Smart city traffic signal lamp management and control system

Abstract

The invention discloses a smart city traffic signal lamp management and control system, and relates to the technical field of traffic signal control. The system comprises a region division module, a traffic flow detection sensor, a region traffic flow analysis module, a signal lamp management and control strategy generation module, an inter-region signal lamp cooperative control module and a variable information sign and information processing and generation module. The system can accurately monitor and analyze urban traffic flow conditions in real time by comprehensively applying an advanced region division algorithm, a traffic flow detection technology and an accurate traffic flow state evaluation model, and can intelligently generate a signal lamp management and control strategy according to an analysis result, so that the traffic lamp management and control efficiency is improved. Comprising the steps that a congestion relieving dynamic timing algorithm is adopted in a congestion area, the number of times of signal lamp switching is reduced, a flow balance guiding timing algorithm is applied in an area with small flow, in addition, cooperative control over signal lamps between the areas can be achieved, and efficient connection and balanced distribution of traffic flow between the areas are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic signal control, and in particular to a smart city traffic signal light management and control system. Background Art

[0002] With the acceleration of urbanization, urban traffic problems have become increasingly prominent. Traffic congestion and frequent traffic accidents have seriously affected citizens' travel experience and the overall operation efficiency of the city. In order to solve these problems, the concept of smart city came into being. Among them, the smart city traffic signal light control system, as a key link, is of great significance to improving the level of urban traffic management and alleviating traffic congestion.

[0003] There are shortcomings in traditional technologies. First, traditional systems are usually based on relatively simple traffic flow detection methods, such as a single induction coil or camera. These methods often cannot fully and accurately reflect the real-time changes in traffic flow, resulting in a lack of precise basis for the formulation of traffic light control strategies. Secondly, traditional systems often use a fixed method for regional division, which lacks flexibility and cannot be dynamically adjusted according to the actual situation of urban traffic flow.

[0004] To sum up, traditional technologies have many shortcomings in dealing with urban traffic congestion problems. Therefore, it is particularly important to develop a smart city traffic signal light control system. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide a smart city traffic signal light control system, which can monitor and analyze the changes in urban traffic flow in real time and accurately by comprehensively using a variety of advanced traffic flow detection technologies and accurate traffic flow status evaluation models. At the same time, the system should also have the ability to generate intelligent signal light control strategies and the ability to coordinate control of signal lights between regions to ensure efficient connection and balanced distribution of traffic flow between regions. Through the joint effect of these measures, the smoothness and efficiency of urban traffic can be significantly improved, the occurrence of traffic congestion can be reduced, and the travel experience of citizens can be enhanced.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a smart city traffic signal light control system, the system includes the following components: a regional division module, a traffic flow detection sensor, a regional traffic flow analysis module, a signal light control strategy generation module and an inter-regional signal light collaborative control module and a variable information sign, an information processing and generation module;

[0007] The regional division module: divides the urban road network into multiple traffic areas. This division is based on comprehensive judgment of multi-dimensional data, including but not limited to road infrastructure data, historical traffic flow big data, geographic information data and urban functional area distribution data. By constructing a regional division algorithm based on the combination of hierarchical analysis method AHP and fuzzy C-means clustering FCM, AHP is first used to determine the relative weight of each evaluation factor. The road infrastructure completeness weight w1 is determined based on expert scoring and statistical analysis of road width, road surface conditions, and traffic facilities. The historical traffic flow weight w2 is based on data analysis of the influence of traffic flow in different time periods and sections on regional division. The geographic information weight w3 considers the impact of topography and water system distribution on traffic flow. The urban functional area weight w4 is determined by analyzing the regional traffic characteristics of commercial areas, residential areas, and industrial areas. These weights are applied to the FCM algorithm. The formula is: Among them J m is the objective function, n is the number of data points, c is the number of cluster centers, u ij is the degree of membership of data point i to cluster j, m is the fuzzy index, x i is the feature vector of data point i, v j is the center vector of cluster j, and J is optimized by continuous iteration. m , until the convergence condition is met and the optimal region division result is obtained;

[0008] The traffic flow detection sensor is set in each traffic area to collect traffic flow data, including vehicle flow, vehicle speed, and vehicle type information. The vehicle flow detection adopts high-precision induction coil and microwave radar fusion technology. The induction coil is buried at a specific depth and position under the road surface. The magnetic field changes caused by the metal parts of the vehicle are detected based on the principle of electromagnetic induction. The microwave radar transmits microwave signals and receives reflected waves, and calculates the vehicle speed through the Doppler effect. The two work together to improve the accuracy of vehicle flow detection. Vehicle type identification uses deep learning image recognition technology and uses convolutional neural networks to extract features and classify vehicle images taken by the camera. The CNN network structure includes multiple convolutional layers, pooling layers, and fully connected layers. The convolution kernel size and number of the convolutional layer and the step size hyperparameters of the pooling layer are determined through training and optimization of a large number of vehicle image samples.

[0009] The regional traffic flow analysis module is connected to the traffic flow detection sensor and is used to analyze the traffic flow situation in each area. The analysis algorithm based on the combination of traffic flow dynamics and data mining technology is used to construct a traffic flow state evaluation model. The formula is: Where S represents the traffic flow state value, Q is the traffic flow, C is the theoretical road capacity, δ is the traffic flow impact weight, V is the average speed, V f is the free flow speed, ∈ is the speed influence weight, Tmix is the equivalent travel time of mixed vehicle types, T t otal is the total observation time, φ is the impact weight of vehicle type, and the model is used to comprehensively evaluate the regional traffic flow status and provide an accurate basis for the signal light control strategy;

[0010] The signal light control strategy generation module is connected to the regional traffic flow analysis module and is used to generate a signal light control strategy based on the analysis results. In congested areas with large traffic flow, a congestion relief dynamic timing algorithm is used. The formula is: Where G1 is the green light duration after adjustment in the congested area, G0 is the basic green light duration, k is the adjustment coefficient, S is the current traffic flow state value, S t is the congestion threshold state value, T c is the duration of the signal light cycle, and the number of signal light switching is reduced. The switching number adjustment formula is: N1 = N0-λ×(SS t ), where N1 is the adjusted switching number, N0 is the basic switching number, and λ is the switching number adjustment coefficient. In areas with low traffic, the flow balance guidance timing algorithm is used, and the formula is: Where G2 is the green light duration after adjustment in the area with low traffic volume, μ is the balance adjustment coefficient, S min is the minimum traffic flow state value in the area;

[0011] The inter-regional traffic light coordination control module is connected to the traffic light control strategy generation module and is used to coordinate the control of traffic lights within and between regions. An inter-regional traffic light coordination model is constructed based on graph theory and dynamic programming principles. Each traffic area is regarded as a node in the graph, and the connecting roads between regions are regarded as edges. The weight of the edge is determined by the road capacity, traffic flow and traffic light timing factors. The formula is: Where W is the weight of the connecting road between area i and area j, C is the road capacity, Q is the traffic flow, α is the impact weight of the capacity, T g is the green light duration from area i to area j, T c is the cycle length of the traffic light, β is the weight of the traffic light timing, and by solving the optimal path of the model, the coordinated switching strategy of traffic lights between regions is determined to achieve efficient connection and balanced distribution of traffic flow between regions;

[0012] The variable information sign is set at the intersection;

[0013] The information processing and generating module is connected to the traffic lights and the traffic flow detection sensor, and is used to generate guidance information according to the timing changes of the traffic lights and the real-time traffic conditions.

[0014] Furthermore, the induction coil in the traffic flow detection sensor adopts a double winding structure, the inner winding adopts a highly sensitive enameled copper wire, which is used to capture the initial signal of the weak magnetic field change when the vehicle passes by, and the outer winding adopts a high temperature resistant and corrosion resistant alloy wire, which is used to enhance the magnetic field signal strength and stabilize the detection signal. The ratio of the inner winding turns to the outer winding turns is determined according to the road type and the expected traffic volume. On the main road with heavy traffic volume and fast speed, the value is 1:3-1:5; on the branch road with less traffic volume, the value is 1:2-1:3. This double winding structure and the design of the turns ratio can effectively improve the accuracy and stability of traffic flow detection, reduce the influence of external electromagnetic interference, and ensure that traffic flow data can be accurately obtained in different traffic environments.

[0015] Furthermore, the calculation of the theoretical road capacity C in the traffic flow state assessment model in the regional traffic flow analysis module takes into account many factors. In addition to the basic parameters of the number of road lanes and width, the influence of road slope, curve curvature and surrounding interference factors is also included. By establishing a capacity calculation model based on multivariate linear regression: C = k1×L+k2×W+k3×G+k4×B+k5×P, where L is the number of lanes, W is the road width, G is the road slope, B is the curve curvature, P is the quantitative value of the surrounding interference factors, and k1-k5 are regression coefficients. These coefficients are obtained by analyzing a large number of The value of k1 is determined by collecting and analyzing the field traffic data of different types of roads, and the value of k2 is between 50-100 (vehicles / hour / meter), which is determined by considering the influence of road width on vehicle driving space and the value of k3, k4 and k5 are determined by the degree of obstruction to traffic flow caused by road slope, curve curvature and surrounding interference factors, respectively, and the value range is between -100 and 100, so as to realize the accurate calculation of theoretical road capacity and improve the accuracy of traffic flow status assessment.

[0016] Furthermore, the congestion threshold state value S in the congestion relief dynamic timing algorithm in the signal light control strategy generation module t The adaptive learning method is used to collect historical traffic flow data within a certain period of time, and the traffic flow status is divided into multiple levels such as smooth, light congestion, moderate congestion, and severe congestion by cluster analysis. Then, the traffic flow status value range of each level is calculated, and the lower limit value of the moderate congestion level is taken as the initial S t During the operation of the system, based on the feedback of actual traffic congestion, the incremental learning algorithm in machine learning is used to t Dynamic adjustment is performed. If a certain area appears multiple times in a specific period of time, the current S t If the traffic is judged as non-congested but the actual traffic is slow, the incremental learning algorithm will be used to appropriately increase St value to better adapt to the dynamic changes of urban traffic conditions and ensure the effectiveness and timeliness of congestion relief strategies.

[0017] Furthermore, the minimum traffic flow state value S in the area in the flow balance guidance timing algorithm in the signal light control strategy generation module min The calculation adopts the sliding window statistical method, setting a time window T w , the value is between 10 and 20 minutes, and the traffic flow status value in the area is counted in each time window, and the minimum value is taken as S in the window min As time goes by, the window keeps sliding and updating, and the latest S min At the same time, in order to avoid the influence of abnormal data, the median filtering technology is used to pre-process the traffic flow state values ​​obtained by statistics, remove data points that obviously deviate from the normal range, and improve S min The accuracy and reliability of the calculation can provide a precise basis for flow balance guidance and timing, and promote the rational distribution of traffic flow between regions.

[0018] Furthermore, the signal light cycle length T in the inter-regional traffic signal light coordination model in the inter-regional signal light coordination control module is c The determination adopts a hierarchical optimization strategy to determine the basic signal light cycle duration range according to the road grade and regional importance. The basic cycle duration of urban trunk roads In 100-150 seconds, secondary roads In 60-100 seconds, considering the real-time changes of traffic flow between regions, the fuzzy logic control algorithm is used to dynamically adjust the basic cycle duration. The input of the fuzzy logic controller is the traffic flow difference rate ΔQ between regions and the traffic flow saturation S within the region. r , the output is the cycle length adjustment ΔT c , by defining the fuzzy rule table, when ΔQ is large and S r High ΔT c Take a larger positive value to increase the duration of the traffic light cycle to relieve traffic pressure, ΔQ is small and S r When low, ΔT c Take a smaller negative value to shorten the traffic light cycle and improve traffic efficiency. After fuzzy reasoning and clarification processing, the final traffic light cycle duration T is obtained. c =T c0 +ΔT c , realizing intelligent optimization and coordinated control of traffic light cycle duration between areas.

[0019] Furthermore, when the information processing and generating module generates the induction information, the information generation of the road congestion ahead adopts a method based on the combination of traffic wave theory and data prediction model, and uses the traffic wave theory to analyze the propagation characteristics of traffic congestion and determine the propagation speed V of the congestion wave. w and congestion density ρ w The relationship between them is: Where V f is the free flow speed, ρ j In order to estimate the congestion density, based on the real-time traffic flow data obtained by the traffic flow detection sensor, the time series prediction model is used to predict the traffic flow changes in the future, and then the propagation range and intensity of the congestion wave are deduced. This information is converted into intuitive graphics or text information and displayed on the variable information sign. Different colored line segments are used to represent different degrees of congested sections. Red indicates severe congestion, yellow indicates moderate congestion, and green indicates smooth traffic. This provides drivers with accurate and timely information on road congestion ahead, helping them plan their routes in advance and avoid getting stuck in congested areas.

[0020] Furthermore, the variable information sign display control module dynamically adjusts the layout of the display content of the variable information sign, determines the size and position of the display area according to the importance and urgency of the inducing information, adopts a full-screen display mode for urgent traffic control information, and highlights it with the largest font and bright color to ensure that the driver can obtain the information in the first time. For recommended driving route information, it is displayed in the main display area of ​​the variable information sign in a combination of dynamic arrows and text. The direction of the arrow indicates the driving direction, and the text marks the road name and distance. For information on congestion conditions on the road ahead, it is scrolled in the form of a small window at the edge of the screen without affecting the display of the main information. At the same time, the display brightness is automatically adjusted according to the ambient light intensity, and a photosensitive sensor is used to detect the light intensity. When the light intensity is strong, the display brightness is increased to enhance visibility. When the light intensity is weak, the display brightness is reduced to reduce energy consumption and light pollution, thereby realizing intelligent layout of display content and adaptive brightness control, and improving the effectiveness and comfort of information transmission.

[0021] Furthermore, the system also includes a system fault diagnosis and self-repair module, which monitors the working status of each component of the system in real time. For traffic flow detection sensors, it determines whether they are working normally by sending detection signals of specific frequencies and receiving feedback signals. If the feedback signal is abnormal, the backup sensor is started, and the cause of the fault is analyzed using a fault diagnosis algorithm. The maintenance task is automatically dispatched according to the diagnosis result. For traffic light controllers, it monitors whether their signal output and control logic are normal. If a fault occurs, it switches to the backup controller and performs remote reset and program repair operations on the faulty controller. For variable information signs, it checks the integrity of the display content and the brightness uniformity index. If an abnormality is found, it automatically adjusts the display parameters or switches to the backup display module. Through this module, the system can achieve high-reliability operation, reduce the risk of traffic management interruption due to equipment failure, and improve the safety and stability of urban transportation.

[0022] Compared with the existing technology, this smart city traffic signal light control system has the following beneficial effects:

[0023] 1. This system can monitor and analyze urban traffic flow in real time and accurately by comprehensively using regional division algorithms, traffic flow detection technology and precise traffic flow status assessment models. According to the analysis results, the system can intelligently generate traffic light control strategies, including the use of congestion relief dynamic timing algorithms in congested areas to reduce the number of traffic light switching times, and the use of flow balance guidance timing algorithms in areas with less traffic. In addition, the system can also achieve coordinated control of traffic lights between regions to ensure efficient connection and balanced distribution of traffic flows between regions. The combined effect of these measures can significantly improve the smoothness and efficiency of urban traffic, reduce traffic congestion, and enhance the travel experience of citizens.

[0024] 2. By adopting adaptive learning methods and incremental learning algorithms, this system can dynamically adjust the congestion threshold status value and key parameters of the signal light cycle duration according to feedback from actual traffic congestion conditions, so as to better adapt to the dynamic changes in urban traffic conditions. At the same time, the system can also dynamically adjust the layout of the display content of variable information signs according to the importance and urgency of the inducing information, and automatically adjust the display brightness according to the ambient light intensity. These intelligent functions enable the system to provide more accurate, timely and humane traffic management services according to different scenarios and needs, thereby enhancing the overall efficiency of urban traffic management.

[0025] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 This is a process operation diagram of a smart city traffic signal light control system. DETAILED DESCRIPTION

[0028] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0029] Embodiment 1

[0030] This embodiment describes a city center business district with narrow roads, numerous shops, and extremely large pedestrian and vehicle flows. Especially during the morning and evening rush hours, traffic congestion is severe and the traffic composition is complex, including a large number of private cars, taxis, and delivery vehicles of different types.

[0031] The commercial district road infrastructure data (road width, road conditions are good, and traffic facilities are well-equipped), historical traffic flow big data (vehicle flow and pedestrian flow are extremely large during peak hours), geographic information data (flat terrain, but there are many parking lots and commercial building entrances around), and urban functional area distribution data (clearly commercial areas) are collected, and the weights are determined using the hierarchical analysis method. The road infrastructure completeness weight w1 is determined based on expert scoring and statistical analysis, combined with the actual situation of the commercial district roads. The historical traffic flow weight w2 is based on the analysis of the influence of peak-hour traffic on regional division; the geographic information weight w3 takes into account the impact of surrounding parking lots and building entrances, and the urban functional area weight w4 is determined based on the traffic characteristics of the commercial district. These weights are applied to the fuzzy C-means clustering algorithm. After iterative optimization, the commercial district road network is divided into multiple traffic areas, such as areas with large shopping malls and office buildings as the core.

[0032] Traffic flow detection sensors are installed at key intersections and sections in each traffic area, using high-precision induction coil and microwave radar fusion technology. High-sensitivity enameled copper wire is used for the inner winding of the induction coil on the main road, and high-temperature and corrosion-resistant alloy wire is used for the outer winding. The ratio of the inner winding turns to the outer winding turns is 1:4 based on the characteristics of large traffic volume and high speed on the main road, and 1:2 for the branch road.

[0033] The induction coil detects the changes in the magnetic field caused by the metal parts of the vehicle based on the principle of electromagnetic induction. The microwave radar transmits microwave signals and receives reflected waves, and calculates the vehicle speed through the Doppler effect. The two work together to accurately collect traffic flow and speed information. At the same time, the camera captures vehicle images and uses convolutional neural networks to identify vehicle types. The hyperparameters of the convolutional layer and pooling layer are optimized through training with a large number of vehicle image samples to ensure accurate identification of different types of vehicles such as cars, buses, and taxis.

[0034] The regional traffic flow analysis module calculates the traffic flow state value S using the traffic flow state evaluation model based on the data collected by the traffic flow detection sensor. The calculation of the theoretical road capacity C takes into account the number of road lanes (mostly two-way four-lane or six-lane), width (wider), slope (smaller), curve curvature (smaller) and surrounding interference factors (such as parking lot entrances and exits, pedestrians crossing). By establishing a capacity calculation model based on multivariate linear regression, the traffic flow impact weight δ, speed impact weight ε and vehicle type impact weight It is determined based on the traffic characteristics of the commercial area. For example, the weight of the impact of vehicle flow is relatively high. It is calculated that the S value of each area is higher during peak hours, indicating that the traffic flow is large and the congestion level is high.

[0035] For areas with severe congestion, such as intersections around shopping malls, a dynamic timing algorithm for congestion relief is used. The congestion threshold state value S is determined by the real-time calculated traffic flow state value S and the adaptive learning method. t (Initially take the lower limit of the moderate congestion level and dynamically adjust it according to the actual congestion feedback) By comparison, through the formula Increase the green light duration and use the formula N1=N0-λ×(SS t ) Reduce the number of signal light switching times to ease congestion. For example, the basic green light duration G0 at a certain intersection is 30 seconds, and the signal light cycle duration T c is 120 seconds, the adjustment coefficient k is 0.5, and the switching frequency adjustment coefficient λ is 0.3. When S is higher than S t The green light duration is appropriately increased and the number of switching times is reduced. For some areas with small traffic volume, the flow balance guidance timing algorithm is used to calculate the time of the green light by using the sliding window statistical method (time window T w Take 15 minutes) to calculate the minimum traffic flow state value S in the area min , and use median filtering technology to preprocess the data to avoid the influence of abnormal data, and then according to the formula Adjust the green light duration to balance the traffic flow between areas. For example, the basic green light duration G0 of a branch road area is 20 seconds, and the balance adjustment coefficient μ is 0.4. When the traffic volume in this area is small, the green light duration is appropriately reduced to guide vehicles to divert to other areas.

[0036] The traffic areas in the commercial district are regarded as nodes, and the connecting roads between the areas are regarded as edges. The inter-area traffic signal coordination model is constructed. The weight of the edge is determined by the road capacity (calculated according to the number of lanes and width), traffic flow (real-time detection) and signal timing factors. For example, a main road connects two important commercial areas. Its road capacity C is relatively high, and the traffic flow Q is relatively large during peak hours. According to the formula Calculate the weights, where the capacity impact weight α and the signal timing impact weight β are determined based on the importance of regional traffic.

[0037] By solving the optimal path of the collaborative model, the coordinated switching strategy of traffic lights between areas is determined. For example, during peak hours, in order to ensure smooth traffic between commercial areas, the main road and branch road traffic lights are coordinated so that vehicles can flow from one commercial area to another in an orderly manner, avoiding information processing and generation and the display of variable information signs.

[0038] The information processing and generation module generates induction information based on the changes in traffic light timing and real-time traffic conditions. For road congestion ahead, it uses traffic wave theory to analyze the congestion propagation characteristics, combined with real-time data obtained by traffic flow detection sensors, and uses a time series prediction model to predict the propagation range and intensity of the congestion wave. For example, it predicts that moderate congestion will occur 500 meters ahead on a main road, and the congestion wave propagation speed is 10 meters per minute.

[0039] The congestion information is converted into graphic or text information and displayed on a variable information sign. Moderately congested sections are represented by yellow lines, which are scrolled at the edge of the screen. The recommended driving route information is displayed in the main display area in a combination of dynamic arrows and text, such as instructing vehicles to avoid congested sections by taking a branch road, and marking the branch road name and distance. At the same time, the display brightness is automatically adjusted according to the ambient light intensity, increasing the brightness when the sunlight is strong and reducing the brightness at night or in dim light, thereby realizing intelligent layout of display content and adaptive brightness control, and preventing local congestion from causing traffic congestion between regions.

[0040] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A smart city traffic signal light control system, characterized in that: The system includes the following components: regional division module, traffic flow detection sensor, regional traffic flow analysis module, signal light control strategy generation module and inter-regional signal light coordination control module as well as variable information sign, information processing and generation module; The regional division module: divides the urban road network into multiple traffic areas. This division is based on comprehensive judgment of multi-dimensional data, including but not limited to road infrastructure data, historical traffic flow big data, geographic information data and urban functional area distribution data. By constructing a regional division algorithm based on the combination of hierarchical analysis method AHP and fuzzy C-means clustering FCM, AHP is first used to determine the relative weight of each evaluation factor. The road infrastructure completeness weight w1 is determined based on expert scoring and statistical analysis of road width, road surface conditions, and traffic facilities. The historical traffic flow weight w2 is based on data analysis of the influence of traffic flow in different time periods and sections on regional division. The geographic information weight w3 considers the impact of topography and water system distribution on traffic flow. The urban functional area weight w4 is determined by analyzing the regional traffic characteristics of commercial areas, residential areas, and industrial areas. These weights are applied to the FCM algorithm. The formula is: Among them J m is the objective function, n is the number of data points, c is the number of cluster centers, u ij is the degree of membership of data point i to cluster j, m is the fuzzy index, x i is the feature vector of data point i, v j is the center vector of cluster j, and J is optimized by continuous iteration. m , until the convergence condition is met and the optimal region division result is obtained; The traffic flow detection sensor is set in each traffic area to collect traffic flow data, including vehicle flow, vehicle speed, and vehicle type information. The vehicle flow detection adopts high-precision induction coil and microwave radar fusion technology. The induction coil is buried at a specific depth and position under the road surface. The magnetic field changes caused by the metal parts of the vehicle are detected based on the principle of electromagnetic induction. The microwave radar transmits microwave signals and receives reflected waves, and calculates the vehicle speed through the Doppler effect. The two work together to improve the accuracy of vehicle flow detection. Vehicle type identification uses deep learning image recognition technology and uses convolutional neural networks to extract features and classify vehicle images taken by the camera. The CNN network structure includes multiple convolutional layers, pooling layers, and fully connected layers. The convolution kernel size and number of the convolutional layer and the step size hyperparameters of the pooling layer are determined through training and optimization of a large number of vehicle image samples. The regional traffic flow analysis module is connected to the traffic flow detection sensor and is used to analyze the traffic flow situation in each area. The analysis algorithm based on the combination of traffic flow dynamics and data mining technology is used to construct a traffic flow state evaluation model. The formula is: Where S represents the traffic flow state value, Q is the traffic flow, C is the theoretical road capacity, δ is the traffic flow impact weight, V is the average speed, V f is the free flow speed, ∈ is the speed influence weight, T mix is the equivalent travel time of mixed vehicle types, T t otal is the total observation time, φ is the impact weight of vehicle type, and the model is used to comprehensively evaluate the regional traffic flow status and provide an accurate basis for the signal light control strategy; The signal light control strategy generation module is connected to the regional traffic flow analysis module and is used to generate a signal light control strategy based on the analysis results. In congested areas with large traffic flow, a congestion relief dynamic timing algorithm is used. The formula is: Where G1 is the green light duration after adjustment in the congested area, G0 is the basic green light duration, k is the adjustment coefficient, S is the current traffic flow state value, S t is the congestion threshold state value, T c is the duration of the signal light cycle, and the number of signal light switching is reduced. The switching number adjustment formula is: N1 = N0-λ×(SS t ), where N1 is the adjusted switching number, N0 is the basic switching number, and λ is the switching number adjustment coefficient. In areas with low traffic, the flow balance guidance timing algorithm is used, and the formula is: Where G2 is the green light duration after adjustment in the area with low traffic volume, μ is the balance adjustment coefficient, S min is the minimum traffic flow state value in the area; The inter-regional traffic light coordination control module is connected to the traffic light control strategy generation module and is used to coordinate the control of traffic lights within and between regions. An inter-regional traffic light coordination model is constructed based on graph theory and dynamic programming principles. Each traffic area is regarded as a node in the graph, and the connecting roads between regions are regarded as edges. The weight of the edge is determined by the road capacity, traffic flow and traffic light timing factors. The formula is: Where W is the weight of the connecting road between area i and area j, C is the road capacity, Q is the traffic flow, α is the impact weight of the capacity, T g is the green light duration from area i to area j, T c is the cycle length of the traffic light, β is the weight of the traffic light timing, and by solving the optimal path of the model, the coordinated switching strategy of traffic lights between regions is determined to achieve efficient connection and balanced distribution of traffic flow between regions; The variable information sign is set at the intersection; The information processing and generating module is connected to the traffic lights and the traffic flow detection sensor, and is used to generate guidance information according to the timing changes of the traffic lights and the real-time traffic conditions.

2. A smart city traffic signal light control system according to claim 1, characterized in that: The induction coil in the traffic flow detection sensor adopts a double winding structure, the inner winding adopts a highly sensitive enameled copper wire, which is used to capture the initial signal of the weak magnetic field change when the vehicle passes by, and the outer winding adopts a high temperature resistant and corrosion resistant alloy wire, which is used to enhance the magnetic field signal strength and stabilize the detection signal. The ratio of the number of turns of the inner winding to the number of turns of the outer winding is determined according to the road type and the expected traffic flow. On the main road with heavy traffic and fast speed, the value is 1:3-1:5; on the branch road with less traffic flow, the value is 1:2-1:

3.

3. A smart city traffic signal light control system according to claim 1, characterized in that: The calculation of the theoretical road capacity C in the traffic flow state assessment model in the regional traffic flow analysis module takes many factors into consideration. In addition to the basic parameters of the number of road lanes and width, the influence of road slope, curve curvature and surrounding interference factors is also included. A capacity calculation model based on multivariate linear regression is established: C=k1×L+k2×W+k3×G+k4×B+k5×P, where L is the number of lanes, W is the road width, G is the road slope, B is the curve curvature, P is the quantitative value of the surrounding interference factors, and k1-k5 are regression coefficients.

4. A smart city traffic signal light control system according to claim 1, characterized in that: The congestion threshold state value S in the congestion relief dynamic timing algorithm in the signal light control strategy generation module t The adaptive learning method is used to collect historical traffic flow data within a certain period of time, and the traffic flow status is divided into multiple levels such as smooth, light congestion, moderate congestion, and severe congestion by cluster analysis. Then, the traffic flow status value range of each level is calculated, and the lower limit value of the moderate congestion level is taken as the initial S t During the operation of the system, based on the feedback of actual traffic congestion, the incremental learning algorithm in machine learning is used to t Dynamic adjustment is performed. If a certain area appears multiple times in a specific period of time, the current S t If the traffic is judged as non-congested but the actual traffic is slow, the incremental learning algorithm will be used to appropriately increase S t The value of .

5. A smart city traffic signal light control system according to claim 1, characterized in that: The minimum traffic flow state value S in the area in the traffic balance guidance timing algorithm in the signal light control strategy generation module min The calculation adopts the sliding window statistical method, setting a time window T w , the value is between 10 and 20 minutes, and the traffic flow status value in the area is counted in each time window, and the minimum value is taken as S in the window min As time goes by, the window keeps sliding and updating, and the latest S min At the same time, in order to avoid the influence of abnormal data, the median filtering technology is used to pre-process the traffic flow state values ​​obtained by statistics, remove data points that obviously deviate from the normal range, and improve S min The accuracy and reliability of the calculation can provide a precise basis for flow balance guidance and timing, and promote the rational distribution of traffic flow between regions.

6. A smart city traffic signal light control system according to claim 1, characterized in that: The signal light cycle duration T in the inter-regional traffic signal light coordination model in the inter-regional signal light coordination control module c The determination adopts a hierarchical optimization strategy to determine the basic signal light cycle duration range according to the road grade and regional importance. The basic cycle duration T of urban trunk roads is c0main In 100-150 seconds, the secondary road T c0neondary In 60-100 seconds, considering the real-time changes of traffic flow between regions, the fuzzy logic control algorithm is used to dynamically adjust the basic cycle duration. The input of the fuzzy logic controller is the traffic flow difference rate ΔQ between regions and the traffic flow saturation S within the region. r , the output is the cycle length adjustment ΔT c , by defining the fuzzy rule table, when ΔQ is large and S r High ΔT c Take a larger positive value to increase the duration of the traffic light cycle to relieve traffic pressure, ΔQ is small and S r When low, ΔT c Take a smaller negative value to shorten the traffic light cycle and improve traffic efficiency. After fuzzy reasoning and clarification processing, the final traffic light cycle duration T is obtained. c =T c0 +ΔT c .

7. A smart city traffic signal light control system according to claim 1, characterized in that: When the information processing and generating module generates the induction information, the information generation of the road congestion ahead adopts a method based on the combination of traffic wave theory and data prediction model, and uses the traffic wave theory to analyze the propagation characteristics of traffic congestion and determine the propagation speed V of the congestion wave. w and congestion density ρ w The relationship between them is: Where V f is the free flow speed, ρ j In order to estimate the congestion density, based on the real-time traffic flow data obtained by the traffic flow detection sensor, the time series prediction model is used to predict the traffic flow changes in the future, and then the propagation range and intensity of the congestion wave are deduced. This information is converted into intuitive graphics or text information and displayed on the variable information sign. Different colored line segments are used to represent different degrees of congested sections. Red indicates severe congestion, yellow indicates moderate congestion, and green indicates smooth traffic. This provides drivers with accurate and timely information on road congestion ahead, helping them plan their routes in advance and avoid getting stuck in congested areas.

8. A smart city traffic signal light control system according to claim 1, characterized in that: The variable information sign display control module dynamically adjusts the layout of the display content of the variable information sign, determines the size and position of the display area according to the importance and urgency of the inductive information, adopts a full-screen display mode for urgent traffic control information, and highlights it with the largest font and bright color. For recommended driving route information, it is displayed in the main display area of ​​the variable information sign in a combination of dynamic arrows and text, with the arrow direction indicating the driving direction and the text marking the road name and distance. For information on congestion on the road ahead, it is scrolled in a small window at the edge of the screen without affecting the display of the main information. At the same time, the display brightness is automatically adjusted according to the ambient light intensity, and a photosensitive sensor is used to detect the light intensity. When the light intensity is strong, the display brightness is increased to enhance visibility. When the light intensity is weak, the display brightness is reduced to reduce energy consumption and light pollution.

9. A smart city traffic signal light control system according to claim 1, characterized in that: The system also includes a system fault diagnosis and self-repair module, which monitors the working status of each component of the system in real time. For traffic flow detection sensors, the module determines whether they are working normally by sending detection signals of specific frequencies and receiving feedback signals. If the feedback signal is abnormal, the backup sensor is started, and the cause of the fault is analyzed using a fault diagnosis algorithm. A maintenance task is automatically dispatched based on the diagnosis result. For traffic light controllers, the module monitors whether their signal output and control logic are normal. If a fault occurs, the module switches to the backup controller and performs remote reset and program repair operations on the faulty controller. For variable information signs, the module checks the display content integrity and brightness uniformity indicators. If an abnormality is found, the display parameters are automatically adjusted or the module switches to the backup display module.

Images