A urban traffic management system and method based on big data
Through big data analysis and prediction of road parking and traffic flow, buses are dispatched, urban traffic congestion problems are solved, road resource utilization is optimized, and the status of public transportation and travel efficiency are improved.
Patent Information
- Application Number
- CN202211077291.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-09-05
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-09-05
AI Technical Summary
The urban traffic congestion problem is serious, mainly due to insufficient parking lots, unreasonable road network structure and low public transportation service quality, and the lack of effective solutions.
The urban traffic management system based on big data, by obtaining road historical parking and traffic data, building prediction models, predicting road parking and traffic flow, setting traffic congestion thresholds, dispatching buses when the threshold is reached, optimizing road resource utilization, and improving public Transportation status.
Effectively predict road traffic congestion, dispatch buses, reduce road congestion, improve travel efficiency, meet residents' needs, and save energy and environmental protection.
Smart Images

Figure CN115457771B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of urban traffic management, and in particular to an urban traffic management system and method based on big data. Background Art
[0002] Entering the 21st century, my country's urbanization process is accelerating, the size of cities is increasing, and the income and living standards of urban residents have also been greatly improved. What comes with it is an increasingly fast pace of life, a stronger sense of time, longer travel distances, an increasing number of private cars, and more frequent economic ties and personnel exchanges between cities and between cities and hinterland areas. The substantial increase in traffic demand has put great pressure on the existing transportation facilities in cities. At present, the development of transportation facilities generally cannot keep up with the growth rate of transportation demand, so traffic congestion has arisen and is becoming increasingly serious. This not only brings inconvenience to people's normal work, study and life, but also leads to a waste of resources, affecting the living environment, air quality and the healthy and rapid development of cities.
[0003] my country's urbanization process is accelerating, the number of private cars has increased rapidly, the total number of residents' trips has grown steadily, and the travel distance has increased rapidly. The main reasons for traffic congestion are: (1) There is a serious shortage of parking lots in my country's cities. Many cities do not have standardized parking lots, which leads to the widespread phenomenon of random parking of motor vehicles, which has caused traffic congestion to a certain extent. (2) There are too many cars and too few roads, the road network structure is unreasonable, and the overall efficiency is low. For a long time, the per capita road area in my country's cities has been at a low level, the proportion of road land to urban land is small, the road network density is low, and the total road capacity supply is seriously insufficient. In the past decade, despite the rapid development of road volume, it still cannot keep up with the annual growth rate of 20% of urban traffic volume. (3) my country's urban public transportation advantages are insufficient, the traffic share rate is low, and the number of cars has increased sharply and the utilization rate is extremely high. The development of urban rail transit is lagging behind, there are problems in the implementation of public transportation priority, and the quality of public transportation services cannot meet people's needs, which has reduced the public transportation share rate. At the same time, the utilization rate of private cars in my country is extremely high.
[0004] The lack of standardized parking lots and unreasonable road network structure have led to motor vehicles being parked randomly on both sides of the road, resulting in a sudden increase in traffic volume during morning and evening rush hours, serious traffic congestion, and enormous pressure on urban traffic. This is a problem that needs to be solved urgently. Among the current technical means, no effective solution has been proposed to alleviate the traffic congestion caused by road parking. Summary of the invention
[0005] The purpose of the present invention is to provide a big data-based urban traffic management system and method to solve the problems raised in the above-mentioned background technology.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for urban traffic management based on big data, the method comprising the following steps:
[0007] Obtain historical road parking data, build a road parking prediction model, and predict the road parking prediction value for the next cycle;
[0008] Obtain historical data on road traffic flow, build a road traffic flow prediction model, and predict the road traffic flow forecast value for the next cycle;
[0009] The road traffic congestion prediction value is obtained based on the road parking prediction value and the road traffic flow prediction value for the next cycle of the road;
[0010] Build an urban traffic management simulation platform, set road traffic congestion thresholds, and dispatch buses when the predicted road traffic congestion value exceeds the road traffic congestion threshold.
[0011] Among them, on the one hand, "public transport priority" can control the total amount of urban traffic to a certain extent, reduce the proportion of individual transportation modes, and improve the status of public transportation in the entire urban transportation system; on the other hand, choosing public transportation can more effectively utilize road resources, reduce road traffic congestion, and be more energy-saving and environmentally friendly.
[0012] According to the above technical solution, in step S1, building a road parking prediction model includes:
[0013] Acquiring historical road parking data, wherein the historical road parking data includes historical parking data of domestic roads and historical parking data of monitored roads;
[0014] The historical parking data of the road includes parking start time, parking end time, parking quantity, parking duration, and parking location;
[0015] The cycle is in hours, one hour is one cycle, and time t is the number of cycles;
[0016] Get the time series X=(X t , X t-1 , ..., X t-N+1 ), as model input;
[0017] A moving average is:
[0018]
[0019] Among them, S t is the moving average of the tth period, X t is the observed value of period t, and N is the moving average period;
[0020] The second moving average of the first moving average of the time series is obtained by performing the second moving average:
[0021]
[0022] The road parking prediction model is:
[0023] S t+M =A t +N t *M
[0024] A t =2S t -H t
[0025]
[0026] Among them, A t is the intercept, i.e. the basic level of the phenomenon in period t, B t is the slope, i.e. the change in the t-th period per unit time, M is the number of future forecast periods, S t is the moving average of the tth period, H t is the quadratic moving average of the tth period, S t+M is the predicted value of road parking in period t+M.
[0027] The quadratic moving average method solves the contradiction that the predicted value lags behind the actual value, and is suitable for the prediction of time series of market phenomena with obvious trend changes. In real life, it can accurately predict road parking and achieve ideal prediction results.
[0028] According to the above technical solution, in step S2, building a road traffic flow prediction model includes:
[0029] Acquiring historical road traffic flow data, wherein the historical road traffic flow data includes historical traffic flow data of domestic roads and historical traffic flow data of monitored roads;
[0030] The historical traffic flow data of the road includes vehicle entry time, vehicle exit time, number of vehicles, mainstream traffic flow, and diversion traffic flow;
[0031] The cycle is in hours, one hour is one cycle, and time t is the number of cycles;
[0032] Get the time series Y = (y1, y2, ..., y n ), as model input;
[0033] The formula for first-order exponential smoothing is:
[0034] S t (1) =ayt +(1-a)S t-1 (1)
[0035] Among them, S t (1) is the exponential smoothing value of the tth period, y t is the observed value of period t, S t-1 (1) is the exponential smoothing value of the t-1 period, and a is the smoothing index;
[0036] The formula for quadratic exponential smoothing is:
[0037] S t (2) =S t (1) +(1-a)S t-1 (2)
[0038] Among them, S t (2) is the quadratic exponential smoothing value of the tth period, S t-1 (2) The quadratic exponential smoothing value of the t-1 period, a is the smoothing index;
[0039] The road traffic flow prediction model is:
[0040] Y t+T =a t +b t *T
[0041] a t =2S t (1) -S t (2)
[0042]
[0043] Among them, a t is the intercept, i.e. the base level in period t, b t is the slope, i.e. the change per unit time in period t, Y t+T is the predicted value of road traffic flow at time t+T.
[0044] The quadratic exponential smoothing method further strengthens the effect of recent observations on the predicted values. The weights assigned to observations at different times are different, thereby increasing the weights of recent observations, so that the predicted values can quickly reflect actual market changes. In the actual urban traffic environment, especially during peak hours in the morning and evening, the use of the quadratic exponential smoothing method can more quickly obtain predicted values based on recent observations, thereby achieving the effect of accurate prediction.
[0045] According to the above technical solution, in step S3, a road traffic congestion prediction value is obtained based on the road parking prediction value and the road traffic flow prediction value of the next cycle of the road;
[0046] Road traffic congestion prediction value Z:
[0047] Z=p*S t+M +q*Y t+T
[0048] Among them, the weight ratio of road parking is recorded as p, the weight ratio of traffic flow is recorded as q, S t+M is the predicted value of road parking in period t+M, Y t+T is the predicted value of road traffic flow at time t+T.
[0049] According to the above technical solution, in step S4, an urban traffic management simulation platform is constructed, a road traffic congestion threshold Z0 is set, and when the road traffic congestion prediction value exceeds the road traffic congestion threshold, a bus is dispatched;
[0050] Set the road traffic congestion numerical threshold Z0; when Z>Z0, buses need to be dispatched;
[0051] The day is divided into three periods: morning peak period, evening peak period and normal period;
[0052] Set the morning peak period to 7am to 9am; set the evening peak period to 5pm to 7pm; set the normal period to other periods except the morning and evening peak periods;
[0053] The number of buses a1 dispatched during the morning and evening peak hours is:
[0054] a1=k1b+k2(Z-Z0);
[0055] The number of buses a2 dispatched during normal times is:
[0056] a2=k1b+k3(Z-Z0);
[0057] Among them, the number of buses under the original dispatch plan is recorded as b, the dispatch coefficient under the original dispatch plan is recorded as k1, the dispatch coefficient during peak hours in the morning and evening is recorded as k2, and the dispatch coefficient during normal periods is recorded as k3.
[0058] Among them, the urban traffic management simulation platform is a simulation platform that integrates modern communications, information, computers, networks, GPS and other high-tech technologies into urban traffic management. It can monitor road parking data and road traffic flow in real time, and dispatch buses as needed.
[0059] According to the above technical solution, the system includes a multi-source data acquisition module, a prediction model construction module, a prediction model analysis module, and an urban traffic management module;
[0060] The multi-source data acquisition module is used to acquire historical road parking data and historical road traffic data to construct a multi-source database; the prediction model construction module is used to construct a road parking prediction model based on the historical road parking data collected by the multi-source data acquisition module and to construct a road traffic flow prediction model based on the historical road traffic flow data collected by the multi-source data acquisition module; the prediction model analysis module is used to predict the road parking prediction value of the next cycle of the road based on the road parking prediction model and to predict the road traffic flow prediction value of the next cycle of the road based on the road traffic flow prediction model; the urban traffic management module is used to construct an urban traffic management simulation platform to dispatch buses to alleviate road traffic congestion according to the road traffic congestion situation;
[0061] The output end of the multi-source data acquisition module is connected to the input end of the prediction model construction module; the output end of the prediction model construction module is connected to the input end of the prediction model analysis module; the output end of the prediction model analysis module is connected to the input end of the urban traffic management module.
[0062] According to the above technical solution, the multi-source data acquisition module includes a domestic road historical parking data acquisition unit and a domestic road historical vehicle flow data acquisition unit;
[0063] The domestic road historical parking data collection unit is used to collect domestic road historical parking data; the road historical parking data includes parking start time, parking end time, parking quantity, parking duration, and parking location;
[0064] The domestic road historical traffic flow data collection unit is used to collect domestic road historical traffic flow data; the road historical traffic flow data includes vehicle entry time, vehicle exit time, vehicle number, mainstream traffic flow, and diversion traffic flow;
[0065] The output end of the domestic road historical parking data collection unit is connected to the input end of the prediction model construction module; the output end of the domestic road historical vehicle flow data collection unit is connected to the input end of the prediction model construction module.
[0066] According to the above technical solution, the prediction model building module includes a road parking prediction model building unit and a road vehicle flow prediction model building unit;
[0067] The road parking prediction model building unit is used to build a road parking prediction model; the road vehicle flow prediction model building unit is used to build a road vehicle flow prediction model;
[0068] The output end of the road parking prediction model construction unit is connected to the input end of the road vehicle flow prediction model construction unit.
[0069] According to the above technical solution, the prediction model analysis module includes a road parking prediction model analysis unit and a road vehicle flow prediction model analysis unit;
[0070] The road parking prediction model analysis unit is used to analyze the road parking prediction model; the road vehicle flow prediction model analysis unit is used to analyze the road vehicle flow prediction model;
[0071] The output end of the road parking prediction model analysis unit is connected to the input end of the road vehicle flow prediction model analysis unit.
[0072] According to the above technical solution, the urban traffic management module includes an urban traffic management simulation platform construction unit and an urban traffic management analysis unit;
[0073] The urban traffic management simulation platform construction unit is used to construct an urban traffic management simulation platform to simulate road bus dispatching; the urban traffic management analysis unit is used to analyze the road traffic congestion situation, carry out different bus dispatching according to the road traffic congestion situation, and feed back the results to the system;
[0074] The output end of the urban traffic management simulation platform construction unit is connected to the input end of the urban traffic management analysis unit.
[0075] Compared with the prior art, the beneficial effects achieved by the present invention are:
[0076] The present invention can provide a road parking prediction model and a road traffic vehicle flow prediction model, conduct a comprehensive analysis based on road parking data and road vehicle flow data, and comprehensively propose a solution to the current situation of road traffic congestion caused by road parking. It can predict road parking data and road vehicle flow data, and when the road parking data reaches a certain threshold, establish a simulation system platform, construct bus points, plan bus travel routes, and dispatch buses, effectively solving the problem of the current situation of road traffic congestion caused by road parking, further improving people's travel rate, and meeting people's daily travel needs. BRIEF DESCRIPTION OF THE DRAWINGS
[0077] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0078] Figure 1 It is a flow chart of a method for urban traffic management based on big data of the present invention;
[0079] Figure 2 It is a structural schematic diagram of a big data-based urban traffic management system of the present invention. DETAILED DESCRIPTION
[0080] 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 only 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.
[0081] See also Figure 1-Figure 2 The present invention provides a technical solution: a method for urban traffic management based on big data, the method comprising the following steps:
[0082] Obtain historical road parking data, build a road parking prediction model, and predict the road parking prediction value for the next cycle;
[0083] Obtain historical data on road traffic flow, build a road traffic flow prediction model, and predict the road traffic flow forecast value for the next cycle;
[0084] The road traffic congestion prediction value is obtained based on the road parking prediction value and the road traffic flow prediction value for the next cycle of the road;
[0085] Build an urban traffic management simulation platform, set road traffic congestion thresholds, and dispatch buses when the predicted road traffic congestion value exceeds the road traffic congestion threshold.
[0086] In step S1, constructing a road parking prediction model includes:
[0087] Acquiring historical road parking data, wherein the historical road parking data includes historical parking data of domestic roads and historical parking data of monitored roads;
[0088] The historical parking data of the road includes parking start time, parking end time, parking quantity, parking duration, and parking location;
[0089] The cycle is in hours, one hour is one cycle, and time t is the number of cycles;
[0090] Get the time series X=(X t , X t-1 , ..., X t-N+1 ), as model input;
[0091] A moving average is:
[0092]
[0093] Among them, S t is the moving average of the tth period, X t is the observed value of period t, and N is the moving average period;
[0094] The second moving average of the first moving average of the time series is obtained by performing the second moving average:
[0095]
[0096] The road parking prediction model is:
[0097] S t+M =A t +N t *M
[0098] A t =2S t -H t
[0099]
[0100] Among them, A t is the intercept, i.e. the basic level of the phenomenon in period t, B t is the slope, i.e. the change in the t-th period per unit time, M is the number of periods predicted in the future, S t is the moving average of the tth period, H t is the quadratic moving average of the tth period, S t+M is the predicted value of road parking in period t+M.
[0101] In step S2, constructing a road traffic flow prediction model includes:
[0102] Acquiring historical road traffic flow data, wherein the historical road traffic flow data includes historical traffic flow data of domestic roads and historical traffic flow data of monitored roads;
[0103] The historical traffic flow data of the road includes vehicle entry time, vehicle exit time, number of vehicles, mainstream traffic flow, and diversion traffic flow;
[0104] The cycle is in hours, one hour is one cycle, and time t is the number of cycles;
[0105] Get the time series Y = (y1, y2, ..., y n ), as model input;
[0106] The formula for first-order exponential smoothing is:
[0107] S t (1) =ay t+(1-a)S t-1 (1)
[0108] Among them, S t (1) is the exponential smoothing value of the tth period, y t is the observed value of period t, S t-1 (1) is the exponential smoothing value of the t-1 period, and a is the smoothing index;
[0109] The formula for quadratic exponential smoothing is:
[0110] S t (2) =S t (1) +(1-a)S t-1 (2)
[0111] Among them, S t (2) is the quadratic exponential smoothing value of the tth period, S t-1 (2) The quadratic exponential smoothing value of the t-1 period, a is the smoothing index;
[0112] The road traffic flow prediction model is:
[0113] Y t+T =a t +b t *T
[0114] a t =2S t (1) -S t (2)
[0115]
[0116] Among them, a t is the intercept, i.e. the base level in period t, b t is the slope, i.e. the change per unit time in period t, Y t+T is the predicted value of road traffic flow at time t+T.
[0117] In step S3, a road traffic congestion prediction value is obtained according to the road parking prediction value and the road traffic flow prediction value of the next cycle of the road;
[0118] Road traffic congestion prediction value Z:
[0119] Z=p*S t+M +q*Y t+T
[0120] Among them, the weight ratio of road parking is recorded as p, the weight ratio of traffic flow is recorded as q, S t+M is the predicted value of road parking in period t+M, Y t+T is the predicted value of road traffic flow at time t+T.
[0121] In step S4, an urban traffic management simulation platform is constructed, a road traffic congestion threshold Z0 is set, and when the road traffic congestion prediction value exceeds the road traffic congestion threshold, a bus is dispatched;
[0122] Set the road traffic congestion numerical threshold Z0; when Z>Z0, buses need to be dispatched;
[0123] The day is divided into three periods: morning peak period, evening peak period and normal period;
[0124] Set the morning peak period to 7am to 9am; set the evening peak period to 5pm to 7pm; set the normal period to other periods except the morning and evening peak periods;
[0125] The number of buses a1 dispatched during the morning and evening peak hours is:
[0126] a1=k1b+k2(Z-Z0);
[0127] The number of buses a2 dispatched during normal times is:
[0128] a2=k1b+k3(Z-Z0);
[0129] Among them, the number of buses under the original dispatch plan is recorded as b, the dispatch coefficient under the original dispatch plan is recorded as k1, the dispatch coefficient during peak hours in the morning and evening is recorded as k2, and the dispatch coefficient during normal periods is recorded as k3.
[0130] The system includes a multi-source data acquisition module, a prediction model construction module, a prediction model analysis module, and an urban traffic management module;
[0131] The multi-source data acquisition module is used to acquire historical road parking data and historical road traffic data to construct a multi-source database; the prediction model construction module is used to construct a road parking prediction model based on the historical road parking data collected by the multi-source data acquisition module and to construct a road traffic flow prediction model based on the historical road traffic flow data collected by the multi-source data acquisition module; the prediction model analysis module is used to predict the road parking prediction value of the next cycle of the road based on the road parking prediction model and to predict the road traffic flow prediction value of the next cycle of the road based on the road traffic flow prediction model; the urban traffic management module is used to construct an urban traffic management simulation platform to dispatch buses to alleviate road traffic congestion according to the road traffic congestion situation;
[0132] The output end of the multi-source data acquisition module is connected to the input end of the prediction model construction module; the output end of the prediction model construction module is connected to the input end of the prediction model analysis module; the output end of the prediction model analysis module is connected to the input end of the urban traffic management module.
[0133] The multi-source data collection module includes a domestic road historical parking data collection unit and a domestic road historical vehicle flow data collection unit;
[0134] The domestic road historical parking data collection unit is used to collect domestic road historical parking data; the domestic road historical vehicle flow data collection unit is used to collect domestic road historical vehicle flow data;
[0135] The output end of the domestic road historical parking data collection unit is connected to the input end of the prediction model construction module; the output end of the domestic road historical vehicle flow data collection unit is connected to the input end of the prediction model construction module.
[0136] The prediction model building module includes a road parking prediction model building unit and a road vehicle flow prediction model building unit;
[0137] The road parking prediction model building unit is used to build a road parking prediction model; the road vehicle flow prediction model building unit is used to build a road vehicle flow prediction model;
[0138] The output end of the road parking prediction model construction unit is connected to the input end of the road vehicle flow prediction model construction unit.
[0139] The prediction model analysis module includes a road parking prediction model analysis unit and a road vehicle flow prediction model analysis unit;
[0140] The road parking prediction model analysis unit is used to analyze the road parking prediction model; the road vehicle flow prediction model analysis unit is used to analyze the road vehicle flow prediction model;
[0141] The output end of the road parking prediction model analysis unit is connected to the input end of the road vehicle flow prediction model analysis unit.
[0142] The urban traffic management module includes an urban traffic management simulation platform construction unit and an urban traffic management analysis unit;
[0143] The urban traffic management simulation platform construction unit is used to construct an urban traffic management simulation platform to simulate road bus dispatching; the urban traffic management analysis unit is used to analyze the road traffic congestion situation, carry out different bus dispatching according to the road traffic congestion situation, and feed back the results to the system;
[0144] The output end of the urban traffic management simulation platform construction unit is connected to the input end of the urban traffic management analysis unit.
[0145] In the examples:
[0146] A road R is provided and monitored;
[0147] Building a road parking prediction model includes:
[0148] Construct a multi-source database based on historical road parking data;
[0149] Get the time series X=(X t , X t-1 , ..., X t-N+1 ), as model input;
[0150] That is, in this embodiment, the historical data of 8 hours on the day are selected to further predict the data of the 12th hour on the day;
[0151] A moving average is:
[0152]
[0153] Among them, S t is the moving average of the tth period, X t is the observed value of period t, and N is the moving average period;
[0154] The second moving average of the first moving average of the time series is obtained by performing the second moving average:
[0155]
[0156] The road parking prediction model is:
[0157] S t+M =A t +B t *M
[0158] A t =2S t -H t
[0159]
[0160] Among them, A t is the intercept, i.e. the basic level of the phenomenon in period t, B t is the slope, i.e. the change in the t-th period per unit time, M is the number of periods predicted in the future, S t is the moving average of the tth period, H t is the quadratic moving average of the tth period, St+M is the predicted value of road parking in period t+M;
[0161] When t=8,M=4,S 12 is the predicted value of road parking at the 12th hour;
[0162] Building a road traffic flow prediction model includes:
[0163] Construct a multi-source database based on historical road traffic data;
[0164] Get the time series Y = (y1, y2, ..., y n ), as model input;
[0165] The formula for first-order exponential smoothing is:
[0166] S t (1) =ay t +(1-a)S t-1 (1)
[0167] Among them, S t (1) is the exponential smoothing value, y t is the observed value of period t, S t-1 (1) is the exponential smoothing value of the t-1 period, and a is the smoothing index;
[0168] The formula for quadratic exponential smoothing is:
[0169] S t (2) =S t (1) +(1-a)S t-1 (2)
[0170] Among them, S t (2) is the quadratic exponential smoothing value of the tth period, S t-1 (2) The quadratic exponential smoothing value of the t-1 period, a is the smoothing index;
[0171] The road traffic flow prediction model is:
[0172] Y t+T =a t +b t *T
[0173] a t =2S t (1) -S t (2)
[0174]
[0175] Among them, a t is the intercept, i.e. the base level in period t, b t is the slope, i.e. the change per unit time in period t, Y t+T is the predicted value of road traffic volume at time t+T;
[0176] When t=8,T=4,Y 12 is the predicted value of road traffic volume at the 12th hour;
[0177] The road traffic congestion prediction value is obtained based on the road parking prediction value and road traffic flow prediction value of the next cycle; the road traffic congestion prediction value Z:
[0178] Z=p*S t+M +q*Y t+T
[0179] Among them, the weight ratio of road parking is recorded as p, the weight ratio of traffic flow is recorded as q, S t+M is the predicted value of road parking in period t+M, Y t+T is the predicted value of road traffic flow at time t+T.
[0180] Set the road traffic congestion numerical threshold Z0; when Z>Z0, buses need to be dispatched;
[0181] The day is divided into three periods: morning peak period, evening peak period and normal period;
[0182] Set the morning peak period to 7am to 9am; set the evening peak period to 5pm to 7pm; set the normal period to other periods except the morning and evening peak periods;
[0183] The number of buses a1 dispatched during the morning and evening peak hours is:
[0184] a1=k1b+k2(Z-Z0);
[0185] The number of buses a2 dispatched during normal times is:
[0186] a2=k1b+k3(Z-Z0);
[0187] Among them, the number of buses under the original dispatch plan is recorded as b, the dispatch coefficient under the original dispatch plan is recorded as k1, the dispatch coefficient during peak hours in the morning and evening is recorded as k2, and the dispatch coefficient during normal periods is recorded as k3.
[0188] The predicted value of road traffic congestion can be obtained based on the predicted value of road parking and the predicted value of road traffic flow, and the road traffic congestion threshold Z0 is set. When the predicted value of road traffic congestion exceeds the road traffic congestion threshold, buses are dispatched to alleviate traffic congestion.
[0189] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0190] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A method for urban traffic management based on big data, characterized in that: The method comprises the following steps: S1: Obtain historical road parking data, build a road parking prediction model, and predict the road parking prediction value for the next cycle; S2: Obtain historical data on road traffic flow, build a road traffic flow prediction model, and predict the road traffic flow prediction value for the next cycle; S3: Calculate the road traffic congestion prediction value according to the road parking prediction value and the road traffic flow prediction value of the next cycle; S4: Build an urban traffic management simulation platform, set a road traffic congestion threshold, and dispatch buses when the road traffic congestion prediction value exceeds the road traffic congestion threshold; In step S1, constructing a road parking prediction model includes: Acquiring historical road parking data, wherein the historical road parking data includes historical parking data of domestic roads and historical parking data of monitored roads; The historical parking data of the road includes parking start time, parking end time, parking quantity, parking duration, and parking location; The cycle is in hours, one hour is one cycle, and time t is the number of cycles; Get the time series X=(X t , X t-1 , ..., X t-N+1 ), as model input; A moving average is: Among them, S t is the moving average of the tth period, X t is the observed value of period t, and N is the moving average period; The second moving average of the first moving average of the time series is obtained by performing the second moving average: The road parking prediction model is: S t+M =A t +B t *M A t =2S t -H t Among them, A t is the intercept, i.e. the basic level of the phenomenon in period t, B t is the slope, i.e. the change in the t-th period per unit time, M is the number of future forecast periods, S t is the moving average of the tth period, H t is the quadratic moving average of the tth period, S t+M is the predicted value of road parking in period t+M; In step S2, constructing a road traffic flow prediction model includes: Acquiring historical road traffic flow data, wherein the historical road traffic flow data includes historical traffic flow data of domestic roads and historical traffic flow data of monitored roads; The cycle is in hours, one hour is one cycle, and time t is the number of cycles; Get the time series Y = (y1, y2, ..., y n ), as model input; The formula for first-order exponential smoothing is: S t (1) =is t +(1-a)S t-1 (1) Among them, S t (1) is the exponential smoothing value of the tth period, y t is the observed value of period t, S t-1 (1) is the exponential smoothing value of the t-1 period, and a is the smoothing index; The formula for quadratic exponential smoothing is: S t (2) =S t (1) +(1-a)S t-1 (2) Among them, S t (2) is the quadratic exponential smoothing value of the tth period, S t-1 (2) The quadratic exponential smoothing value of the t-1 period, a is the smoothing index; The road traffic flow prediction model is: Y t+T =a t +b t *T a t =2S t (1) -S t (2) Among them, a t is the intercept, i.e. the base level in period t, b t is the slope, i.e. the change per unit time in the tth period, T is the expected lead period number; Y t+T is the predicted value of road traffic volume at the time t+T period; In step S3, a road traffic congestion prediction value is obtained according to the road parking prediction value and the road traffic flow prediction value of the next cycle of the road; Road traffic congestion prediction value Z: Z=p*S t+M +q*Y t+T Among them, the weight ratio of road parking is recorded as p, the weight ratio of traffic flow is recorded as q, S t+M is the predicted value of road parking in period t+M, Y t+T is the predicted value of road traffic volume at time t+T; In step S4, an urban traffic management simulation platform is constructed, a road traffic congestion threshold Z0 is set, and when the road traffic congestion prediction value exceeds the road traffic congestion threshold, a bus is dispatched; Set the road traffic congestion numerical threshold Z0; when Z>Z0, buses need to be dispatched; The day is divided into three periods: morning peak period, evening peak period and normal period; Set the morning peak period to 7am to 9am; set the evening peak period to 5pm to 7pm; set the normal period to other periods except the morning and evening peak periods; The number of buses a1 dispatched during the morning and evening peak hours is: a1=k1b+k2(Z-Z0); The number of buses a2 dispatched during normal times is: a2=k1b+k3(Z-Z0); Among them, the number of buses under the original dispatch plan is recorded as b, the dispatch coefficient under the original dispatch plan is recorded as k1, the dispatch coefficient during peak hours in the morning and evening is recorded as k2, and the dispatch coefficient during normal periods is recorded as k3.
