM) network, especially the long-term dependency, so as to improve the accuracy of demand forecasting. By introducing the gating mechanism, the model overcomes the shortcomings of traditional RNN in dealing with long-term dependency problems and can effectively model complex time series relationships. In addition, the double-layer LSTM can realize that the first layer LSTM captures local time series features in the data, such as daily fluctuations and random changes; the second layer LSTM further abstracts on this basis and extracts higher-level features, such as weekly and monthly cycle trends and long-term change patterns, so as to provide more robust and accurate prediction results.

[0029] Furthermore, the method uses SSA (Sparrow Search Algorithm) to optimize the hyperparameters in the LSTM neural network model. Through the intelligent optimization algorithm, the optimal model hyperparameter combination is automatically searched and found, thereby improving the prediction performance of the double-layer LSTM model. The performance of the LSTM model is highly dependent on multiple key hyperparameters, such as learning rate, number of iterations, number of hidden layer nodes, etc. However, the manual adjustment of these hyperparameters is time-consuming and difficult to find the global optimal solution. Therefore, the method uses the SSA algorithm to simplify the hyperparameter optimization process. SSA is an intelligent optimization algorithm that simulates the foraging behavior of sparrows. It uses population intelligence to find the global optimal solution in a complex search space through cooperation and competition between individuals. In the algorithm, the population is divided into "discoverers" and "followers". "Discoverers" are responsible for exploring new search areas, while "followers" follow or make local adjustments based on the position of "discoverers". The algorithm also avoids falling into local optimality by dynamically updating the positions of discoverers and followers, as well as simulating early warning mechanisms, and finally achieves the global optimal hyperparameter search. Compared with traditional parameter adjustment methods, SSA's global search capability and dynamic adjustment mechanism enable it to obtain better hyperparameter combinations at less computational cost, thereby improving the performance of the two-layer LSTM model and ultimately improving the accuracy and robustness of taxi demand forecasting.

[0030] It can be understood that the beneficial effects of the second aspect mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0031] In summary, the present invention improves the sensitivity to changes in demand for the number of taxis at high-speed railway stations, further improves the accuracy of prediction, and helps to alleviate the contradiction caused by the mismatch between the supply of taxi resources at railway stations and the growing passenger demand.

[0032] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is the internal structure diagram of the LSTM loop unit; Figure 2 This is the flow chart of VMD-SSA-LSTM prediction model; Figure 3 A schematic diagram of a computer device provided by an embodiment of the present invention; Figure 4 The present invention is a block diagram of an electronic device provided according to an embodiment. DETAILED DESCRIPTION

[0034] 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.

[0035] In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0036] 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.

[0037] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects are in an "or" relationship.

[0038] It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0039] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0040] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0041] The present invention provides a taxi travel demand prediction method based on deep learning. Based on the VMD-SSA-LSTM taxi demand prediction model that takes into account real-time weather and holiday characteristics, the taxi demand per hour at the high-speed railway station the next day is predicted, so as to improve the sensitivity to changes in the number of taxis at the high-speed railway station, make timely adjustments to the taxi demand forecast, better adapt to the characteristic differences of the taxi demand at the high-speed railway station, and improve the prediction accuracy.

[0042] See also Figure 2 The present invention provides a taxi travel demand prediction method based on deep learning, comprising the following steps: S1. Use sensors and manual data collection at high-speed railway stations to collect the historical number of taxis (PV) and their actual number of passengers (APC), the number of new queued passengers (AP) and the number of people in the queue (LP) during the time interval, the number of new queued vehicles (AV) and the number of vehicles in the queue (LV) during the time interval, and the weather conditions (weather) and holidays (is_holiday) to construct a data set; S101, calculating the collected data to fit the actual demand for taxis at the railway station. The specific calculation process is as follows: (1) S102. During the calculation process, the L value needs to be rounded up. Due to equipment errors and human errors, there are abnormal data in the data. The abnormal data needs to be captured and deleted to avoid affecting the subsequent experimental results, and the average value of the data before and after the period is used as supplementary data. The specific process is as follows: (2) S103, the collected weather conditions and holiday conditions text content are processed into data, and the specific mapping is shown in the following table:

[0043]

[0044] S104. Construct a data set based on the collected historical information , dataset Each data in ( ), where and They represent the weather and taxi demand at the current time period on the current date. Indicates whether the current date is a holiday, and divides the data set into 8:2 Divide into training set With the test set .

[0045] S2, using VMD technology to decompose the taxi demand data at high-speed railway stations; S201, input signal taxi demand data is , decompose it into The intrinsic mode function components of the finite bandwidth ,right The center frequency corresponding to the component , unit pulse function Initialize and determine the hyperparameter values ​​of the algorithm, including the number of modes (K), the initial center constraint strength (alpha), and the gradient descent learning rate of the dual term (tau); S202, transform each modal component using Hilbert transform to obtain the analytical function of each mode, thereby obtaining the single-sided spectrum of each modal component, and adding an adjustment item to move the spectrum of the modal component to the corresponding baseband. The specific calculation process is: (3) make is the unit pulse function, represents the partial derivative function of time t, * represents convolution, Indicates The center frequency corresponding to each component.

[0046] Using Lagrangian operator With quadratic penalty factor Reconstruct the above formula and transform the original problem into an unconstrained problem: (4) S203, based on Fourier isometric transformation, using the multiplication operator alternating direction method , , renew, express The Wiener filter, The corresponding Fourier transform is , , , the update process is: (5) (6) (7) S204, the iteration termination condition is , the specific calculation process is: (10) Otherwise, continue to execute step S203.

[0047] S3, construct a two-layer LSTM prediction model to predict the number of taxi demands; The LSTM network structure is as follows: At time t, LSTM receives the hidden layer from the previous unit. , the memory unit of the previous unit and the input layer There are three parts, and the output at time t is two parts: the hidden layer of this unit and the memory unit of this unit , and then pass it to the next unit.

[0048] (11) (12) (13) (14) (15) in, is the sigmoid activation function, is the current input vector, It's the Gate of Forgetfulness. is the current hidden layer vector, , , They are the bias, input weight and recurrent weight of the forget gate respectively. It is an external input between 0 and 1 controlled by the sigmoid activation function.

[0049] S4, use the SSA algorithm to find the optimal hyperparameters in the LSTM neural network model; The SSA algorithm needs to determine the optimal hyperparameters, including learning rate, number of iterations, number of nodes in the first hidden layer, and number of nodes in the second hidden layer. The data is input into the SSA-LSTM model, and the SSA algorithm will use the degree of fit of the prediction results as an evaluation indicator to continuously adjust the parameters.

[0050] S401. Use the SSA algorithm to optimize the LSTM model. First, initialize the sparrow population, including the number of sparrows as n, the proportion of discoverers in the population, and the alarm threshold for warning sparrows as , The maximum number of iterations in the whole process is t. The Adam algorithm is used and the parameter range is defined to ensure that the parameters in the optimization process do not exceed the specified range. S402: The degree of fit between the initial prediction result of the model and the true value is taken as the initial value of the fitness, which is specifically calculated as follows: (16) in, represents the predicted value within the prediction time interval, Represents the true value within the prediction time interval; S403, dividing the sparrows into finders and followers according to the fitness values, and updating the finder positions. The specific process is as follows: (17) Where t represents the number of iterations, is a random number, Indicates a danger signal value, represents the safety value, Q is a random number that follows a normal distribution, and L is a 1×d matrix whose elements are all 1. When , it means the current environment is safe and the sparrow can continue to search the current area. When it is on, it means the sparrow has detected dangerous information and issued an alarm. All sparrows in the current area will be moved to other safe areas. S404: The identities of the follower and the discoverer will be transformed, and the specific process of the follower position update is as follows: (19) (18) in, represents the optimal finder position, Represents the worst global position, A is an element randomly 1 is a 1×d size matrix.

[0051] when When , it indicates that the adaptability of the ith follower is poor and needs to move to another area; S405. Assuming that the number of sparrows aware of danger in the population is 10% to 20% and their initial positions are randomly generated, the specific process after the warning sparrows discover the danger is as follows: (20) is the global optimal position, is the fitness of the i-th sparrow, is the global optimal fitness in the current situation, is the global optimal fitness in the current situation, is a random number that follows a normal distribution, is a random number representing the step size control, is a constant.

[0052] when When When , it indicates that the sparrow is aware of the danger and moves with K as the movement parameter to get closer to other sparrows; S406. When the number of iterations is exhausted, the parameters stop updating and the optimal hyperparameters found by the SSA algorithm are output, namely, the optimal learning rate, number of iterations, number of nodes in the first hidden layer, and number of nodes in the second hidden layer of the LSTM network model.

[0053] S5, the modal components and variables processed by VMD are used as input data and input into the SSA-LSTM model, and the prediction values ​​corresponding to each modal component are output through the fully connected layer; S6. Superimpose the predicted values ​​of each modal component predicted by the SSA-LSTM model to obtain the final prediction result, and select a variety of evaluation indicators to compare the model effects.

[0054] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be collectively referred to herein as "circuits", "modules" or "platforms".

[0055] In another embodiment of the present invention, a taxi travel demand prediction system based on deep learning is provided, which can be used to implement the above-mentioned taxi travel demand prediction method based on deep learning. Specifically, the taxi travel demand prediction system based on deep learning includes a decomposition module, a construction module, a training module and a prediction module.

[0056] Among them, the decomposition module uses VMD technology to decompose the taxi demand quantity data of high-speed railway stations to obtain modal components and variables; Building modules and constructing a two-layer LSTM prediction model; The training module uses the SSA algorithm to find the optimal hyperparameters in the LSTM neural network model, trains the double-layer LSTM prediction model, and obtains the SSA-LSTM model; The prediction module uses the modal components and variables processed by VMD technology as input data and inputs them into the SSA-LSTM model. The prediction values ​​corresponding to each modal component are output through the fully connected layer, and the prediction values ​​corresponding to each modal component are superimposed to obtain the final prediction result.

[0057] In another embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the taxi travel demand prediction method based on deep learning, including: A two-layer LSTM prediction model is constructed. The SSA algorithm is used to find the optimal hyperparameters in the LSTM neural network model, and the two-layer LSTM prediction model is trained to obtain the SSA-LSTM model. The modal components and variables processed by VMD technology are used as input data and input into the SSA-LSTM model. The prediction values ​​corresponding to each modal component are output through the fully connected layer, and the prediction values ​​corresponding to each modal component are superimposed to obtain the final prediction result.

[0058] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device, and can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that more specific examples (non-exhaustive list) of the computer-readable storage medium here include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0059] Computer readable storage media also include data signals propagated in baseband or as part of a carrier wave, which carry readable program codes. Such propagated data signals can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0060] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0061] The processor may load and execute one or more instructions stored in a computer-readable storage medium to implement the corresponding steps of the taxi travel demand prediction method based on deep learning in the above embodiment; one or more instructions in the computer-readable storage medium are loaded and executed by the processor as follows: A two-layer LSTM prediction model is constructed. The SSA algorithm is used to find the optimal hyperparameters in the LSTM neural network model, and the two-layer LSTM prediction model is trained to obtain the SSA-LSTM model. The modal components and variables processed by VMD technology are used as input data and input into the SSA-LSTM model. The prediction values ​​corresponding to each modal component are output through the fully connected layer, and the prediction values ​​corresponding to each modal component are superimposed to obtain the final prediction result.

[0062] See also Figure 3 , the terminal device is a computer device. The computer device 60 of this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, the method for predicting taxi travel demand based on deep learning in the embodiment is implemented. To avoid repetition, it is not described one by one here. Alternatively, when the computer program 63 is executed by the processor 61, the functions of each model / unit in the system for predicting taxi travel demand based on deep learning in the embodiment are implemented. To avoid repetition, it is not described one by one here.

[0063] The computer device 60 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will appreciate that Figure 3 This is only an example of the computer device 60 and does not constitute a limitation of the computer device 60. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0064] The processor 61 may be a central processing unit (CPU), or other general-purpose processors, central processing units, graphics processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, data processing logic based on quantum computing, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0065] The memory 62 may be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 60.

[0066] Furthermore, the memory 62 may include both an internal storage unit of the computer device 60 and an external storage device. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0067] Any reference to a memory, database or other medium used in the embodiments provided in this application may include at least one of a non-volatile and a volatile memory. Non-volatile memory may include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory, an optical memory, a high-density embedded non-volatile memory, a resistive random access memory (ReRAM), a magnetic random access memory (MRAM), a ferroelectric random access memory (FRAM), a phase change memory (PCM), a graphene memory, etc. Volatile memory may include a random access memory (RAM) or an external cache memory, etc. As an illustration and not limitation, RAM may be in various forms, such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), etc.

[0068] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0069] See also Figure 4 The terminal device 600 is an electronic device, and the electronic device is in the form of a general computing device. The components of the electronic device may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0070] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present invention described in the above method section of this specification. For example, the processing unit 610 can perform the following steps: Figure 1 Follow the steps shown in .

[0071] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0072] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0073] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0074] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0075] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can usually be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0076] In the taxi demand simulation data, the model can make demand forecasts based on weather, holidays and other characteristics. The goodness of fit of the final test results on this simulation data set can reach above 0.99, and the MAE is 0.92, which shows an overall improvement compared to the traditional deep learning model; and the prediction error percentage of this model fluctuates less. The model that integrates multiple features has the narrowest prediction interval and the narrowest error density distribution. Compared with other traditional deep learning models under the same conditions, it can improve the prediction accuracy and is more suitable for the prediction of the number of taxi demands at high-speed railway stations.

[0077] In summary, the present invention is a method and system for predicting taxi travel demand based on deep learning, which realizes high-precision prediction of taxi travel demand by combining VMD, double-layer LSTM and SSA optimization technology; VMD decomposition improves the feature extraction ability of complex time series data, double-layer LSTM captures short-term fluctuations and long-term trends, and SSA automatically optimizes hyperparameters to ensure the optimal performance of the model. On the whole, the method of the present invention shows strong prediction accuracy and robustness in complex and changeable environments, and provides effective support for actual taxi dispatching. Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit, and the above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other and are not used to limit the protection scope of this application. The specific working process of the units and modules in the above system can refer to the corresponding process in the above method embodiment, which will not be repeated here.

[0078] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0079] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. 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.

[0080] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0081] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0082] 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.

[0083] If the integrated module / 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 present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0084] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0087] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A taxi travel demand prediction method based on deep learning, characterized in that: The following steps are involved: The VMD technology is used to decompose the taxi demand data at the high-speed railway station to obtain the modal components and variables; Construct a two-layer LSTM prediction model; The SSA algorithm is used to find the optimal hyperparameters in the LSTM neural network model, and the double-layer LSTM prediction model is optimized to obtain the SSA-LSTM model; The modal components and variables processed by VMD technology are used as input data and input into the SSA-LSTM model. The predicted values ​​corresponding to each modal component are output through the fully connected layer. The predicted values ​​corresponding to each modal component are superimposed to obtain the final prediction result.

2. The taxi travel demand prediction method based on deep learning according to claim 1 is characterized in that: The specific data collected on the number of taxis required at high-speed railway stations are as follows: Collect the historical number of taxis and their actual number of passengers, the number of new queued passengers and the number of people currently queuing in the time interval, the number of new queued vehicles and the number of vehicles currently queuing in the time interval, and the weather and holiday conditions, calculate the collected data, and fit the actual demand L for taxis at the railway station; The actual demand L is rounded up; Process the collected weather and holiday text content into data; Constructing a data set based on the collected historical information , dataset Each data in ( ), and They represent the weather and taxi demand at the current time period on the current date. Indicates whether the current date is a holiday, and divides the data set into 8:2 Divide into training set With the test set .

3. The taxi travel demand prediction method based on deep learning according to claim 2 is characterized in that: The abnormal data in the actual demand L is captured and deleted, and the average value of the data before and after the time period is used as supplementary data.

4. The taxi travel demand prediction method based on deep learning according to claim 1 is characterized in that: The VMD technology is used to decompose the demand data of taxis at high-speed railway stations into the following details: Taxi demand data Decompose into The intrinsic mode function components of the finite bandwidth ,right The center frequency corresponding to the component , unit pulse function Initialize and determine the hyperparameter values ​​of the algorithm, including the number of modes, the initial center constraint strength, and the gradient descent learning rate of the dual term; Each modal component is transformed using the Hilbert transform to obtain the analytical function of each mode, the single-sided spectrum of each modal component is obtained, and the adjustment term is added to move the spectrum of the modal component to the corresponding baseband. With quadratic penalty factor Reconstruct the original problem into an unconstrained problem; Based on Fourier isometric transform, the multiplication operator alternating direction method is used to , , renew, express The Wiener filter, The corresponding Fourier transform is , , ; Iterate until .

5. The taxi travel demand prediction method based on deep learning according to claim 4 is characterized in that: The introduction of Lagrange multiplication operator and penalty coefficient transforms the original problem into an unconstrained problem, and the augmented Lagrange is obtained as follows: in, is the unit pulse function, represents the partial derivative function of time t, * represents convolution, Indicates The center frequency corresponding to each component is Is an imaginary unit.

6. The taxi travel demand prediction method based on deep learning according to claim 4 is characterized in that: Termination Condition for: in, For the The first mode The value after iterations, For the The first mode The value after iterations.

7. The taxi travel demand prediction method based on deep learning according to claim 1 is characterized in that: The specific double-layer LSTM prediction model is: in, It's the Gate of Forgetfulness. It is an external input between 0 and 1 controlled by the sigmoid activation function. is the sigmoid activation function, is the current input vector, is the current hidden layer vector, , , are the bias, input weight, and loop weight of the forget gate, respectively. is the memory unit of this unit, is an initial output obtained by the activation function sigmoid, is the hidden layer of the previous unit.

8. The taxi travel demand prediction method based on deep learning according to claim 1 is characterized in that: The SSA algorithm is used to find the optimal hyperparameters in the LSTM neural network model, and the two-layer LSTM prediction model is optimized as follows: Initialize the sparrow population; take the degree of fit between the model's initial prediction results and the true value as the initial value of fitness; According to the fitness value, the sparrows are divided into discoverers and followers, and the discoverer position is updated. Safety value When , it means the current environment is safe and the sparrow can continue to search the current area. When it is on, it means the sparrow has detected dangerous information and issued an alarm. All sparrows in the current area will be moved to other safe areas. The identity of the follower and the discoverer changes, and the follower position is updated. When , it indicates that the adaptability of the ith follower is poor and needs to move to another area; Assume that the number of sparrows aware of danger in the population is 10% to 20%, and their initial positions are randomly generated. The warning sparrows discover the danger. When the fitness of the i-th sparrow The global best fitness under the current situation When the sparrow is at the edge of the group, When , it indicates that the sparrow is aware of the danger and moves with K as the movement parameter to get closer to other sparrows; When the number of iterations is exhausted, the parameters stop updating and the optimal hyperparameters found by the SSA algorithm are output, namely the optimal learning rate, number of iterations, number of nodes in the first hidden layer, and number of nodes in the second hidden layer of the LSTM network model.

9. The taxi travel demand prediction method based on deep learning according to claim 8 is characterized in that: The specific process after the early warning sparrow discovers danger is: in, is the global optimal position, is the global optimal fitness in the current situation, is a random number that follows a normal distribution, is a random number representing the step size control, is a constant.

10. A taxi travel demand prediction system based on deep learning, characterized in that: include: Decomposition module, which uses VMD technology to decompose the taxi demand data at the high-speed railway station to obtain modal components and variables; Building modules and constructing a two-layer LSTM prediction model; The training module uses the SSA algorithm to find the optimal hyperparameters in the LSTM neural network model, trains the double-layer LSTM prediction model, and obtains the SSA-LSTM model; The prediction module uses the modal components and variables processed by VMD technology as input data and inputs them into the SSA-LSTM model. The prediction values ​​corresponding to each modal component are output through the fully connected layer, and the prediction values ​​corresponding to each modal component are superimposed to obtain the final prediction result.

Images