Congestion prediction method, congestion prediction device and computer storage medium
By acquiring and processing the regional information around the dock and generating a fused average itinerary speed, the problem of inability to predict dock congestion in the prior art is solved, and accurate prediction of dock congestion is achieved, providing valuable data for decision-making.
Patent Information
- Application Number
- CN202411439943.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-15
- Publication Date
- 2025-02-28
AI Technical Summary
The existing technology cannot achieve advance prediction of terminal congestion, resulting in lagging congestion decisions and unable to effectively solve the problem of terminal congestion.
By obtaining the area information of the area to be predicted, a predicted average travel speed of the target section is generated, and the predicted speed is corrected using the correction value to obtain the fused average travel speed, thereby generating road congestion information.
In advance and accurate prediction of congestion conditions in the target area is achieved, relevant data is provided for congestion decisions, and the accuracy and reliability of congestion prediction are improved.
Smart Images
Figure CN119541191A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to a congestion prediction method, a congestion prediction device and a computer storage medium. Background Art The closure of terminals due to extreme weather has caused passengers and cargo to be stranded, posing severe challenges to urban management and may cause potential safety incidents in counties. During extreme weather periods, a large amount of manpower and financial resources are required for county management.
[0007] However, at present, the above-mentioned congestion problem can only be discovered and dealt with in a short period of time when the congestion occurs, but it is impossible to predict the congestion in advance, resulting in delayed congestion decision-making and failure to effectively solve the port congestion problem. Summary of the invention
[0008] In order to solve the above technical problems, the present application proposes a congestion prediction method, a congestion prediction device and a computer storage medium.
[0009] In order to solve the above technical problems, the present application proposes a congestion prediction method, which includes:
[0010] Obtain regional information of the area to be predicted;
[0011] Based on the area information, generating a predicted average travel speed of a target road section in the area to be predicted;
[0012] Generating a target road section correction value of the area to be predicted based on the predicted average travel speed of the target road section;
[0013] The predicted average travel speed of the target section is corrected by using the target section correction value to obtain a fused average travel speed of the target section;
[0014] The target road section is combined with the average travel speed to generate road congestion information of the area to be predicted.
[0015] Among them, the area information includes the current average vehicle speed of the target section of the area to be predicted, and the environmental information of the area to be predicted, and the environmental information includes: weather data, time data, checkpoint data, port capacity data, parking lot data, and / or road data.
[0016] The step of generating a predicted average travel speed of a target road section in the area to be predicted based on the area information includes:
[0017] Inputting the regional information into a speed prediction model to extract environmental impact features from the regional information;
[0018] The speed prediction model is used to predict the environmental impact characteristics to generate a predicted average travel speed of the target road section in the area to be predicted.
[0019] Wherein, the congestion prediction method further includes:
[0020] Initializing model hyperparameters of the speed prediction model;
[0021] Inputting training data into the speed prediction model to obtain a predicted value of the training data;
[0022] Obtaining training residuals based on the true values and predicted values of the training data;
[0023] Using the training residuals to train a decision tree of the speed prediction model, and obtaining node weights of leaf nodes of the decision tree;
[0024] The model parameters of the speed prediction model are updated using the node weights.
[0025] The step of generating a target road section correction value of the area to be predicted based on the predicted average travel speed of the target road section includes:
[0026] Inputting the predicted average travel speed of the target road section into a speed correction model to extract vehicle speed sequence features in the regional information;
[0027] The speed correction model is used to predict the vehicle speed sequence characteristics to generate a target road section correction value for the area to be predicted.
[0028] Wherein, the congestion prediction method further includes:
[0029] Obtain the predicted value and true value of the speed prediction model for the training sample;
[0030] Constructing a training residual sequence using the predicted values and true values of the training samples;
[0031] Determining whether the training residual sequence is stable;
[0032] If yes, performing differential processing on the training residual sequence until the training residual sequence reaches a stable state;
[0033] The speed correction model is trained using a stationary training residual sequence.
[0034] After the target road section is used to fuse the average travel speed to generate the road congestion information of the area to be predicted, the congestion prediction method further includes:
[0035] generating a number of congestion handling decision plans based on the road congestion information;
[0036] Based on each congestion handling decision plan and the area information of the area to be predicted, generating prediction area information corresponding to each congestion handling decision plan;
[0037] The predicted area information of the several congestion handling decision plans are compared to obtain the optimal congestion handling decision plan.
[0038] In order to solve the above technical problems, the present application also proposes a congestion prediction device, which includes a region acquisition module, a speed prediction module, a speed correction module, a speed fusion module, and a congestion prediction module; wherein,
[0039] The region acquisition module is used to acquire region information of the region to be predicted;
[0040] The speed prediction module is used to generate a predicted average travel speed of a target road section in the area to be predicted based on the area information;
[0041] The speed correction module is used to generate a target section correction value of the area to be predicted based on the predicted average travel speed of the target section;
[0042] The speed fusion module is used to correct the predicted average travel speed of the target section by using the target section correction value to obtain the fused average travel speed of the target section;
[0043] The congestion prediction module is used to generate road congestion information of the area to be predicted by integrating the target road section with the average travel speed.
[0044] To solve the above technical problems, the present application also proposes a congestion prediction device, which includes a memory and a processor coupled to the memory; wherein the memory is used to store program data, and the processor is used to execute the program data to implement the congestion prediction method as described above.
[0045] In order to solve the above technical problems, the present application also proposes a computer storage medium, which is used to store program data. When the program data is executed by a computer, it is used to implement the above congestion prediction method.
[0046] Compared with the prior art, the beneficial effects of the present application are as follows: the congestion prediction device obtains the regional information of the area to be predicted; based on the regional information, generates the predicted average travel speed of the target road section of the area to be predicted; based on the predicted average travel speed of the target road section, generates the target road section correction value of the area to be predicted; uses the target road section correction value to correct the predicted average travel speed of the target road section to obtain the fused average travel speed of the target road section; uses the fused average travel speed of the target road section to generate the road congestion information of the area to be predicted. Through the above congestion prediction method, the congestion situation of the target area can be predicted in advance and accurately, providing relevant data for congestion decision-making, and realizing accurate and reliable congestion prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0048] in:
[0049] Figure 1 It is a flowchart of an embodiment of a congestion prediction method provided by the present application;
[0050] Figure 2 This is a schematic diagram of the modeling process of an embodiment of a congestion prediction model for the periphery of a port and a wharf provided by the present application;
[0051] Figure 3 It is a schematic diagram of the training process of an embodiment of a speed prediction model provided by the present application;
[0052] Figure 4 It is a training process diagram of an embodiment of a speed correction model provided by the present application;
[0053] Figure 5 It is a flowchart of another embodiment of the congestion prediction method provided by the present application;
[0054] Figure 6 It is a structural schematic diagram of an embodiment of a congestion prediction device provided by the present application;
[0055] Figure 7 is a structural schematic diagram of another embodiment of the congestion prediction device provided by the present application;
[0056] Figure 8 It is a structural diagram of an embodiment of a computer storage medium provided by the present application. DETAILED DESCRIPTION
[0057] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0058] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0059] In order to solve the problems of the prior art, this application provides a congestion prediction method to alleviate the congestion problem of county passenger and cargo roll-on / roll-off terminals from multiple aspects such as traffic carrying capacity calculation, congestion prediction, congestion relief, and extreme weather detention management.
[0060] It should be noted that the following uses a county passenger and cargo roll-on / roll-off terminal as a specific embodiment of the area to be predicted. In other embodiments, the area to be predicted may also be other workspaces, including but not limited to: factories, parking lots, shopping malls, etc., which are not listed here one by one.
[0061] Please refer to Figure 1 , Figure 1 It is a flowchart of an embodiment of a congestion prediction method provided by the present application.
[0062] The congestion prediction method of the present application is applied to a congestion prediction device, wherein the congestion prediction device of the present application can be a server, a terminal device, or a system in which a server and a terminal device cooperate with each other. Accordingly, the various parts of the congestion prediction device, such as various units, subunits, modules, and submodules, can all be set in the server, can all be set in the terminal device, or can be set in the server and the terminal device respectively.
[0063] Furthermore, the above-mentioned server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules, such as software or software modules used to provide distributed servers, or it can be implemented as a single software or software module, which is not specifically limited here.
[0064] like Figure 1 As shown, the specific steps are as follows:
[0065] Step S11: Obtaining the area information of the area to be predicted.
[0066] In an embodiment of the present application, the congestion prediction device obtains regional information about the county passenger and cargo roll-on / roll-off terminal, and the regional information represents the surrounding environment that may affect the congestion of the county passenger and cargo roll-on / roll-off terminal. Compared with urban road congestion, the traffic congestion of the county passenger and cargo roll-on / roll-off terminal will be somewhat different. Specifically, the congestion of the county passenger and cargo roll-on / roll-off terminal is affected by the upstream highway exit vehicles and upstream road traffic volume on the one hand, and by the downstream port capacity and the saturation of the downstream port internal parking lot on the other hand. Therefore, launching data collection on upstream and downstream influencing factors is a prerequisite for improving the accuracy of congestion prediction. The data that need to be collected in this application mainly include weather, vehicle travel time, road traffic volume, average vehicle speed, highway exit flow, parking lot vehicle passage records, port capacity, etc.
[0067] Therefore, the regional information of the county passenger and cargo ro-ro terminals includes but is not limited to the following information:
[0068] Weather data: hourly wind speed (alternative indicator: wind force level), rainfall, temperature, visibility, and whether it is a typhoon day.
[0069] Time data: vehicle travel time.
[0070] Checkpoint data: traffic volume and average speed of the corresponding lanes monitored by the checkpoints, checkpoint monitoring data of the highway intersecting with the access highway, and checkpoint distribution.
[0071] Highway exit traffic data.
[0072] Port capacity data: ship departure time, ship load at departure: number of vehicles, including large and small vehicles (alternative indicator: number of vehicles leaving the port per hour).
[0073] Parking lot data: data on vehicles entering and leaving the parking lot, vehicle entry and exit times, and parking lot capacity.
[0074] Road data: road length, road carrying capacity, number of lanes.
[0075] After obtaining the above regional information, the congestion prediction device may also perform data preprocessing on the regional information before inputting the regional information into the congestion prediction model. The data preprocessing provided in this application includes but is not limited to:
[0076] Data cleaning: handling missing values and removing noisy data.
[0077] Data integration: entity recognition, removing redundant data, handling numerical conflicts.
[0078] Data transformation: data normalization, discretization, and stratification.
[0079] Feature construction: Use date data to construct holiday identification features, week features, and hour features; use checkpoint traffic flow data to construct the one-period lagged traffic flow features of each section; use port shift data to construct the port's hourly capacity features, etc.
[0080] Feature encoding: Encode hour features and week features.
[0081] Step S12: Based on the area information, generate a predicted average travel speed of the target road section in the area to be predicted.
[0082] In the embodiment of the present application, the congestion prediction device is pre-installed with a congestion prediction model around the port. Figure 2 , Figure 2 It is a schematic diagram of the modeling process of an embodiment of a congestion prediction model around a port and a wharf provided in the present application.
[0083] Depend on Figure 2 It can be seen that the congestion prediction model of the present application mainly includes two functions, namely, vehicle speed trend prediction and vehicle speed residual prediction. The congestion prediction model first obtains a basic predicted vehicle speed through vehicle speed trend prediction, then obtains the possible residual of the basic predicted vehicle speed through vehicle speed residual prediction, and finally corrects the basic predicted vehicle speed through the residual to obtain an accurate predicted vehicle speed for evaluating the congestion situation.
[0084] Therefore, the congestion prediction model of the present application can be divided into a speed prediction model and a speed correction model according to its functions, wherein the speed prediction model performs vehicle speed trend prediction, and the speed correction model performs vehicle speed residual prediction.
[0085] Specifically, this application proposes Figure 2The modeling idea shown is as follows: using variables such as weather, time, traffic volume in different lanes of each section, and average travel speed as input, the XGboost (eXtreme Gradient Boosting) + ARIMA (Autoregressive Integrated Moving Average Model) algorithm is used to build a fusion prediction model. The influence of various factors other than the average vehicle speed is introduced through the XGboost algorithm, and the detailed characteristics of the average vehicle speed sequence itself are captured through ARIMA. By combining internal and external methods, the congestion situation of Jingang Avenue is accurately predicted.
[0086] The specific approach is to use the XGboost model to predict the average travel speed of the target road section in the next hour, use the ARIMA model to fit the prediction residual of the XGboost model, and use the ARIMA prediction result to correct the XGboost prediction result to obtain the fused average travel speed prediction result. The road congestion situation can be directly judged based on the speed.
[0087] In other embodiments, other algorithms may also be used to achieve both the vehicle speed prediction function and the vehicle speed correction function.
[0088] In the embodiment of the present application, the congestion prediction device inputs the regional information of step S11 into the speed prediction model to extract the environmental impact characteristics in the regional information. The speed prediction model predicts the average travel speed of the target road section around the county passenger and cargo roll-on / roll-off terminal based on the environmental impact characteristics, which is used to determine the road congestion situation.
[0089] At this time, the input and output data model of the speed prediction model is as follows:
[0090]
[0091]
[0092] Please continue reading Figure 3 , Figure 3 It is a schematic diagram of the training process of an embodiment of a speed prediction model provided by the present application.
[0093] like Figure 3 As shown, the specific steps are as follows:
[0094] Step S21: Initialize model hyperparameters of the speed prediction model.
[0095] In the embodiment of the present application, the XGboost prediction model in the above example is used as the speed prediction model here to specifically introduce the model training process.
[0096] Specifically, the XGboost prediction model is a machine learning algorithm based on a gradient boosting decision tree, which integrates multiple weak learners to form a strong learner to improve prediction accuracy. The congestion prediction device initializes the model hyperparameters of the XGboost prediction model, such as the number, depth, and learning rate of the decision tree, and initializes the model parameters of the XGboost prediction model, such as the initial weights of each network layer.
[0097] Step S22: input the training data into the speed prediction model to obtain the predicted value of the training data.
[0098] Step S23: Obtain training residuals based on the true values and predicted values of the training data.
[0099] In the embodiment of the present application, the congestion prediction device inputs the training data, i.e., the traffic congestion data collected by the smart city data platform, into the speed prediction model to obtain the prediction result of the speed prediction model at the current stage, while the true value can be directly obtained from the traffic congestion data. Then, the congestion prediction device calculates the residual according to the prediction result of the speed prediction model at the current stage, i.e., the predicted value and the true value.
[0100] Step S24: using the training residuals to train the decision tree of the speed prediction model, and obtaining the node weights of the leaf nodes of the decision tree.
[0101] In an embodiment of the present application, the congestion prediction device uses the current residual to train a decision tree, divides the samples into leaf nodes according to the structure of the decision tree, and calculates the weight of the leaf node according to the samples and residuals on the leaf node.
[0102] Step S25: using the node weights to update the model parameters of the speed prediction model.
[0103] In an embodiment of the present application, the congestion prediction device adds the prediction result of the current tree and the weight of the leaf node to the prediction result of the current model, and updates the model parameters of the speed prediction model.
[0104] Furthermore, the congestion prediction device also needs to use a test set to evaluate the performance of the speed prediction model. If the speed prediction model performs poorly, the model parameters can be adjusted, such as increasing the number of trees, reducing the learning rate, etc.
[0105] Step S13: generating a target road section correction value of the area to be predicted based on the predicted average travel speed of the target road section.
[0106] In the embodiments of the present application, Figure 2 As shown, the congestion prediction device inputs the target section predicted average travel speed output by the speed prediction model into the speed correction model to predict the speed residual of the target section predicted average travel speed as the target section correction value.
[0107] Specifically, the congestion prediction device captures the detailed characteristics of the average vehicle speed sequence itself through the speed correction model, and predicts the correction value of the target road section through the speed correction model, which is used to accurately predict the congestion situation of the entrance avenue through a combination of internal and external methods.
[0108] Please continue reading Figure 4 , Figure 4 It is a training flow diagram of an embodiment of a speed correction model provided by the present application.
[0109] like Figure 4 As shown, the specific steps are as follows:
[0110] Step S31: Obtain the predicted value and true value of the speed prediction model for the training sample.
[0111] Step S32: construct a training residual sequence using the predicted values and true values of the training samples.
[0112] Step S33: Determine whether the training residual sequence is stationary.
[0113] Step S34: performing differential processing on the training residual sequence until the training residual sequence reaches a stable state.
[0114] Step S35: Train the speed correction model using the steady-state training residual sequence.
[0115] In the embodiment of the present application, ARIMA is a commonly used prediction algorithm based on time series analysis. The model consists of three parts: autoregressive model (AR), difference model (I) and moving average model (MA). The model construction process of the ARIMA model is as follows:
[0116] First, the congestion prediction device constructs a residual sequence based on the predicted value output by the XGboost model and the actual value, and checks the stationarity of the residual sequence. If it is not stationary, it performs a difference, that is, enters step S34, and repeats this step until it is stable.
[0117] Then, the congestion prediction device uses ACF / PACF to determine the model order, evaluates the model according to the BIC criterion, and finally uses the speed correction model to perform residual prediction.
[0118] Step S14: using the target section correction value to correct the predicted average travel speed of the target section to obtain the fused average travel speed of the target section.
[0119] In the embodiment of the present application, the congestion prediction device fuses the prediction results of the XGboost model and the ARIMA model to obtain the final prediction result. Specifically, the congestion prediction device uses the target section correction value to correct the predicted average travel speed of the target section to obtain the fused average travel speed of the target section.
[0120] Step S15: The target road section is integrated with the average travel speed to generate road congestion information of the area to be predicted.
[0121] In an embodiment of the present application, the congestion prediction device can directly judge the road congestion situation based on the average travel speed integrated with the target road section, that is, the greater the average travel speed of the predicted target road section, the less serious the road congestion situation; the smaller the average travel speed of the predicted target road section, the more serious the road congestion situation.
[0122] In the present application, the congestion prediction device obtains the regional information of the area to be predicted; based on the regional information, generates the predicted average travel speed of the target road section of the area to be predicted; based on the predicted average travel speed of the target road section, generates the target road section correction value of the area to be predicted; uses the target road section correction value to correct the predicted average travel speed of the target road section to obtain the fused average travel speed of the target road section; uses the fused average travel speed of the target road section to generate the road congestion information of the area to be predicted. Through the above congestion prediction method, the congestion situation of the target area can be predicted in advance and accurately, providing relevant data for congestion decision-making, and realizing accurate and reliable congestion prediction.
[0123] Furthermore, the congestion prediction device performs daily congestion management based on the road congestion information calculated by the above embodiment, including but not limited to the following functions:
[0124] (1) Congestion warning
[0125] Use intelligent outbound calling technology, text messages, station notifications, APP push, etc. to send congestion warning information to notify traffic police, ports, urban management and other relevant departments.
[0126] Through the cockpit, mobile terminal, etc., you can view the real-time heat maps of the main access highways, trunk roads, port transport capacity schedules, number of parking spaces around the port, etc. in the current terminal and port area, and dynamically display the congestion trend forecast and congestion relief forecast for a period of time in the future based on the digital twin three-dimensional visualization.
[0127] (2) Congestion handling decisions
[0128] County decision-makers can initiate online decision-making meetings through the video conferencing system in the command hall, instruct relevant departments to formulate strategies to alleviate congestion, and watch the real-time handling process and real-time congestion trends in the cockpit.
[0129] Through the large-screen display of model data, auxiliary decision-making data for congestion is provided, such as: whether the traffic lights need to be adjusted, whether the high-speed large screen needs to be linked to issue congestion prompts, whether the checkpoint needs to limit the flow, etc.
[0130] (3) Congestion relief
[0131] For severely congested roads entering the port, the traffic police signal light system can be linked to make dynamic adjustments, and the traffic conditions can be viewed in real time through video.
[0132] Congestion information can be released through the traffic police system and displayed on a large screen on the highway.
[0133] Through operator linkage, reminder text messages are sent to vehicles and car owners within a 10-kilometer radius around the port.
[0134] Traffic police and urban management guidance and arrangement: The conditions of areas near docks and ports, parking lots and parking spaces within the county are displayed in real time and guidance routes are planned. The traffic police, urban management and other personnel who are guiding on site are notified, and intelligent outbound voice suggestions are made when necessary.
[0135] Ro-Ro Ship Scheduling: By analyzing and calculating data such as the Ro-Ro Ship area and the size of vehicles to be loaded and unloaded based on a linear programming algorithm, reasonable shipping scheduling information is calculated, the ship and the shipping vehicle are bound together, and the scheduling information is output to the port staff. The staff will make reasonable scheduling arrangements for the vehicles based on the shipping schedule recommendations, thereby improving the port's shipping efficiency.
[0136] (4) Congestion analysis
[0137] By conducting relevant data analysis on the process data of each congestion, county decision-makers can summarize the results of the data analysis and make improvements on the key nodes of congestion.
[0138] (5) Management of stranded operations after suspension due to extreme weather conditions.
[0139] In extreme weather conditions, docks and ports will stop operating and severe congestion will become inevitable. At this time, the congestion prediction model will become invalid. The focus of governance is to properly resettle people and vehicles stranded within the county, avoid the occurrence of safety incidents, and avoid affecting normal county order.
[0140] (6) Detention warning
[0141] When an extreme weather warning occurs, intelligent outbound calling technology, text messages, station notifications, APP push, etc. are used to send detention warnings to traffic police, ports, urban management and other relevant departments.
[0142] Through the cockpit, mobile terminal, etc., you can view the real-time heat maps of the main access highways, trunk roads, port transport capacity schedules, number of parking spaces around the port, etc. in the current terminal and port area, and dynamically display the congestion trend forecast and congestion relief forecast for a period of time in the future based on the digital twin three-dimensional visualization.
[0143] (7) Decision on detention and disposal
[0144] Statistics of stranded vehicles: Drones are used to cruise the main roads into the port, and AI algorithms are used to identify and count stranded vehicles. The approximate number of stranded vehicles, vehicle types, number of people, and other data are calculated to assist decision makers.
[0145] County decision-makers can initiate online decision-making meetings through the video conferencing system in the command hall, instruct relevant departments to formulate strategies for dealing with stranded people and vehicles, and watch the real-time handling process and real-time congestion trends in real time through the cockpit.
[0146] (8) Management of stranded people and vehicles
[0147] Warning for vehicles entering the port: congestion information can be released through the traffic police system and displayed on the large screen on the highway; through the linkage with operators, reminder text messages are sent to vehicles and car owners within 10 kilometers around the terminal and port, and subsequent vehicles are diverted in advance.
[0148] Traffic police and urban management guidance and arrangement: The conditions of areas near docks and ports, parking lots and parking spaces within the county are displayed in real time and guidance routes are planned. The traffic police, urban management and other personnel who are guiding on site are notified, and intelligent outbound voice suggestions are made when necessary.
[0149] Arrangements for stranded drivers: Stranded drivers will be notified of guidance information and resettlement arrangements through drone cruise announcements, smart outbound calls, text messages, mini-programs, etc., their vehicles will be arranged in nearby parking lots, and information suggestions such as nearby hotels will be pushed to drivers; and stranded people will be notified of real-time government messages.
[0150] (9) Shipping arrangement and dispatch after sailing:
[0151] Through data analysis, we can provide the port entry plan and route arrangement for stranded vehicles after the sailing starts in advance, provide suggestions to decision makers and restore port traffic as soon as possible.
[0152] By analyzing and calculating the data such as the area of ro-ro ships and the size of vehicles to be loaded and unloaded based on linear programming algorithms, reasonable shipping arrangement information is calculated, the relationship between the ship and the shipping vehicle is bound, and the scheduling information is output to the port staff. The staff will make reasonable scheduling arrangements for the vehicles based on the shipping arrangement recommendations, thereby improving the port shipping efficiency.
[0153] (10) Retention analysis
[0154] Relevant data analysis will be conducted on the process data of each detention. County decision-makers can summarize the results of the data analysis and make improvements on the key nodes of detention disposal.
[0155] Regarding the above congestion handling decision, this application also provides a solution for optimizing the congestion handling decision. For details, please refer to Figure 5 , Figure 5 It is a flowchart of another embodiment of the congestion prediction method provided by the present application.
[0156] like Figure 5 As shown, the specific steps are as follows:
[0157] Step S41: Generate several congestion handling decision plans based on road congestion information.
[0158] In an embodiment of the present application, the congestion prediction device can generate several congestion handling decision plans based on road congestion information.
[0159] Specifically, the congestion prediction device can obtain detailed information about the port area based on the road congestion information, namely, the specific distribution of congested roads, the non-congested areas of congested roads, the start time of congestion on congested roads, and the duration of congestion. The congestion handling decision plan generated by the congestion prediction device is to guide vehicles on the congested roads to the non-congested areas and to relieve the congested roads. In addition, the congestion handling decision plan can also include any one or more of the above congestion management plans, and their combination plans are not listed here one by one.
[0160] Step S42: Based on each congestion handling decision plan and the area information of the area to be predicted, generate prediction area information corresponding to each congestion handling decision plan.
[0161] In the embodiment of the present application, the congestion prediction device generates predicted area information according to each generated congestion handling decision plan and the current area information. The predicted area information is the area change in the future period after the terminal executes the congestion handling decision plan, which is used to verify whether the congestion handling decision plan produces good results.
[0162] Step S43: Compare the predicted area information of several congestion handling decision plans to obtain the optimal congestion handling decision plan.
[0163] In the embodiment of the present application, the congestion prediction device can obtain the advantages and disadvantages of each congestion handling decision plan by comparing the predicted area information of each congestion handling decision plan, thereby determining the optimal congestion handling decision plan to execute.
[0164] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.
[0165] In order to implement the above congestion prediction method, this application also proposes a congestion prediction device, please refer to Figure 6 , Figure 6 It is a structural diagram of an embodiment of a congestion prediction device provided by the present application.
[0166] The congestion prediction device 500 of this embodiment includes a region acquisition module 51 , a speed prediction module 52 , a speed correction module 53 , a speed fusion module 54 , and a congestion prediction module 55 .
[0167] The region acquisition module 51 is used to acquire region information of the region to be predicted.
[0168] The speed prediction module 52 is used to generate a predicted average travel speed of a target road section in the area to be predicted based on the area information.
[0169] The speed correction module 53 is used to generate a target section correction value of the area to be predicted based on the predicted average travel speed of the target section.
[0170] The speed fusion module 54 is used to correct the predicted average travel speed of the target section by using the target section correction value to obtain a fused average travel speed of the target section.
[0171] The congestion prediction module 55 is used to generate the road congestion information of the area to be predicted by integrating the target road section with the average travel speed.
[0172] In order to implement the above congestion prediction method, this application also proposes another congestion prediction device, please refer to Figure 7 , Figure 7 It is a structural diagram of another embodiment of the congestion prediction device provided by the present application.
[0173] The congestion prediction device 600 of this embodiment includes a processor 61 , a memory 62 , an input / output device 63 , and a bus 64 .
[0174] The processor 61 , the memory 62 , and the input / output device 63 are respectively connected to the bus 64 . The memory 62 stores program data, and the processor 61 is used to execute the program data to implement the congestion prediction method described in the above embodiment.
[0175] In the embodiment of the present application, the processor 61 may also be referred to as a CPU (Central Processing Unit). The processor 61 may be an integrated circuit chip having the ability to process signals. The processor 61 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, or discrete hardware components. A general-purpose processor may be a microprocessor or the processor 61 may also be any conventional processor, etc.
[0176] This application also provides a computer storage medium, please continue to refer to Figure 8 , Figure 8 It is a structural diagram of an embodiment of a computer storage medium provided in the present application. The computer storage medium 700 stores a computer program 71. When the computer program 71 is executed by a processor, it is used to implement the congestion prediction method of the above embodiment.
[0177] When the embodiments of the present application are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.
[0178] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A congestion prediction method, characterized in that: The congestion prediction method comprises: Obtain regional information of the area to be predicted; Based on the area information, generating a predicted average travel speed of a target road section in the area to be predicted; Generating a target road section correction value of the area to be predicted based on the predicted average travel speed of the target road section; The predicted average travel speed of the target section is corrected by using the target section correction value to obtain a fused average travel speed of the target section; The target road section is combined with the average travel speed to generate road congestion information of the area to be predicted.
2. The congestion prediction method according to claim 1, characterized in that: The area information includes the current average vehicle speed of the target road section of the area to be predicted, and the environmental information of the area to be predicted, and the environmental information includes: weather data, time data, checkpoint data, port capacity data, parking lot data, and / or road data.
3. The congestion prediction method according to claim 2, characterized in that: The generating, based on the area information, a predicted average travel speed of a target road section in the area to be predicted, comprises: Inputting the regional information into a speed prediction model to extract environmental impact features from the regional information; The speed prediction model is used to predict the environmental impact characteristics to generate a predicted average travel speed of the target road section in the area to be predicted.
4. The congestion prediction method according to claim 3, characterized in that: The congestion prediction method further includes: Initializing model hyperparameters of the speed prediction model; Inputting training data into the speed prediction model to obtain a predicted value of the training data; Obtaining training residuals based on the true values and predicted values of the training data; Using the training residuals to train a decision tree of the speed prediction model, and obtaining node weights of leaf nodes of the decision tree; The model parameters of the speed prediction model are updated using the node weights.
5. The congestion prediction method according to claim 2, characterized in that: The generating the target road section correction value of the area to be predicted based on the predicted average travel speed of the target road section includes: Inputting the predicted average travel speed of the target road section into a speed correction model to extract vehicle speed sequence features in the regional information; The speed correction model is used to predict the vehicle speed sequence characteristics to generate a target road section correction value for the area to be predicted.
6. The congestion prediction method according to claim 5, characterized in that: The congestion prediction method further includes: Obtain the predicted value and true value of the speed prediction model for the training sample; Constructing a training residual sequence using the predicted values and true values of the training samples; Determining whether the training residual sequence is stable; If yes, performing differential processing on the training residual sequence until the training residual sequence reaches a stable state; The speed correction model is trained using a stationary training residual sequence.
7. The congestion prediction method according to claim 1, characterized in that: After the target road section is used to fuse the average travel speed to generate the road congestion information of the area to be predicted, the congestion prediction method further includes: generating a number of congestion handling decision plans based on the road congestion information; Based on each congestion handling decision plan and the area information of the area to be predicted, generating prediction area information corresponding to each congestion handling decision plan; The predicted area information of the several congestion handling decision plans are compared to obtain the optimal congestion handling decision plan.
8. A congestion prediction device, characterized in that: The congestion prediction device includes an area acquisition module, a speed prediction module, a speed correction module, a speed fusion module, and a congestion prediction module; wherein, The region acquisition module is used to acquire region information of the region to be predicted; The speed prediction module is used to generate a predicted average travel speed of a target road section in the area to be predicted based on the area information; The speed correction module is used to generate a target section correction value of the area to be predicted based on the predicted average travel speed of the target section; The speed fusion module is used to correct the predicted average travel speed of the target section by using the target section correction value to obtain the fused average travel speed of the target section; The congestion prediction module is used to generate road congestion information of the area to be predicted by integrating the target road section with the average travel speed.
9. A congestion prediction device, characterized in that: The congestion prediction device includes a memory and a processor coupled to the memory; The memory is used to store program data, and the processor is used to execute the program data to implement the congestion prediction method according to any one of claims 1 to 7.
10. A computer storage medium, characterized in that: The computer storage medium is used to store program data, and when the program data is executed by a computer, it is used to implement the congestion prediction method according to any one of claims 1 to 7.