Control method and system for adjusting road tunnel illumination based on dynamic traffic

Abstract

The invention belongs to the technical field of road tunnel illumination, and particularly relates to a control method and system for adjusting road tunnel illumination based on dynamic traffic, which comprises the following steps: firstly, acquiring vehicle information through a high-definition camera array, a microwave radar detector and an illumination intensity sensor in a multi-source manner, then preprocessing the multi-source data, and applying an algorithm to adjust the illumination of a road tunnel; traffic flow statistics and trend prediction, vehicle driving state analysis and vehicle type accurate classification statistics are realized; based on this, the traffic flow is counted and predicted in real time through the dynamic traffic information acquisition and analysis module, the illumination brightness reference value is set and corrected based on the traffic flow by combining the multi-dimensional illumination parameter intelligent decision module, and then the illumination execution and dynamic adjustment module accurately executes the brightness instruction. The method has the advantages that the illumination brightness is dynamically matched with the traffic flow, the traffic safety requirement during high traffic flow is met, and energy waste during low traffic flow is avoided.

Description

Technical Field

[0001] This invention relates to the field of highway tunnel lighting technology, specifically to a control method and system for adjusting highway tunnel lighting based on dynamic traffic. Background Technology

[0002] As a special type of transportation infrastructure, highway tunnels have significantly different internal lighting environments compared to the external natural environment. The rationality of the lighting system directly affects driving safety, traffic efficiency, and energy consumption.

[0003] In actual operation, traffic conditions inside the tunnel are constantly changing, such as differences in traffic flow at different times (up to 800 vehicles per hour during the morning peak and only 200 vehicles per hour during the off-peak), fluctuations in vehicle speed (speeds as low as 10 km / h during congestion and up to 80 km / h during smooth traffic), the diversity of vehicle types (small cars, large trucks, buses, etc.) and the impact of special weather conditions (rainy days, foggy days) on light propagation inside the tunnel.

[0004] Currently, most traditional highway tunnel lighting control systems adopt fixed-mode control, that is, adjusting the lighting brightness according to preset time periods or simple external light intensity thresholds. This cannot make flexible and precise adjustments based on real-time dynamic traffic conditions inside the tunnel, making it difficult to achieve an effective balance between meeting driving safety requirements and reducing energy consumption. Specifically, it has the following problems:

[0005] 1. The lighting brightness is not matched with traffic flow, posing safety hazards and wasting energy.

[0006] Traditional control systems typically operate at fixed brightness levels. When traffic volume in a tunnel surges, this fixed brightness may not meet drivers' needs to discern road conditions and vehicle distances, increasing the risk of rear-end collisions, scrapes, and other traffic accidents. For example, during the morning rush hour, a two-way four-lane tunnel experienced traffic flow of 700 vehicles per hour with relatively small distances between vehicles. However, the traditional system maintained off-peak lighting levels, making it difficult for drivers to clearly observe the brake lights and turn signals of vehicles ahead, resulting in multiple rear-end collisions in that tunnel. Conversely, when traffic volume is extremely low (such as only 50 vehicles per hour in the early morning), the lighting system maintains high brightness, leading to significant energy waste. Statistics show that this type of energy waste accounts for more than 30% of the total energy consumption of tunnel lighting.

[0007] 2. Unable to dynamically adjust lighting response speed according to vehicle speed.

[0008] Drivers have varying perceptual needs regarding changes in tunnel lighting conditions at different driving speeds. At high speeds (e.g., 60-80 km / h), the lighting system needs to respond quickly to changes in light to ensure sufficient reaction time for the driver. At low speeds (e.g., 10-20 km / h in congested conditions), the lighting system does not require such a rapid response. However, current control systems have fixed lighting response speeds. For example, a tunnel lighting system might have a fixed response delay of 2 seconds. At high speeds, this 2-second delay means that when the driver enters the tunnel's transition zone between light and dark, the lighting brightness may not have adjusted in time, creating a "black hole effect" and affecting visual adaptation. Conversely, at low speeds, an excessively fast response speed can cause frequent fluctuations in lighting brightness, increasing driver fatigue.

[0009] 3. Lack of targeted lighting adjustments for different vehicle models.

[0010] Different vehicle types vary significantly in height, width, and headlight illumination range, resulting in different requirements for tunnel lighting brightness and uniformity. Large trucks, with their higher profiles, can obstruct smaller vehicles behind them, and their lower headlight angle necessitates higher brightness from the tunnel lighting system to ensure drivers of smaller vehicles can clearly see the road ahead. Conversely, small cars have relatively lower lighting requirements. Current control systems do not consider these vehicle differences, employing a uniform lighting standard. For example, a tunnel might provide 200 cd / m² of illumination for all vehicle types. This results in insufficient light for drivers of smaller vehicles behind large trucks, making it difficult to clearly identify road obstacles, while small cars experience over-illumination when driving alone.

[0011] 4. Insufficient early warning and emergency response capabilities for lighting system malfunctions.

[0012] Tunnel lighting systems consist of numerous lamps. During long-term operation, these lamps may age, break down, or malfunction, leading to reduced brightness or complete darkness in certain areas. Current control systems lack effective fault warning mechanisms, typically requiring regular inspections by staff to detect faults, resulting in delayed discovery. For example, in one tunnel, 10 lamps failed due to wiring faults, causing the brightness of that section to drop to 50 cd / m², below the national standard of 100 cd / m². However, the control system failed to issue any warning signals, and the issue was only discovered during a staff inspection three days later, posing a significant safety hazard during that period. Furthermore, emergency lighting switching is slow in the event of sudden malfunctions. For instance, in one tunnel where the main lighting system experienced a power outage, the emergency lighting system only activated after 5 seconds, causing temporary darkness and resulting in traffic congestion and driver panic.

[0013] Based on the above, a control method and system for regulating highway tunnel lighting based on dynamic traffic is invented. Summary of the Invention

[0014] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0015] The control method for regulating highway tunnel lighting based on dynamic traffic includes the following specific steps:

[0016] S1: First, vehicle information is collected from multiple sources using a high-definition camera array, microwave radar detector, and light intensity sensor. Then, after preprocessing the multi-source data, algorithms are used to achieve traffic flow statistics and trend prediction, vehicle driving status analysis, and accurate classification statistics of vehicle types.

[0017] S2: First, set a brightness baseline value based on traffic flow and then adjust it in conjunction with light intensity; then optimize the lighting uniformity according to vehicle type distribution and vehicle position; then set and adjust the lighting response speed according to vehicle speed; at the same time, develop exclusive lighting strategies for special scenarios to ensure that lighting parameters are adapted to real-time traffic conditions.

[0018] S3: First, through multi-channel control, industrial Ethernet communication and PWM dimming, the lights in each zone are precisely controlled; then the tunnel is divided into five functional zones and the zone brightness command is executed; at the same time, the lighting effect is monitored in real time with the help of the brightness sensor and the uniformity is calculated; then, through dynamic feedback adjustment, the brightness deviation is corrected and the response speed is optimized, forming a command-execution-monitoring-correction closed loop to ensure that the lighting effect meets the standard.

[0019] S4: First, collect lighting parameters and monitor lighting status through the intelligent controller; then, use the fault tree algorithm to diagnose the fault type and trigger a three-level warning according to the degree of impact; then, when a fault occurs, switch emergency lighting and adjust it in zones, linking the ventilation and traffic signal systems; at the same time, connect with the operation and maintenance and traffic control platform to schedule operation and maintenance resources, propose control suggestions, verify the effect after the fault is repaired and archive it to reduce the risk and impact of the fault.

[0020] As a preferred embodiment of the control method for regulating highway tunnel lighting based on dynamic traffic according to the present invention, the specific steps of S1 are as follows:

[0021] S11: First, a high-definition camera array is used to capture the dynamic information of vehicles inside the tunnel in real time; then, a microwave radar detector is used to accurately detect the speed and position of vehicles even in bad weather or low light conditions; finally, a light intensity sensor is used to collect data on the natural light intensity inside and outside the tunnel in real time.

[0022] S12: First, the Kalman filter algorithm is used to reduce noise in the collected raw data; then, the data in different formats collected by the high-definition camera, microwave radar detector, and light intensity sensor are converted into a unified JSON format to facilitate subsequent data transmission and analysis.

[0023] S13: First, based on the preprocessed camera data, target detection and tracking algorithms are used to count the number of vehicles passing through the tunnel per unit time to obtain real-time traffic flow data; then, time series analysis algorithms are used to predict the traffic flow trend in the next 10 minutes based on the real-time traffic flow data of the past 30 minutes.

[0024] S14: First, perform statistical analysis on the driving speed data of all vehicles collected, calculate the average speed, maximum speed, minimum speed of vehicles in the tunnel, and the proportion of vehicles in different speed ranges; then, based on the vehicle position data collected by the camera, calculate the distance between adjacent vehicles and compare it with the safe distance standard to analyze the current safety level of vehicle driving in the tunnel.

[0025] S15: First, based on the vehicle image data collected by the camera, a deep learning image classification algorithm is used to accurately classify the vehicles into small cars, medium-sized buses, large trucks, and special vehicles; then, the number of each type of vehicle and its proportion passing through the tunnel per unit time are counted.

[0026] As a preferred embodiment of the control method for regulating highway tunnel lighting based on dynamic traffic according to the present invention, the specific steps of S2 are as follows:

[0027] S21: First, establish a mapping model between traffic flow and lighting brightness benchmark values ​​to determine the lighting brightness benchmark values ​​for the current and future periods based on real-time traffic flow data and traffic trend prediction data; then, correct the brightness benchmark values ​​based on the light intensity data collected by the light intensity sensor inside and outside the tunnel.

[0028] S22: First, based on the vehicle type ratio data, optimize the lighting uniformity to meet the lighting needs of different vehicle types; then, based on the vehicle location data, if a certain area is found to have a high vehicle density, automatically increase the brightness of the surrounding lights in that area to compensate for the local lighting uniformity and ensure uniform brightness distribution in the vehicle driving area.

[0029] S23: First, establish a correlation model between the average vehicle speed and the lighting response speed; then, based on the vehicle speed distribution data, if it is found that there are large differences in vehicle speed within the tunnel, the response speed corresponding to the speed range with the highest proportion is used as the basis for correction, taking into account the lighting needs of vehicles with different speeds.

[0030] S24: When severe weather is detected in the tunnel via camera or external meteorological data interface, the special weather lighting mode is automatically activated; then, when a special vehicle is detected entering the tunnel, a priority passage route is automatically planned for the special vehicle, and the lighting brightness on that route is adjusted.

[0031] As a preferred embodiment of the control method for regulating highway tunnel lighting based on dynamic traffic according to the present invention, the specific steps of S3 are as follows:

[0032] S31: First, each intelligent lighting controller has at least 16 independent control channels to control different areas and different types of lights in the tunnel, achieving precise control of different lights; then, industrial Ethernet is used for data communication to receive lighting parameter commands; at the same time, the operating status data of the lights is fed back to form a data closed loop; then, the PWM dimming module is used to continuously adjust the power of the lights according to the brightness command.

[0033] S32: First, the tunnel is divided into five areas according to function and location: entrance transition area, entrance section, middle section, exit section, and exit transition area. Each area is equipped with an independent lighting control circuit, which is controlled by an intelligent lighting controller. Then, the brightness of the lights in each area is controlled according to the zoning brightness instructions issued by the intelligent lighting controller.

[0034] S33: First, based on the brightness sensor, monitor the actual lighting brightness of each area in real time and transmit the monitoring data to the intelligent lighting controller; then, based on the brightness data collected by each brightness sensor, calculate the lighting uniformity of each area in real time and compare it with the set uniformity standard. If the uniformity is found to be unsatisfactory, an adjustment signal is generated.

[0035] S34: First, compare the actual brightness collected with the set target brightness, calculate the brightness deviation value, and provide feedback; then, based on the feedback brightness adjustment process data, analyze the actual response speed of the lighting system and compare it with the set response speed; if the actual response speed is greater than the set value, optimize the PWM dimming frequency of the intelligent lighting controller to shorten the response time; if the actual response speed is too small, causing brightness fluctuations, appropriately reduce the dimming frequency to optimize the response speed.

[0036] As a preferred embodiment of the control method for regulating highway tunnel lighting based on dynamic traffic according to the present invention, the specific steps of S4 are as follows:

[0037] S41: First, the operating parameters of each lamp are collected through the intelligent lighting controller to fully understand the operating status of the lamp; then, the normal threshold range of each parameter is set so that when the parameter of a lamp exceeds the threshold range, the lamp is immediately marked as abnormal and the abnormal information is recorded.

[0038] S42: First, based on the collected abnormal parameters, the fault tree analysis algorithm is used to accurately diagnose the fault type; then, a three-level early warning mechanism is set according to the scope and severity of the fault; then, machine learning algorithms are used to predict the probability of the lamp's failure in the next 7 days based on historical fault data and real-time monitoring parameters.

[0039] S43: When a Level 3 fault or sudden power outage is detected in the main lighting system, the emergency lighting system can be immediately activated through the dual power supply automatic switching switch to ensure that there are no dark intervals in the tunnel; then, the emergency lighting will be zoned and the brightness will be optimized according to the location of the fault area; after that, when the emergency lighting is activated, the tunnel ventilation system and traffic signal system can be automatically linked.

[0040] S44: First, it connects with the GPS dispatch platform of the maintenance team to automatically match the maintenance vehicle and personnel closest to the tunnel after receiving a fault warning, generate the optimal maintenance route, and synchronize the fault location, fault type, and required spare parts information to the maintenance personnel's mobile terminal to improve maintenance efficiency; then, in the event of a level 3 fault, it will automatically send traffic control suggestions to the traffic management department's command platform; after the maintenance personnel complete the fault repair and upload photos and test data of the repair completion through the mobile APP, it can automatically collect the repaired data for verification, confirm that the lighting parameters have returned to normal, turn off the emergency lighting system, switch back to the main lighting mode, archive the fault handling results to the database, and update the equipment maintenance record.

[0041] A control system for regulating highway tunnel lighting based on dynamic traffic includes:

[0042] The dynamic traffic information collection and analysis module is used to first collect vehicle information from multiple sources, including a high-definition camera array, a microwave radar detector, and a light intensity sensor. Then, after preprocessing the multi-source data, algorithms are used to realize traffic flow statistics and trend prediction, vehicle driving status analysis, and accurate classification statistics of vehicle types.

[0043] The multi-dimensional lighting parameter intelligent decision-making module is used to first set the brightness benchmark value based on traffic flow and then combine it with light intensity correction; then optimize the lighting uniformity according to vehicle type distribution and vehicle position; then set and correct the lighting response speed according to vehicle speed; and at the same time, formulate exclusive lighting strategies for special scenarios to ensure that the lighting parameters are adapted to real-time traffic conditions.

[0044] The lighting execution and dynamic adjustment module is used to first precisely control the lighting fixtures in each zone through multi-channel control, industrial Ethernet communication and PWM dimming; then it divides the tunnel into five functional zones and executes the zone brightness commands; at the same time, it uses brightness sensors to monitor the lighting effect in real time and calculate the uniformity; then it uses dynamic feedback adjustment to correct brightness deviation and optimize response speed, forming a command-execution-monitoring-correction closed loop to ensure that the lighting effect meets the standards.

[0045] The fault warning and emergency handling module is used to first collect lighting parameters and monitor the status of the lighting through the intelligent controller; then, it uses a fault tree algorithm to diagnose the fault type and trigger a three-level warning according to the degree of impact; then, when a fault occurs, it switches emergency lighting and adjusts it in different zones, and links the ventilation and traffic signal systems; at the same time, it connects to the operation and maintenance and traffic control platform to schedule operation and maintenance resources, propose control suggestions, verify the effect after the fault is repaired and archive it, thereby reducing the risk and impact of the fault.

[0046] As a preferred embodiment of the control system for regulating highway tunnel lighting based on dynamic traffic as described in this invention, the dynamic traffic information acquisition and analysis module includes:

[0047] The multi-source traffic data acquisition unit is used to first capture the dynamic information of vehicles in the tunnel in real time based on a high-definition camera array; then, based on a microwave radar detector, it can accurately detect the speed and position of vehicles even in bad weather or low light conditions; and finally, based on a light intensity sensor, it collects the natural light intensity data inside and outside the tunnel in real time.

[0048] The data preprocessing unit first uses the Kalman filter algorithm to reduce noise in the collected raw data; then it converts the data in different formats collected by the high-definition camera, microwave radar detector, and light intensity sensor into a unified JSON format to facilitate subsequent data transmission and analysis.

[0049] The traffic flow dynamic analysis unit first uses target detection and tracking algorithms to count the number of vehicles passing through the tunnel per unit time based on preprocessed camera data, thus obtaining real-time traffic flow data; then, it uses time series analysis algorithms to predict the traffic flow trend in the next 10 minutes based on the real-time traffic flow data of the past 30 minutes.

[0050] The vehicle driving status analysis unit is used to first perform statistical analysis on the driving speed data of all vehicles collected, and calculate the average speed, maximum speed, minimum speed and the proportion of vehicles in different speed ranges in the tunnel; then, based on the vehicle position data collected by the camera, it calculates the distance between adjacent vehicles and compares it with the safe distance standard to analyze the safety level of vehicle driving in the tunnel.

[0051] The vehicle classification and statistics unit is used to first accurately classify vehicles based on the vehicle image data collected by the camera, using a deep learning image classification algorithm, into small cars, medium-sized buses, large trucks, and special vehicles; then it counts the number of each type of vehicle and its proportion passing through the tunnel per unit time.

[0052] As a preferred embodiment of the control system for regulating highway tunnel lighting based on dynamic traffic as described in this invention, the multi-dimensional lighting parameter intelligent decision-making module includes:

[0053] The lighting brightness reference value determination unit is used to first establish a mapping relationship model between traffic flow and lighting brightness reference value, and then determine the lighting brightness reference value for the current period and a future period based on the real-time traffic flow data and traffic trend prediction data output by the traffic flow dynamic analysis unit; then, the brightness reference value is corrected based on the light intensity data inside and outside the tunnel collected by the light intensity sensor.

[0054] The lighting uniformity optimization unit first optimizes the lighting uniformity based on the vehicle proportion data output by the vehicle classification statistics unit to meet the lighting needs of different vehicle models. Then, based on the vehicle location data output by the vehicle driving status analysis unit, if a certain area has a high vehicle density, it automatically increases the brightness of the surrounding lights in that area to compensate for the local lighting uniformity and ensure uniform brightness distribution within the vehicle driving area.

[0055] The lighting response speed adjustment unit is used to first establish a correlation model between the average vehicle speed and the lighting response speed; then, based on the vehicle speed distribution data output by the vehicle driving status analysis unit, if it is found that there are large differences in vehicle speed within the tunnel, the response speed corresponding to the speed range with the highest proportion is used as the basis for adjustment, taking into account the lighting needs of vehicles with different speeds.

[0056] The special scene lighting decision unit is used to automatically activate the special weather lighting mode when severe weather is detected in the tunnel through a camera or external meteorological data interface; then, when the vehicle type classification and statistics unit identifies a special vehicle entering the tunnel, it automatically plans a priority passage route for the special vehicle and corrects the lighting brightness on that route.

[0057] As a preferred embodiment of the control system for regulating highway tunnel lighting based on dynamic traffic as described in this invention, the lighting execution and dynamic adjustment module includes:

[0058] The intelligent lighting controller unit is designed to first enable each intelligent lighting controller to have at least 16 independent control channels to control different areas and different types of lighting fixtures within the tunnel, achieving precise control of different lighting fixtures. Next, it uses industrial Ethernet to communicate with the multi-dimensional lighting parameter intelligent decision-making module to receive lighting parameter commands. Simultaneously, it feeds back the operating status data of the lighting fixtures to the multi-dimensional lighting parameter intelligent decision-making module, forming a data closed loop. Then, using a PWM dimming module, it continuously adjusts the power of the lighting fixtures according to brightness commands, ensuring that the lighting brightness accurately reaches the target value set by the multi-dimensional lighting parameter intelligent decision-making module.

[0059] The zoned lighting execution unit is used to first divide the tunnel into five zones according to function and location: entrance transition zone, entrance section, middle section, exit section, and exit transition zone. Each zone is equipped with an independent lighting control loop, which is controlled by an intelligent lighting controller. Then, according to the zoned brightness instructions issued by the intelligent lighting controller, the brightness of the lights in each zone is controlled.

[0060] The real-time lighting effect monitoring unit first monitors the actual lighting brightness of each area in real time based on the brightness sensor and transmits the monitoring data to the intelligent lighting controller. Then, based on the brightness data collected by each brightness sensor, it calculates the lighting uniformity of each area in real time and compares it with the uniformity standard set by the multi-dimensional lighting parameter intelligent decision module. If the uniformity is found to be unsatisfactory, an adjustment signal is generated.

[0061] The dynamic feedback adjustment unit first compares the actual brightness collected by the real-time lighting effect monitoring unit with the target brightness set by the multi-dimensional lighting parameter intelligent decision-making module, calculates the brightness deviation value, and feeds it back. Then, based on the brightness adjustment process data fed back by the real-time lighting effect monitoring unit, it analyzes the actual response speed of the lighting system and compares it with the response speed set by the multi-dimensional lighting parameter intelligent decision-making module. If the actual response speed is greater than the set value, the PWM dimming frequency of the intelligent lighting controller is optimized to shorten the response time. If the actual response speed is too small, causing brightness fluctuations, the dimming frequency is appropriately reduced to optimize the response speed.

[0062] As a preferred embodiment of the control system for regulating highway tunnel lighting based on dynamic traffic according to the present invention, the fault early warning and emergency handling module includes:

[0063] The lighting status monitoring unit is used to first collect the operating parameters of each lighting fixture through the intelligent lighting controller to fully understand the operating status of the lighting fixture; then, it sets the normal threshold range for each parameter so that when the parameter of a certain lighting fixture exceeds the threshold range, the lighting fixture is immediately marked as abnormal and the abnormal information is recorded.

[0064] The fault early warning analysis unit is used to first accurately diagnose the fault type based on the abnormal parameters collected by the lighting fixture status monitoring unit and the fault tree analysis algorithm; then, a three-level early warning mechanism is set according to the scope and severity of the fault; and then, machine learning algorithms are used to predict the probability of the lighting fixture's failure in the next 7 days based on historical fault data and real-time monitoring parameters.

[0065] The emergency lighting control unit is used to immediately activate the emergency lighting system through a dual-power automatic switching switch when a level 3 fault or sudden power outage is detected in the main lighting system, ensuring that there are no dark intervals in the tunnel; then, it will optimize the brightness of the emergency lighting according to the location of the fault area; after that, when the emergency lighting is activated, it can automatically link the tunnel ventilation system and traffic signal system.

[0066] The fault handling linkage unit first connects with the maintenance team's GPS dispatch platform to automatically match the nearest maintenance vehicle and personnel to the tunnel upon receiving a fault warning, generate the optimal maintenance route, and synchronize the fault location, fault type, and required spare parts information to the maintenance personnel's mobile devices to improve maintenance efficiency. Then, in the event of a level 3 fault, it automatically sends traffic control suggestions to the traffic management department's command platform. After the maintenance personnel complete the fault repair and upload photos and test data of the repair completion via the mobile APP, it can automatically collect the repaired data for verification. Once it confirms that the lighting parameters have returned to normal, it shuts down the emergency lighting system, switches back to the main lighting mode, archives the fault handling results to the database, and updates the equipment maintenance records.

[0067] Compared with existing technologies:

[0068] 1. The system uses a dynamic traffic information collection and analysis module to collect and predict traffic flow in real time. Combined with a multi-dimensional lighting parameter intelligent decision-making module, it sets and corrects the lighting brightness benchmark value based on traffic flow. Then, the lighting execution and dynamic adjustment module accurately executes the brightness command. This system has the advantages of dynamically adapting lighting brightness to traffic flow, meeting driving safety requirements during high traffic flow, and avoiding energy waste during low traffic flow.

[0069] 2. By analyzing the average vehicle speed and speed distribution through the dynamic traffic information collection and analysis module, and establishing a correlation model based on vehicle speed and dynamically correcting the lighting response speed through the multi-dimensional lighting parameter intelligent decision module, and further optimizing the response speed execution process through the lighting execution and dynamic adjustment module, it has the advantages of flexibly adjusting the lighting response speed according to the vehicle speed, taking into account the visual response requirements at high speeds and the visual comfort at low speeds.

[0070] 3. The system accurately identifies vehicle types and calculates their proportions through a dynamic traffic information collection and analysis module. It optimizes lighting uniformity based on vehicle type distribution and adjusts lighting strategies according to the needs of different vehicle types through a multi-dimensional lighting parameter intelligent decision-making module. Furthermore, it implements zoned lighting according to the strategy through a lighting execution and dynamic adjustment module. This system has the advantages of providing targeted lighting based on vehicle type differences and meeting the personalized needs of drivers of different vehicle types for lighting brightness and uniformity.

[0071] 4. The fault warning and emergency handling module monitors the operating parameters of the lighting fixtures in real time, diagnoses the fault type and triggers graded warnings, and quickly switches emergency lighting and links with the ventilation and traffic signal systems. At the same time, it connects with the operation and maintenance and traffic control platform, which has the advantages of enabling early warning of lighting system faults, rapid activation of emergency handling measures and reducing the impact of faults on driving safety. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of the overall framework of the present invention;

[0073] Figure 2 This is a schematic diagram of the framework of the dynamic traffic information collection and analysis module of the present invention;

[0074] Figure 3 This is a schematic diagram of the framework of the intelligent decision-making module for multi-dimensional lighting parameters of the present invention;

[0075] Figure 4 This is a schematic diagram of the lighting execution and dynamic adjustment module framework of the present invention;

[0076] Figure 5 This is a schematic diagram of the fault warning and emergency handling module framework of the present invention. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0078] This invention provides a control method for regulating highway tunnel lighting based on dynamic traffic. Please refer to [link / reference]. Figures 1-4 The specific steps are as follows:

[0079] S1: First, vehicle information is collected from multiple sources using a high-definition camera array, microwave radar detector, and light intensity sensor. Then, after preprocessing the multi-source data, algorithms are used to achieve traffic flow statistics and trend prediction, vehicle driving status analysis, and accurate classification statistics of vehicle types.

[0080] The specific steps of S1 are as follows:

[0081] S11: First, a high-definition camera array is used to capture the dynamic information of vehicles inside the tunnel in real time; then, a microwave radar detector is used to accurately detect the speed and position of vehicles even in adverse weather conditions (rain, fog) or low light conditions; then, a light intensity sensor is used to collect data on the natural light intensity inside and outside the tunnel in real time.

[0082] S12: First, the Kalman filter algorithm is used to denoise the raw data such as traffic flow, vehicle speed, and light intensity, removing abnormal data caused by equipment vibration and external interference (such as electromagnetic waves in the tunnel). For example, if the vehicle speed data collected by the camera suddenly spikes to 150km / h (far exceeding the tunnel speed limit of 80km / h), the system determines that the data is an outlier, removes it using the Kalman filter algorithm, and smooths it using normal data from before and after. Then, the data in different formats (such as video streams, analog signals, and digital signals) collected by the high-definition camera, microwave radar detector, and light intensity sensor are converted into a unified JSON format for easy data transmission and analysis in the future.

[0083] S13: First, based on the preprocessed camera data, target detection and tracking algorithms (such as YOLOv8 algorithm) are used to count the number of vehicles passing through the tunnel per unit time (1 minute as a statistical period) to obtain real-time traffic flow data; then, time series analysis algorithms (such as LSTM neural network) are used to predict the traffic flow trend in the next 10 minutes based on the real-time traffic flow data of the past 30 minutes. For example, it is predicted that the traffic flow will increase from the current 300 vehicles / hour to 500 vehicles / hour in the next 10 minutes, providing support for adjusting lighting parameters in advance.

[0084] S14: First, statistically analyze the collected vehicle speed data to calculate the average speed, maximum speed, minimum speed, and vehicle percentage in different speed ranges (e.g., 0-20km / h, 20-40km / h, 40-60km / h, 60-80km / h). For example, if the analysis shows that the current average speed in the tunnel is 30km / h, and the vehicles in the 20-40km / h range account for 70%, it is determined that the tunnel is currently congested. Next, based on the vehicle position data collected by the camera, calculate the distance between adjacent vehicles (head-to-head distance) and compare it with the safe distance standard (e.g., a safe distance of 30 meters at a speed of 30km / h) to analyze the safety level of vehicle travel in the tunnel. If multiple groups of vehicles are found to have distances less than the safe distance, it is determined that there is a risk of rear-end collision.

[0085] S15: First, based on the vehicle image data collected by the camera, a deep learning image classification algorithm (such as the ResNet50 algorithm) is used to accurately classify the vehicles into small cars (length < 6 meters, width < 1.8 meters), medium-sized buses (6 meters ≤ length < 10 meters, 1.8 meters ≤ width < 2.5 meters), large trucks (length ≥ 10 meters, width ≥ 2.5 meters), and special vehicles (such as fire trucks and ambulances), with an accuracy rate of no less than 95%. Then, the number and proportion of each type of vehicle passing through the tunnel within a unit of time (5 minutes as a statistical period) are counted. For example, if the count shows that there are 20 small cars, 5 medium-sized buses, and 3 large trucks in the current 5 minutes, the proportions of the vehicle types are 74%, 18.5%, and 7.5%, respectively, providing a basis for targeted lighting adjustments.

[0086] S2: First, set a brightness baseline value based on traffic flow and then adjust it in conjunction with light intensity; then optimize the lighting uniformity according to vehicle type distribution and vehicle position; then set and adjust the lighting response speed according to vehicle speed; at the same time, develop exclusive lighting strategies for special scenarios to ensure that lighting parameters are adapted to real-time traffic conditions.

[0087] The specific steps of S2 are as follows:

[0088] S21: First, establish a mapping model between traffic flow and lighting brightness benchmark values. For example, when traffic flow is <200 vehicles / hour, the lighting brightness benchmark value is set to 100 cd / m²; when 200 vehicles / hour ≤ traffic flow < 500 vehicles / hour, the benchmark value is set to 150 cd / m²; when traffic flow is ≥500 vehicles / hour, the benchmark value is set to 200 cd / m². Based on real-time traffic flow data and traffic trend prediction data, determine the lighting brightness benchmark values ​​for the current and future periods. Then, based on... Based on the light intensity data collected by the light intensity sensor inside and outside the tunnel, the brightness benchmark value is corrected. For example, when the light intensity outside the tunnel is strong (such as at noon on a sunny summer day, the light intensity > 10000 cd / ㎡), in order to avoid the "black hole effect" when drivers enter the tunnel, the brightness benchmark value of the entrance section is increased by 20%; when the light intensity outside the tunnel is weak (such as on a cloudy day or at night, the light intensity < 1000 cd / ㎡), the brightness benchmark value of the entrance section is decreased by 10%.

[0089] S22: First, based on the vehicle type proportion data, optimize the lighting uniformity to meet the lighting needs of different vehicle types. When the proportion of large trucks is >10%, since large trucks will block light, the lighting uniformity in the lower part of the tunnel (height 1-3 meters) needs to be improved, requiring the ratio of maximum brightness to minimum brightness in this area to be ≤1.5. When the proportion of small cars is >80%, focus on ensuring the lighting uniformity in the middle part of the tunnel (height 0.5-2 meters), with a ratio ≤2.0. Then, based on the vehicle location data, if a large vehicle density is found in a certain area (e.g., more than 10 vehicles within 50 meters), the brightness of the surrounding lights in that area will be automatically increased to compensate for the local lighting uniformity and ensure uniform brightness distribution in the vehicle driving area.

[0090] S23: First, establish a correlation model between average vehicle speed and lighting response speed. For example, when the average vehicle speed is <30km / h (congested state), the lighting response speed is set to 500ms to avoid frequent fluctuations in brightness; when 30km / h ≤ average vehicle speed <60km / h, the response speed is set to 300ms; when the average vehicle speed is ≥60km / h, the response speed is set to 100ms to ensure that the lighting brightness quickly follows changes in light and meets the driver's visual needs. Then, based on the vehicle speed distribution data, if it is found that there are large differences in vehicle speed within the tunnel (such as some vehicles traveling at 60km / h and some vehicles traveling at 20km / h), the response speed corresponding to the highest proportion of vehicle speed range is used as the basis for a ±50ms correction to take into account the lighting needs of vehicles at different speeds.

[0091] S24: When severe weather such as rain or fog is detected in the tunnel via a camera or external meteorological data interface, a special weather lighting mode is automatically activated. For example, in rainy weather, due to the reduced reflectivity of the road surface, the overall lighting brightness in the tunnel is increased by 30% from the baseline value, and the contour lights on the tunnel sidewalls are turned on to enhance the visibility of the tunnel boundaries. In foggy weather, the lighting brightness is reduced (to avoid strong light scattering affecting visibility), the baseline value is reduced by 20%, and yellow fog lights are turned on to improve light penetration. Then, when special vehicles (such as fire trucks and ambulances) are detected entering the tunnel, a priority passage route is automatically planned for the special vehicles, and the lighting brightness on that route is corrected, such as increasing it to 250 cd / m², while the guide lights (red or blue) on both sides of the route are turned on to guide the special vehicles to pass quickly and ensure the efficiency of emergency rescue.

[0092] S3: First, through multi-channel control, industrial Ethernet communication and PWM dimming, the lights in each zone are precisely controlled; then the tunnel is divided into five functional zones and the zone brightness command is executed; at the same time, the lighting effect is monitored in real time with the help of the brightness sensor and the uniformity is calculated; then, through dynamic feedback adjustment, the brightness deviation is corrected and the response speed is optimized, forming a command-execution-monitoring-correction closed loop to ensure that the lighting effect meets the standard.

[0093] The specific steps of S3 are as follows:

[0094] S31: First, each intelligent lighting controller has at least 16 independent control channels to control different areas and types of lights within the tunnel (such as entrance lights, middle lights, exit lights, contour lights, and fog lights), achieving precise control of different lights. Next, industrial Ethernet is used for data communication to receive lighting parameter commands (such as brightness values, response speed, and light switch status), with a communication delay ≤50ms. Simultaneously, the operating status data of the lights (such as current brightness, power, and temperature) is fed back to form a data closed loop. Then, a PWM (Pulse Width Modulation) dimming module is used to continuously adjust the power of the lights according to the brightness commands, with a dimming range of 0-100% and an adjustment accuracy of 1%.

[0095] S32: First, the tunnel is divided into five zones according to function and location: entrance transition zone, entrance section, middle section, exit section, and exit transition zone. Each zone has an independent lighting control loop, controlled by an intelligent lighting controller. For example, in a 1000-meter-long tunnel, the entrance transition zone is 0-50 meters, the entrance section is 50-200 meters, the middle section is 200-800 meters, the exit section is 800-950 meters, and the exit transition zone is 950-1000 meters. Then, based on the zoning brightness instructions issued by the intelligent lighting controller, the brightness of the lights in each zone is controlled. For example, the brightness instruction for the entrance transition zone is 300 cd / m², the entrance section is 200 cd / m², the middle section is 150 cd / m², the exit section is 200 cd / m², and the exit transition zone is 300 cd / m². The zoning lighting execution unit then adjusts the brightness of the lights in each zone to the corresponding values ​​to ensure that the lighting in each zone meets the decision requirements.

[0096] S33: First, based on the brightness sensor, monitor the actual lighting brightness of each area in real time. The monitoring frequency is 1 time / second, and the monitoring accuracy is ±5cd / ㎡. The monitoring data is then transmitted to the intelligent lighting controller. Next, based on the brightness data collected by each brightness sensor, calculate the lighting uniformity (the ratio of maximum brightness to minimum brightness) of each area in real time and compare it with the set uniformity standard. If the uniformity does not meet the requirements (e.g., the ratio > the set threshold), an adjustment signal is generated.

[0097] S34: First, compare the collected actual brightness with the set target brightness, calculate the brightness deviation value (actual brightness - target brightness), and provide feedback. If the absolute value of the deviation value is >10 cd / m², the dynamic feedback adjustment unit sends a correction command to the intelligent lighting controller to adjust the lamp power until the deviation value between the actual brightness and the target brightness is ≤5 cd / m². For example, if the target brightness is 150 cd / m², the actual brightness is 135 cd / m², and the deviation value is -15 cd / m², then a command is sent to increase the lamp power by 10%, raising the actual brightness to approximately 150 cd / m². Next, based on the feedback brightness adjustment process data, analyze the actual response speed of the lighting system (the time from receiving the command to the actual brightness reaching the target value) and compare it with the set response speed. If the actual response speed is greater than the set value (e.g., set to 300 ms, actual 400 ms), optimize the PWM dimming frequency of the intelligent lighting controller to shorten the response time. If the actual response speed is too small, causing brightness fluctuations, appropriately reduce the dimming frequency to optimize the response speed.

[0098] S4: First, collect lighting parameters and monitor lighting status through the intelligent controller; then, use the fault tree algorithm to diagnose the fault type and trigger a three-level warning according to the degree of impact; then, when a fault occurs, switch emergency lighting and adjust it in zones, linking the ventilation and traffic signal systems; at the same time, connect with the operation and maintenance and traffic control platform to schedule operation and maintenance resources, propose control suggestions, verify the effect after the fault is repaired and archive it to reduce the risk and impact of the fault.

[0099] The specific steps of S4 are as follows:

[0100] S41: First, the intelligent lighting controller collects the operating parameters of each lamp, including the actual operating voltage (monitoring range 0-250V, accuracy ±0.5V), operating current (monitoring range 0-5A, accuracy ±0.01A), surface temperature (monitoring range -20℃-80℃, accuracy ±1℃), and brightness output value (calibrated in conjunction with brightness sensor data, accuracy ±5cd / ㎡). The monitoring frequency is 1 time / second to fully understand the operating status of the lamps. Then, the normal threshold range of each parameter is set (e.g., voltage 198V-242V, current 0.5A-2A, temperature 25℃-60℃). When the parameters of a lamp exceed the threshold range (e.g., voltage drops to 180V, temperature rises to 75℃), the lamp is immediately marked as abnormal, and the time of the abnormality and the trend of parameter changes are recorded to provide data support for fault analysis.

[0101] S42: First, based on the collected abnormal parameters, the fault tree analysis (FTA) algorithm is used to accurately diagnose the fault type. For example, when the brightness output value of the lamp drops sharply and the current is 0, it is determined to be a "circuit open circuit fault". When the brightness drops by more than 10% but the current and voltage are normal and the temperature rises above the threshold, it is determined to be a "lamp aging fault". When multiple adjacent lamps have abnormal voltage at the same time, it is determined to be a "regional power supply fault". The diagnostic accuracy rate is not less than 90%. Then, a three-level early warning mechanism is set according to the scope and severity of the fault.

[0102] Level 1 Warning (Minor Fault): A single non-critical light fixture malfunctions (such as an aging light fixture in the middle section of a tunnel), which has a minor impact on the overall lighting. The system automatically sends a text warning message to the operation and maintenance platform, prompting staff to replace or repair the fixture within 24 hours.

[0103] Level 2 Warning (Moderate Fault): When 3 or more lights in the same area fail (e.g., the circuit of 3 lights in the entrance section is broken), causing the lighting brightness in the area to decrease by more than 20%, the system sends a text and voice warning to the operation and maintenance platform and triggers the electronic display screen in the tunnel to show "Lighting ahead is weakened, drive with caution".

[0104] Level 3 warning (serious fault): When 10 or more lights fail or a batch of lights fail in key areas (entrance transition area, exit transition area), resulting in the area lighting brightness being less than 50% of the national standard, the system will immediately send an emergency warning to the operation and maintenance platform and traffic management department, and automatically activate the emergency lighting plan.

[0105] Then, machine learning algorithms (such as random forest models) are used to predict the probability of failure of the lamps in the next 7 days based on historical failure data (failure type, occurrence time, and lamp age) and real-time monitoring parameters. For example, if a batch of lamps that have been used for 5 years has recently experienced an increase in the frequency of temperature fluctuations, the system predicts that their failure probability is 60% and pushes replacement suggestions in advance.

[0106] S43: Upon detecting a Level 3 fault or sudden power outage in the main lighting system, the emergency lighting system (powered by a UPS) can be immediately activated via a dual-power automatic transfer switch (switching time ≤ 0.5 seconds) to ensure no dark intervals within the tunnel. The emergency lighting system uses LED emergency lights, distributed at 1-meter intervals along the tunnel ceiling, with each light having a brightness ≥ 100 cd / m², meeting basic driving lighting needs. Subsequently, the brightness of the emergency lighting will be optimized based on the location of the fault area; for example, if the main lighting in the entrance transition area fails, the brightness of the emergency lights in the entrance transition area will be increased to 200 cd / m², and the entrance lighting will be activated simultaneously. The system features strong guide lights to help drivers quickly adapt to changes in light levels. In the event of a malfunction in the middle of the tunnel, it maintains the basic brightness of the emergency lights, but focuses on increasing the brightness of the emergency lights within a 50-meter radius before and after the malfunction, creating a "lighting buffer zone." When the emergency lighting is activated, it automatically activates the tunnel ventilation and traffic signal systems. For example, it increases the ventilation system's airflow (by 30%) to prevent smoke accumulation that may result from the malfunction; it controls the traffic lights at the tunnel entrance to flash yellow to remind vehicles entering the tunnel to slow down; and it simultaneously pushes navigation app suggestions (such as those from Gaode and Baidu Maps) to passing vehicles, recommending alternative routes (if any).

[0107] S44: First, it interfaces with the maintenance team's GPS dispatch platform to automatically match the nearest maintenance vehicle and personnel to the tunnel upon receiving a fault warning, generating the optimal repair route (avoiding congested sections). It also synchronizes information such as fault location, fault type, and required spare parts (e.g., LED lights, wiring connectors) to the maintenance personnel's mobile devices, improving repair efficiency. For example, upon receiving a level 2 warning, the system matches a maintenance vehicle 3 kilometers away, plans a route to arrive within 15 minutes, and prompts the user to bring the corresponding model of LED lights. Then, in the event of a level 3 fault, it automatically sends traffic control suggestions to the traffic management department's command platform, such as… It is recommended to implement half-width traffic control in the tunnel and set up temporary flow restriction signs at the tunnel entrance. At the same time, the highway toll system should be linked to display a message on the toll station screen ahead that "the tunnel lighting ahead is faulty, vehicles should proceed slowly," guiding vehicles to slow down in advance or choose other exits. After maintenance personnel complete the fault repair and upload photos and test data (such as the brightness of the area after repair) through the mobile APP, the system can automatically collect the repaired data for verification. After confirming that the lighting parameters have returned to normal, the emergency lighting system should be turned off, the main lighting mode should be switched back, and the fault handling results should be archived in the database and the equipment maintenance record updated.

[0108] A control system for regulating highway tunnel lighting based on dynamic traffic includes:

[0109] The dynamic traffic information collection and analysis module is used to first collect vehicle information from multiple sources, including a high-definition camera array, a microwave radar detector, and a light intensity sensor. Then, after preprocessing the multi-source data, algorithms are used to realize traffic flow statistics and trend prediction, vehicle driving status analysis, and accurate classification statistics of vehicle types.

[0110] The multi-dimensional lighting parameter intelligent decision-making module is used to first set the brightness benchmark value based on traffic flow and then combine it with light intensity correction; then optimize the lighting uniformity according to vehicle type distribution and vehicle position; then set and correct the lighting response speed according to vehicle speed; and at the same time, formulate exclusive lighting strategies for special scenarios to ensure that the lighting parameters are adapted to real-time traffic conditions.

[0111] The lighting execution and dynamic adjustment module is used to first precisely control the lighting fixtures in each zone through multi-channel control, industrial Ethernet communication and PWM dimming; then it divides the tunnel into five functional zones and executes the zone brightness commands; at the same time, it uses brightness sensors to monitor the lighting effect in real time and calculate the uniformity; then it uses dynamic feedback adjustment to correct brightness deviation and optimize response speed, forming a command-execution-monitoring-correction closed loop to ensure that the lighting effect meets the standards.

[0112] The fault warning and emergency handling module is used to first collect lighting parameters and monitor the status of the lighting through the intelligent controller; then, it uses a fault tree algorithm to diagnose the fault type and trigger a three-level warning according to the degree of impact; then, when a fault occurs, it switches emergency lighting and adjusts it in different zones, and links the ventilation and traffic signal systems; at the same time, it connects to the operation and maintenance and traffic control platform to schedule operation and maintenance resources, propose control suggestions, verify the effect after the fault is repaired and archive it, thereby reducing the risk and impact of the fault.

[0113] The dynamic traffic information collection and analysis module includes:

[0114] The multi-source traffic data acquisition unit is used to first capture the dynamic information of vehicles in the tunnel in real time based on a high-definition camera array; then, based on a microwave radar detector, it can accurately detect the speed and position of vehicles even in adverse weather conditions (rain, fog) or low light conditions; and finally, based on a light intensity sensor, it collects the natural light intensity data inside and outside the tunnel in real time.

[0115] The data preprocessing unit first uses a Kalman filter algorithm to denoise the collected raw data such as traffic flow, vehicle speed, and light intensity, removing abnormal data caused by equipment vibration and external interference (such as electromagnetic waves in the tunnel). For example, if the vehicle speed data collected by the camera suddenly spikes to 150 km / h (far exceeding the tunnel speed limit of 80 km / h), the system determines this data as an outlier, removes it using the Kalman filter algorithm, and smooths it using normal data from before and after. Then, it converts the data in different formats (such as video streams, analog signals, and digital signals) collected by the high-definition camera, microwave radar detector, and light intensity sensor into a unified JSON format for easy subsequent data transmission and analysis.

[0116] The traffic flow dynamic analysis unit first uses target detection and tracking algorithms (such as YOLOv8) to count the number of vehicles passing through the tunnel within a unit of time (1 minute as a statistical cycle) based on preprocessed camera data, thus obtaining real-time traffic flow data. Then, it uses time series analysis algorithms (such as LSTM neural networks) to predict the traffic flow trend in the next 10 minutes based on the real-time traffic flow data of the past 30 minutes. For example, it predicts that the traffic flow will increase from the current 300 vehicles / hour to 500 vehicles / hour in the next 10 minutes, providing support for adjusting lighting parameters in advance.

[0117] The vehicle driving status analysis unit first performs statistical analysis on the collected vehicle speed data, calculating the average speed, maximum speed, minimum speed, and vehicle percentage in different speed ranges (e.g., 0-20km / h, 20-40km / h, 40-60km / h, 60-80km / h). For example, if the analysis shows that the current average speed in the tunnel is 30km / h, and 70% of the vehicles are in the 20-40km / h range, the tunnel is determined to be congested. Next, based on the vehicle position data collected by the camera, the unit calculates the distance between adjacent vehicles (head-to-head distance) and compares it with the safe distance standard (e.g., 30 meters at a speed of 30km / h) to analyze the safety level of vehicle driving in the tunnel. If multiple groups of vehicles are found to be less than the safe distance, a rear-end collision risk is determined.

[0118] The vehicle classification and statistics unit first uses deep learning image classification algorithms (such as ResNet50) to accurately classify vehicles based on vehicle image data collected by cameras. Vehicles are categorized into small cars (length < 6 meters, width < 1.8 meters), medium-sized buses (6 meters ≤ length < 10 meters, 1.8 meters ≤ width < 2.5 meters), large trucks (length ≥ 10 meters, width ≥ 2.5 meters), and special vehicles (such as fire trucks and ambulances), with an accuracy rate of no less than 95%. Next, it counts the number and percentage of each vehicle type passing through the tunnel within a unit of time (5 minutes per statistical period). For example, if the count shows 20 small cars, 5 medium-sized buses, and 3 large trucks in the current 5 minutes, representing 74%, 18.5%, and 7.5% respectively, this provides a basis for targeted lighting adjustments.

[0119] The multi-dimensional lighting parameter intelligent decision-making module includes:

[0120] The lighting luminance benchmark value determination unit is used to first establish a mapping relationship model between traffic flow and lighting luminance benchmark values. For example, when traffic flow is <200 vehicles / hour, the lighting luminance benchmark value is set to 100 cd / m²; when traffic flow is ≤200 vehicles / hour and <500 vehicles / hour, the benchmark value is set to 150 cd / m²; and when traffic flow is ≥500 vehicles / hour, the benchmark value is set to 200 cd / m². Based on the real-time traffic flow data and traffic trend prediction data output by the traffic flow dynamic analysis unit, the unit determines the current and future lighting luminance benchmark values ​​for the current and future periods. The system first establishes a baseline value for the tunnel's lighting brightness. Then, based on the light intensity data collected by the light intensity sensor inside and outside the tunnel, the baseline value is adjusted. For example, when the light intensity outside the tunnel is strong (e.g., at noon on a sunny summer day, light intensity > 10000 cd / m²), to avoid a "black hole effect" for drivers entering the tunnel, the baseline value for the lighting brightness of the entrance section is increased by 20%; when the light intensity outside the tunnel is weak (e.g., on a cloudy day or at night, light intensity < 1000 cd / m²), the baseline value for the lighting brightness of the entrance section is decreased by 10%.

[0121] The lighting uniformity optimization unit first optimizes the lighting uniformity based on the vehicle proportion data output by the vehicle classification statistics unit, targeting the lighting needs of different vehicle types. When the proportion of large trucks is >10%, since large trucks will block light, the lighting uniformity in the lower part of the tunnel (height 1-3 meters) needs to be improved, requiring the ratio of maximum brightness to minimum brightness in this area to be ≤1.5. When the proportion of small cars is >80%, the focus is on ensuring the lighting uniformity in the middle part of the tunnel (height 0.5-2 meters), with a ratio ≤2.0. Then, based on the vehicle position data output by the vehicle driving status analysis unit, if a high vehicle density is found in a certain area (e.g., more than 10 vehicles within 50 meters), the brightness of the surrounding lights in that area is automatically increased to compensate for the local lighting uniformity, ensuring uniform brightness distribution in the vehicle driving area.

[0122] The lighting response speed adjustment unit first establishes a correlation model between the average vehicle speed and the lighting response speed. For example, when the average vehicle speed is <30km / h (congested state), the lighting response speed is set to 500ms to avoid frequent fluctuations in brightness; when 30km / h ≤ average vehicle speed <60km / h, the response speed is set to 300ms; when the average vehicle speed is ≥60km / h, the response speed is set to 100ms to ensure that the lighting brightness quickly follows changes in light and meets the driver's visual needs. Then, based on the vehicle speed distribution data output by the vehicle driving status analysis unit, if it is found that there are large differences in vehicle speed within the tunnel (such as some vehicles traveling at 60km / h and some vehicles traveling at 20km / h), the response speed corresponding to the highest proportion of vehicle speed range is used as the basis for a ±50ms correction to take into account the lighting needs of vehicles at different speeds.

[0123] The special scene lighting decision unit is used to automatically activate special weather lighting modes when adverse weather conditions such as rain or fog are detected in the tunnel through cameras or external meteorological data interfaces. For example, in rainy weather, due to the reduced reflectivity of the road surface, the overall lighting brightness in the tunnel is increased by 30% from the baseline value, and the contour lights on the tunnel side walls are turned on to enhance the visibility of the tunnel boundaries. In foggy weather, the lighting brightness is reduced (to avoid strong light scattering affecting visibility), the baseline value is reduced by 20%, and yellow fog lights are turned on to improve light penetration. Then, when the vehicle classification and statistics unit identifies special vehicles (such as fire trucks and ambulances) entering the tunnel, it automatically plans a priority passage route for the special vehicles and corrects the lighting brightness on that route, such as increasing it to 250 cd / m², while turning on the guide lights (red or blue) on both sides of the route to guide the special vehicles to pass quickly and ensure emergency rescue efficiency.

[0124] The lighting execution and dynamic adjustment module includes:

[0125] The intelligent lighting controller unit first enables each intelligent lighting controller to have at least 16 independent control channels to control different areas and types of lighting fixtures (such as entrance section lighting fixtures, middle section lighting fixtures, exit section lighting fixtures, contour lights, and fog lights) within the tunnel, achieving precise control of different lighting fixtures. Next, it uses industrial Ethernet to communicate with the multi-dimensional lighting parameter intelligent decision module, receiving lighting parameter commands (such as brightness values, response speed, and lighting fixture on / off status) with a communication latency of ≤50ms. Simultaneously, it feeds back the operating status data of the lighting fixtures (such as current brightness, power, and temperature) to the multi-dimensional lighting parameter intelligent decision module, forming a data closed loop. Then, using a PWM (Pulse Width Modulation) dimming module, it continuously adjusts the power of the lighting fixtures according to the brightness commands, with a dimming range of 0-100% and an adjustment accuracy of 1%, ensuring that the lighting brightness accurately reaches the target value set by the multi-dimensional lighting parameter intelligent decision module.

[0126] The zoned lighting execution unit first divides the tunnel into five zones based on function and location: entrance transition zone, entrance section, middle section, exit section, and exit transition zone. Each zone has an independent lighting control loop, controlled separately by an intelligent lighting controller. For example, in a 1000-meter-long tunnel, the entrance transition zone is 0-50 meters, the entrance section is 50-200 meters, the middle section is 200-800 meters, the exit section is 800-950 meters, and the exit transition zone is 950-1000 meters. Then, based on the zoned brightness commands issued by the intelligent lighting controller, the unit controls the brightness of the lights in each zone. For example, if the brightness command for the entrance transition zone is 300 cd / m², the entrance section is 200 cd / m², the middle section is 150 cd / m², the exit section is 200 cd / m², and the exit transition zone is 300 cd / m², the zoned lighting execution unit adjusts the brightness of the lights in each zone to the corresponding values ​​to ensure that the lighting in each zone meets the decision requirements.

[0127] The real-time lighting effect monitoring unit first monitors the actual lighting brightness of each area in real time based on the brightness sensor. The monitoring frequency is 1 time / second, and the monitoring accuracy is ±5cd / ㎡. The monitoring data is then transmitted to the intelligent lighting controller. Next, based on the brightness data collected by each brightness sensor, the unit calculates the lighting uniformity (the ratio of maximum brightness to minimum brightness) of each area in real time and compares it with the uniformity standard set by the multi-dimensional lighting parameter intelligent decision module. If the uniformity does not meet the requirements (e.g., the ratio > the set threshold), an adjustment signal is generated.

[0128] The dynamic feedback adjustment unit first compares the actual brightness collected by the real-time lighting effect monitoring unit with the target brightness set by the multi-dimensional lighting parameter intelligent decision module, calculates the brightness deviation value (actual brightness - target brightness), and feeds it back. If the absolute value of the deviation value is >10cd / ㎡, the dynamic feedback adjustment unit sends a correction command to the intelligent lighting controller to adjust the lamp power until the deviation value between the actual brightness and the target brightness is ≤5cd / ㎡. For example, if the target brightness is 150 cd / m², the actual brightness is 135 cd / m², and the deviation is -15 cd / m², then a command is sent to increase the lamp power by 10%, raising the actual brightness to approximately 150 cd / m². Next, based on the brightness adjustment process data fed back by the real-time lighting effect monitoring unit, the actual response speed of the lighting system (the time from receiving the command to the actual brightness reaching the target value) is analyzed and compared with the response speed set by the multi-dimensional lighting parameter intelligent decision-making module. If the actual response speed is greater than the set value (e.g., set at 300 ms, actual at 400 ms), the PWM dimming frequency of the intelligent lighting controller is optimized to shorten the response time. If the actual response speed is too low, causing brightness fluctuations, the dimming frequency is appropriately reduced to optimize the response speed.

[0129] The fault early warning and emergency handling module includes:

[0130] The lighting status monitoring unit first collects the operating parameters of each lighting fixture through the intelligent lighting controller, including the actual operating voltage (monitoring range 0-250V, accuracy ±0.5V), operating current (monitoring range 0-5A, accuracy ±0.01A), surface temperature (monitoring range -20℃-80℃, accuracy ±1℃), and brightness output value (calibrated in conjunction with brightness sensor data, accuracy ±5cd / ㎡). The monitoring frequency is 1 time / second to comprehensively understand the operating status of the lighting fixtures. Then, it sets the normal threshold range for each parameter (e.g., voltage 198V-242V, current 0.5A-2A, temperature 25℃-60℃). When the parameters of a certain lighting fixture exceed the threshold range (e.g., voltage drops to 180V, temperature rises to 75℃), the lighting fixture is immediately marked as abnormal, and information such as the time of the abnormality and the trend of parameter changes is recorded to provide data support for fault analysis.

[0131] The fault early warning analysis unit is used to accurately diagnose fault types based on abnormal parameters collected by the lighting status monitoring unit using the fault tree analysis (FTA) algorithm. For example, when the brightness output value of a lighting fixture drops sharply and the current is 0, it is determined to be a "circuit open circuit fault"; when the brightness drops by more than 10% but the current and voltage are normal, and the temperature rises above the threshold, it is determined to be a "lighting fixture aging fault"; when multiple adjacent lighting fixtures simultaneously experience voltage abnormalities, it is determined to be a "regional power supply fault", with a diagnostic accuracy rate of no less than 90%. Then, a three-level early warning mechanism is set according to the scope and severity of the fault impact.

[0132] Level 1 Warning (Minor Fault): A single non-critical light fixture malfunctions (such as an aging light fixture in the middle section of a tunnel), which has a minor impact on the overall lighting. The system automatically sends a text warning message to the operation and maintenance platform, prompting staff to replace or repair the fixture within 24 hours.

[0133] Level 2 Warning (Moderate Fault): When 3 or more lights in the same area fail (e.g., the circuit of 3 lights in the entrance section is broken), causing the lighting brightness in the area to decrease by more than 20%, the system sends a text and voice warning to the operation and maintenance platform and triggers the electronic display screen in the tunnel to show "Lighting ahead is weakened, drive with caution".

[0134] Level 3 warning (serious fault): When 10 or more lights fail or a batch of lights fail in key areas (entrance transition area, exit transition area), resulting in the area lighting brightness being less than 50% of the national standard, the system will immediately send an emergency warning to the operation and maintenance platform and traffic management department, and automatically activate the emergency lighting plan.

[0135] Then, machine learning algorithms (such as random forest models) are used to predict the probability of failure of the lamps in the next 7 days based on historical failure data (failure type, occurrence time, and lamp age) and real-time monitoring parameters. For example, if a batch of lamps that have been used for 5 years has recently experienced an increase in the frequency of temperature fluctuations, the system predicts that their failure probability is 60% and pushes replacement suggestions in advance.

[0136] The emergency lighting control unit is designed to immediately activate the emergency lighting system (powered by a UPS) via an automatic dual-power switch (switching time ≤ 0.5 seconds) when a level 3 fault or sudden power outage is detected in the main lighting system, ensuring no dark intervals within the tunnel. The emergency lighting system uses LED emergency lights, distributed at 1-meter intervals along the tunnel ceiling, with each light having a brightness ≥ 100 cd / m², meeting basic driving lighting needs. Subsequently, the system optimizes the brightness of the emergency lighting based on the location of the fault area; for example, if the main lighting in the entrance transition area fails, the brightness of the emergency lights in the entrance transition area will be increased to 200 cd / m², and simultaneously... The high-intensity guide lights at the entrance help drivers quickly adapt to changes in light intensity. In the event of a malfunction in the middle of the tunnel, the emergency lights maintain their basic brightness, with a focus on increasing the brightness of emergency lights within a 50-meter radius before and after the malfunction, creating a "lighting buffer zone." When the emergency lighting is activated, it automatically activates the tunnel ventilation and traffic signal systems. For example, it increases the ventilation system's airflow (by 30%) to prevent smoke accumulation that may result from the malfunction; it controls the traffic lights at the tunnel entrance to flash yellow, reminding vehicles entering the tunnel to slow down; and it simultaneously pushes navigation app suggestions (such as Gaode or Baidu Maps) to passing vehicles, recommending alternative routes (if available).

[0137] The fault handling linkage unit first interfaces with the maintenance team's GPS dispatch platform. Upon receiving a fault warning, it automatically matches the nearest maintenance vehicle and personnel to the tunnel, generates the optimal repair route (avoiding congested sections), and synchronizes information such as fault location, fault type, and required spare parts (e.g., LED lights, wiring connectors) to the maintenance personnel's mobile devices, improving maintenance efficiency. For example, upon receiving a level 2 warning, the system matches a maintenance vehicle 3 kilometers away, plans a route to arrive within 15 minutes, and prompts the user to bring the corresponding model of LED lights. Then, in the event of a level 3 fault, it automatically sends traffic control recommendations to the traffic management department's command platform. Suggestions include "implementing half-width traffic control in the tunnel" and "setting up temporary flow restriction signs at the tunnel entrance." Simultaneously, the system should be linked to the highway toll system, displaying a message on the tollbooth screen ahead stating "Tunnel lighting malfunction ahead, vehicles should proceed slowly," guiding vehicles to slow down or choose alternative exits. After maintenance personnel complete the repairs and upload photos and test data (such as the brightness of the repaired area) via a mobile app, the system should automatically collect and verify the repaired data. Once the lighting parameters are confirmed to be normal, the emergency lighting system should be shut down, switching back to main lighting mode, and the fault handling results should be archived in the database, updating the equipment maintenance records.

[0138] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A control method for regulating highway tunnel lighting based on dynamic traffic, characterized in that, The specific steps are as follows: S1: First, vehicle information is collected from multiple sources using a high-definition camera array, microwave radar detector, and light intensity sensor. Then, after preprocessing the multi-source data, algorithms are used to achieve traffic flow statistics and trend prediction, vehicle driving status analysis, and accurate classification statistics of vehicle types. S2: First, set a brightness baseline value based on traffic flow and then adjust it in conjunction with light intensity; then optimize the lighting uniformity according to vehicle type distribution and vehicle position; then set and adjust the lighting response speed according to vehicle speed; at the same time, develop exclusive lighting strategies for special scenarios to ensure that lighting parameters are adapted to real-time traffic conditions. S3: First, through multi-channel control, industrial Ethernet communication and PWM dimming, the lights in each zone are precisely controlled; then the tunnel is divided into five functional zones and the zone brightness command is executed; at the same time, the lighting effect is monitored in real time with the help of the brightness sensor and the uniformity is calculated; then, through dynamic feedback adjustment, the brightness deviation is corrected and the response speed is optimized, forming a command-execution-monitoring-correction closed loop to ensure that the lighting effect meets the standard. S4: First, the intelligent controller collects the lamp parameters and monitors the lamp status; then, the fault tree algorithm is used to diagnose the fault type and trigger a three-level warning according to the degree of impact. In the event of a failure, emergency lighting is switched and zoned, and ventilation and traffic signal systems are coordinated. At the same time, the system connects with the operation and maintenance and traffic control platform to dispatch operation and maintenance resources, propose control suggestions, verify the effect after the failure is repaired and archive the results, thereby reducing the risk and impact of the failure.

2. The control method for regulating highway tunnel lighting based on dynamic traffic according to claim 1, characterized in that, The specific steps of S1 are as follows: S11: First, a high-definition camera array is used to capture the dynamic information of vehicles inside the tunnel in real time; then, a microwave radar detector is used to accurately detect the speed and position of vehicles even in bad weather or low light conditions; finally, a light intensity sensor is used to collect data on the natural light intensity inside and outside the tunnel in real time. S12: First, the Kalman filter algorithm is used to reduce noise in the collected raw data; then, the data in different formats collected by the high-definition camera, microwave radar detector, and light intensity sensor are converted into a unified JSON format to facilitate subsequent data transmission and analysis. S13: First, based on the preprocessed camera data, target detection and tracking algorithms are used to count the number of vehicles passing through the tunnel per unit time to obtain real-time traffic flow data; then, time series analysis algorithms are used to predict the traffic flow trend in the next 10 minutes based on the real-time traffic flow data of the past 30 minutes. S14: First, perform statistical analysis on the driving speed data of all vehicles collected, calculate the average speed, maximum speed, minimum speed of vehicles in the tunnel, and the proportion of vehicles in different speed ranges; then, based on the vehicle position data collected by the camera, calculate the distance between adjacent vehicles and compare it with the safe distance standard to analyze the current safety level of vehicle driving in the tunnel. S15: First, based on the vehicle image data collected by the camera, a deep learning image classification algorithm is used to accurately classify the vehicles into small cars, medium-sized buses, large trucks, and special vehicles; then, the number of each type of vehicle and its proportion passing through the tunnel per unit time are counted.

3. The control method for regulating highway tunnel lighting based on dynamic traffic according to claim 1, characterized in that, The specific steps of S2 are as follows: S21: First, establish a mapping model between traffic flow and lighting brightness benchmark values ​​to determine the lighting brightness benchmark values ​​for the current and future periods based on real-time traffic flow data and traffic trend prediction data; then, correct the brightness benchmark values ​​based on the light intensity data collected by the light intensity sensor inside and outside the tunnel. S22: First, based on the vehicle type ratio data, optimize the lighting uniformity to meet the lighting needs of different vehicle types; then, based on the vehicle location data, if a certain area is found to have a high vehicle density, automatically increase the brightness of the surrounding lights in that area to compensate for the local lighting uniformity and ensure uniform brightness distribution in the vehicle driving area. S23: First, establish a correlation model between the average vehicle speed and the lighting response speed; then, based on the vehicle speed distribution data, if it is found that there are large differences in vehicle speed within the tunnel, the response speed corresponding to the speed range with the highest proportion is used as the basis for correction, taking into account the lighting needs of vehicles with different speeds. S24: When severe weather is detected in the tunnel via camera or external meteorological data interface, the special weather lighting mode is automatically activated; then, when a special vehicle is detected entering the tunnel, a priority passage route is automatically planned for the special vehicle, and the lighting brightness on that route is adjusted.

4. The control method for regulating highway tunnel lighting based on dynamic traffic according to claim 1, characterized in that, The specific steps of S3 are as follows: S31: First, each intelligent lighting controller has at least 16 independent control channels to control different areas and different types of lights in the tunnel, achieving precise control of different lights; then, industrial Ethernet is used for data communication to receive lighting parameter commands; at the same time, the operating status data of the lights is fed back to form a data closed loop. Then, using a PWM dimming module, the power of the lamp is continuously adjusted according to the brightness command; S32: First, the tunnel is divided into five areas according to function and location: entrance transition area, entrance section, middle section, exit section, and exit transition area. Each area is equipped with an independent lighting control circuit, which is controlled by an intelligent lighting controller. Then, the brightness of the lights in each area is controlled according to the zoning brightness instructions issued by the intelligent lighting controller. S33: First, based on the brightness sensor, monitor the actual lighting brightness of each area in real time and transmit the monitoring data to the intelligent lighting controller; then, based on the brightness data collected by each brightness sensor, calculate the lighting uniformity of each area in real time and compare it with the set uniformity standard. If the uniformity is found to be unsatisfactory, an adjustment signal is generated. S34: First, compare the actual brightness collected with the set target brightness, calculate the brightness deviation value, and feed it back; then, based on the feedback brightness adjustment process data, analyze the actual response speed of the lighting system and compare it with the set response speed. If the actual response speed is greater than the set value, the PWM dimming frequency of the intelligent lighting controller will be optimized to shorten the response time. If the actual response speed is too slow, causing brightness fluctuations, the dimming frequency can be appropriately reduced to optimize the response speed.

5. The control method for regulating highway tunnel lighting based on dynamic traffic according to claim 1, characterized in that, The specific steps of S4 are as follows: S41: First, the operating parameters of each lamp are collected through the intelligent lighting controller to fully understand the operating status of the lamp; then, the normal threshold range of each parameter is set so that when the parameter of a lamp exceeds the threshold range, the lamp is immediately marked as abnormal and the abnormal information is recorded. S42: First, based on the collected abnormal parameters, the fault tree analysis algorithm is used to accurately diagnose the fault type; then, a three-level early warning mechanism is set according to the scope and severity of the fault; then, machine learning algorithms are used to predict the probability of the lamp's failure in the next 7 days based on historical fault data and real-time monitoring parameters. S43: When a Level 3 fault or sudden power outage is detected in the main lighting system, the emergency lighting system can be immediately activated through the dual power supply automatic switching switch to ensure that there are no dark intervals in the tunnel; then, the emergency lighting will be zoned and the brightness will be optimized according to the location of the fault area; after that, when the emergency lighting is activated, the tunnel ventilation system and traffic signal system can be automatically linked. S44: First, it connects with the GPS dispatch platform of the maintenance team to automatically match the maintenance vehicle and personnel closest to the tunnel after receiving a fault warning, generate the optimal maintenance route, and synchronize the fault location, fault type, and required spare parts information to the mobile terminal of the maintenance personnel to improve maintenance efficiency; then, when a level 3 fault occurs, it will automatically send traffic control suggestions to the command platform of the traffic management department. After maintenance personnel complete the fault repair and upload photos and test data of the repair completion via the mobile APP, the system can automatically collect the repaired data for verification. Once it confirms that the lighting parameters have returned to normal, the system will turn off the emergency lighting system, switch back to the main lighting mode, and archive the fault handling results to the database and update the equipment maintenance records.

6. A control system for regulating highway tunnel lighting based on dynamic traffic, characterized in that, include: The dynamic traffic information collection and analysis module is used to first collect vehicle information from multiple sources, including a high-definition camera array, a microwave radar detector, and a light intensity sensor. Then, after preprocessing the multi-source data, algorithms are used to realize traffic flow statistics and trend prediction, vehicle driving status analysis, and accurate classification statistics of vehicle types. The multi-dimensional lighting parameter intelligent decision module is used to first set the brightness baseline value based on traffic flow and then combine it with light intensity correction; then optimize the lighting uniformity according to vehicle type distribution and vehicle position; and finally set and correct the lighting response speed according to vehicle speed. At the same time, we develop exclusive lighting strategies for special scenarios to ensure that lighting parameters are adapted to real-time traffic conditions; The lighting execution and dynamic adjustment module is used to first precisely control the lighting fixtures in each zone through multi-channel control, industrial Ethernet communication and PWM dimming; then it divides the tunnel into five functional zones and executes the zone brightness commands; at the same time, it uses brightness sensors to monitor the lighting effect in real time and calculate the uniformity; then it uses dynamic feedback adjustment to correct brightness deviation and optimize response speed, forming a command-execution-monitoring-correction closed loop to ensure that the lighting effect meets the standards. The fault warning and emergency handling module is used to first collect lamp parameters through the intelligent controller and monitor the lamp status; then, it uses a fault tree algorithm to diagnose the fault type and trigger a three-level warning according to the degree of impact. In the event of a failure, emergency lighting is switched and zoned, and ventilation and traffic signal systems are coordinated. At the same time, the system connects with the operation and maintenance and traffic control platform to dispatch operation and maintenance resources, propose control suggestions, verify the effect after the failure is repaired and archive the results, thereby reducing the risk and impact of the failure.

7. The control system for regulating highway tunnel lighting based on dynamic traffic according to claim 6, characterized in that, The dynamic traffic information collection and analysis module includes: The multi-source traffic data acquisition unit is used to first capture the dynamic information of vehicles in the tunnel in real time using a high-definition camera array; then, using a microwave radar detector, it can accurately detect the speed and position of vehicles even in bad weather or low light conditions; and finally, using a light intensity sensor, it collects the natural light intensity data inside and outside the tunnel in real time. The data preprocessing unit first uses the Kalman filter algorithm to reduce noise in the collected raw data; then it converts the data in different formats collected by the high-definition camera, microwave radar detector, and light intensity sensor into a unified JSON format to facilitate subsequent data transmission and analysis. The traffic flow dynamic analysis unit first uses target detection and tracking algorithms to count the number of vehicles passing through the tunnel per unit time based on preprocessed camera data, thus obtaining real-time traffic flow data; then, it uses time series analysis algorithms to predict the traffic flow trend in the next 10 minutes based on the real-time traffic flow data of the past 30 minutes. The vehicle driving status analysis unit is used to first perform statistical analysis on the driving speed data of all vehicles collected, and calculate the average speed, maximum speed, minimum speed and the proportion of vehicles in different speed ranges in the tunnel; then, based on the vehicle position data collected by the camera, it calculates the distance between adjacent vehicles and compares it with the safe distance standard to analyze the safety level of vehicle driving in the tunnel. The vehicle classification and statistics unit is used to first accurately classify vehicles based on the vehicle image data collected by the camera, using a deep learning image classification algorithm, into small cars, medium-sized buses, large trucks, and special vehicles; then it counts the number of each type of vehicle and its proportion passing through the tunnel per unit time.

8. The control system for regulating highway tunnel lighting based on dynamic traffic according to claim 6, characterized in that, The multi-dimensional lighting parameter intelligent decision-making module includes: The lighting brightness reference value determination unit is used to first establish a mapping relationship model between traffic flow and lighting brightness reference value, and then determine the lighting brightness reference value for the current period and a future period based on the real-time traffic flow data and traffic trend prediction data output by the traffic flow dynamic analysis unit; then, the brightness reference value is corrected based on the light intensity data inside and outside the tunnel collected by the light intensity sensor. The lighting uniformity optimization unit first optimizes the lighting uniformity based on the vehicle proportion data output by the vehicle classification statistics unit to meet the lighting needs of different vehicle models. Then, based on the vehicle location data output by the vehicle driving status analysis unit, if a certain area has a high vehicle density, it automatically increases the brightness of the surrounding lights in that area to compensate for the local lighting uniformity and ensure uniform brightness distribution within the vehicle driving area. The lighting response speed adjustment unit is used to first establish a correlation model between the average vehicle speed and the lighting response speed; then, based on the vehicle speed distribution data output by the vehicle driving status analysis unit, if it is found that there are large differences in vehicle speed within the tunnel, the response speed corresponding to the speed range with the highest proportion is used as the basis for adjustment, taking into account the lighting needs of vehicles with different speeds. The special scene lighting decision unit is used to automatically activate the special weather lighting mode when severe weather is detected in the tunnel through a camera or external meteorological data interface; then, when the vehicle type classification and statistics unit identifies a special vehicle entering the tunnel, it automatically plans a priority passage route for the special vehicle and corrects the lighting brightness on that route.

9. The control system for regulating highway tunnel lighting based on dynamic traffic according to claim 6, characterized in that, The lighting execution and dynamic adjustment module includes: The intelligent lighting controller unit is designed to first enable each intelligent lighting controller to have at least 16 independent control channels to control different areas and different types of lighting fixtures within the tunnel, achieving precise control of different lighting fixtures. Next, it uses industrial Ethernet to communicate with the multi-dimensional lighting parameter intelligent decision-making module to receive lighting parameter commands. Simultaneously, it feeds back the operating status data of the lighting fixtures to the multi-dimensional lighting parameter intelligent decision-making module, forming a data closed loop. Then, using a PWM dimming module, it continuously adjusts the power of the lighting fixtures according to brightness commands, ensuring that the lighting brightness accurately reaches the target value set by the multi-dimensional lighting parameter intelligent decision-making module. The zoned lighting execution unit is used to first divide the tunnel into five zones according to function and location: entrance transition zone, entrance section, middle section, exit section, and exit transition zone. Each zone is equipped with an independent lighting control loop, which is controlled by an intelligent lighting controller. Then, according to the zoned brightness instructions issued by the intelligent lighting controller, the brightness of the lights in each zone is controlled. The real-time lighting effect monitoring unit first monitors the actual lighting brightness of each area in real time based on the brightness sensor and transmits the monitoring data to the intelligent lighting controller. Then, based on the brightness data collected by each brightness sensor, it calculates the lighting uniformity of each area in real time and compares it with the uniformity standard set by the multi-dimensional lighting parameter intelligent decision module. If the uniformity is found to be unsatisfactory, an adjustment signal is generated. The dynamic feedback adjustment unit first compares the actual brightness collected by the real-time lighting effect monitoring unit with the target brightness set by the multi-dimensional lighting parameter intelligent decision-making module, calculates the brightness deviation value, and feeds it back. Then, based on the brightness adjustment process data fed back by the real-time lighting effect monitoring unit, it analyzes the actual response speed of the lighting system and compares it with the response speed set by the multi-dimensional lighting parameter intelligent decision-making module. If the actual response speed is greater than the set value, the PWM dimming frequency of the intelligent lighting controller is optimized to shorten the response time. If the actual response speed is too small, causing brightness fluctuations, the dimming frequency is appropriately reduced to optimize the response speed.

10. The control system for regulating highway tunnel lighting based on dynamic traffic according to claim 6, characterized in that, The fault early warning and emergency handling module includes: The lighting status monitoring unit is used to first collect the operating parameters of each lighting fixture through the intelligent lighting controller to fully understand the operating status of the lighting fixture; then, it sets the normal threshold range for each parameter so that when the parameter of a certain lighting fixture exceeds the threshold range, the lighting fixture is immediately marked as abnormal and the abnormal information is recorded. The fault early warning analysis unit is used to first accurately diagnose the fault type based on the abnormal parameters collected by the lighting fixture status monitoring unit and the fault tree analysis algorithm; then, a three-level early warning mechanism is set according to the scope and severity of the fault; and then, machine learning algorithms are used to predict the probability of the lighting fixture's failure in the next 7 days based on historical fault data and real-time monitoring parameters. The emergency lighting control unit is used to immediately activate the emergency lighting system through a dual-power automatic switching switch when a level 3 fault or sudden power outage is detected in the main lighting system, ensuring that there are no dark intervals in the tunnel; then, it will optimize the brightness of the emergency lighting according to the location of the fault area; after that, when the emergency lighting is activated, it can automatically link the tunnel ventilation system and traffic signal system. The fault handling linkage unit first connects with the maintenance team's GPS dispatch platform to automatically match the nearest maintenance vehicle and personnel to the tunnel upon receiving a fault warning, generate the optimal maintenance route, and synchronize the fault location, fault type, and required spare parts information to the maintenance personnel's mobile devices to improve maintenance efficiency. Then, in the event of a level 3 fault, it automatically sends traffic control suggestions to the traffic management department's command platform. After the maintenance personnel complete the fault repair and upload photos and test data of the repair completion via the mobile APP, it can automatically collect the repaired data for verification. Once it confirms that the lighting parameters have returned to normal, it shuts down the emergency lighting system, switches back to the main lighting mode, archives the fault handling results to the database, and updates the equipment maintenance records.

Images