Tunnel car light cooperative adaptive energy-saving lighting and safety guiding method and system

Abstract

The application provides a tunnel vehicle lamp adaptive energy-saving lighting and safety guiding method and system, relates to the technical field of tunnel traffic lighting and intelligent power distribution, and constructs a four-in-one collaborative architecture of vehicle perception, lighting regulation and control, energy storage scheduling and safety guiding, proposes a dynamic lighting matrix of light-electricity-vehicle space-time matching, an energy storage and lighting micro-grid autonomous operation mode and a multi-modal visual guiding light coding technology, realizes single-vehicle-level fine lighting, distributed energy storage intelligent scheduling and lighting integrated safety guiding, finally achieves the three goals of energy-saving maximization, safety optimization and cost minimization, and provides core technical support for urban tunnel intelligent lighting and low-carbon operation.

Description

Technical Field

[0001] This invention relates to the technical field of tunnel traffic lighting and intelligent power distribution, specifically to an adaptive energy-saving lighting and safety guidance method and system for coordinated vehicle lighting in tunnels. Background Technology

[0002] By the end of 2024, the total length of highway tunnels in China exceeded 32,600 kilometers, with urban road tunnels accounting for more than 35%. Tunnel lighting, as a core energy-consuming unit in tunnel operation, accounts for 60%-70% of the total energy consumption of tunnel operations, with annual lighting electricity costs for a single kilometer-long urban tunnel reaching several million yuan. The low-carbon transformation of tunnel lighting systems has become a core task in upgrading urban transportation infrastructure.

[0003] Meanwhile, with the rapid development of vehicle-to-everything (V2X), edge computing, and distributed energy storage technologies, tunnel lighting has evolved from traditional timed and quantitative static control to dynamic intelligent control based on perception, decision-making, and execution. However, existing technologies generally suffer from systemic problems such as separation of vehicle and lamp, disconnect between lamp and energy, and separation of lighting and guidance, making it difficult to achieve a synergistic improvement in energy efficiency and safety levels.

[0004] Currently, tunnel lighting and safety guidance technologies can be mainly divided into the following four categories, all of which have fundamental shortcomings:

[0005] (1) Traditional timed illumination dimming system: Based on a preset time schedule or the illumination at the opening, the dimming is graded and the energy saving rate is limited. It cannot sense the real-time status of vehicles, and there is a reverse logic phenomenon of lights on when there are no vehicles and lights off when there are vehicles. In addition, it does not consider the peak and valley characteristics of the power grid and the coordinated optimization of the energy storage system, resulting in low energy utilization efficiency.

[0006] (2) Traffic flow-based area dimming system: The brightness of a zone is adjusted by detecting traffic flow in the area using coils and cameras. However, it has drawbacks such as slow response, coarse control granularity, and inability to distinguish between different vehicle types. It is prone to blind spots or over-illumination and cannot provide personalized safety guidance for individual vehicles.

[0007] (3) Vehicle-road cooperative lighting system: The system uses V2X technology to obtain vehicle location information for following lighting. However, it only achieves basic linkage of "the light turns on when the vehicle arrives" and does not integrate vehicle speed, vehicle type, and driving status for fine-grained brightness and color temperature control, nor does it consider the impact of lighting pattern on driver visual comfort; at the same time, it relies entirely on grid power supply and does not combine energy storage system to achieve load optimization.

[0008] (4) Tunnel safety guidance system: Safety prompts are mostly provided by independent variable information signs and guide signs, which are physically separated from the lighting system and controlled independently. Dynamic safety guidance cannot be achieved through changes in lighting color and brightness, resulting in limited guidance effect and increasing the construction and maintenance costs of tunnel infrastructure.

[0009] Existing research indicates that lighting conditions are a significant factor contributing to traffic accidents in highway tunnels, while inadequate lighting control strategies also lead to substantial energy waste. Current technologies fail to simultaneously resolve the core contradictions of energy efficiency versus safety versus energy efficiency, and they also fail to deeply integrate lighting, vehicle, and energy systems. Therefore, developing a vehicle-lighting-energy-coordinated adaptive energy-saving lighting and safety guidance system has significant engineering and application value for reducing tunnel operating energy consumption, improving driving safety, and optimizing power grid load distribution. Summary of the Invention

[0010] This invention was made to solve the above-mentioned problems, and its purpose is to provide an adaptive energy-saving lighting and safety guidance method and system that enables coordinated use of vehicle lights in tunnels.

[0011] This invention provides an adaptive energy-saving lighting and safety guidance method for vehicle-light coordination in tunnels, characterized by the following steps: S1: Multi-source data acquisition and fusion step, using sensing devices deployed in the tunnel to synchronously acquire multi-source data, and performing time alignment and preprocessing on the acquired multi-source data to obtain a standardized time-series dataset; the multi-source data includes vehicle data, environmental data, energy data, and equipment data; S2: Dynamic lighting matrix management step, dividing the tunnel into multiple basic lighting units at preset intervals, dynamically dividing the tunnel into multiple lighting areas based on the basic lighting units and vehicle data, according to the real-time vehicle distribution, and identifying vehicle driving status based on vehicle data; S3: Adaptive lighting control step, calculating based on vehicle data, vehicle driving status, environmental data, and tunnel location... S4: Lighting parameters of each lamp in each lighting area generate lighting control instructions; S5: Energy storage microgrid scheduling step, based on real-time electricity price, grid load status, state of charge of the tunnel's energy storage system, and lighting load prediction determined based on lighting control instructions, generates energy storage charging and discharging strategies and corresponding energy storage scheduling instructions; S6: Light-coded safety guidance step, converts safety events in the tunnel into light codes represented by lighting color, brightness, and flickering mode, generates multiple guidance modes, and makes decisions on multiple guidance modes according to preset priority rules, generating safety guidance instructions that integrate lighting and safety guidance; S7: Collaborative execution control step, sends lighting control instructions, energy storage scheduling instructions, and safety guidance instructions to corresponding equipment for execution, and collects execution effect data in real time for feedback to continuously optimize subsequent instructions.

[0012] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination provided by the present invention may also have the following features: the method for time alignment of multi-source data in S1 is as follows: multi-source data is synchronized at the microsecond level based on the PTP precise time protocol, and the method for preprocessing multi-source data is as follows: outliers are removed based on the 3σ principle through the edge gateway, multi-source data with ≤3 consecutive missing sampling points are filled by linear interpolation, multi-source data with >3 consecutive missing sampling points are marked as invalid and replaced by historical multi-source data, and multi-source data with different sampling frequencies are uniformly resampled to the same frequency to complete the spatiotemporal alignment and feature fusion.

[0013] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle-light coordination provided by this invention may also have the following features: The division of multiple lighting areas in S2 follows these rules: Single-vehicle scenario: Centered on the vehicle, a core lighting area is divided along the tunnel length according to a preset number of basic lighting units. Transitional lighting areas are divided upstream and downstream of the core lighting area. The brightness of the transitional lighting area decreases linearly with a gradient away from the vehicle. The remaining areas are basic safety lighting areas, with brightness lower than that of the transitional lighting areas. Multi-vehicle scenario: When the distance between adjacent vehicles is less than a preset distance threshold, the overlapping core lighting area and transitional lighting area of ​​adjacent vehicles are merged into a continuous high-brightness lighting area. The brightness of the high-brightness lighting area is consistent with that of the core lighting area. No-vehicle scenario: All basic lighting units within the tunnel are set as basic safety lighting areas.

[0014] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle-light coordination provided by the present invention may also have the following features: wherein, the lighting parameters in S3 include: target brightness, calculated based on vehicle type and speed in vehicle data, ambient illuminance in environmental data and tunnel location; target color temperature, adjusted based on vehicle driving status and current time period; and target light emission angle, adjusted based on vehicle type in vehicle data to avoid glare.

[0015] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights coordinated with the present invention may also have the following feature: wherein the calculation formula for the target brightness in S3 is:

[0016]

[0017] In the formula, Let be the luminance (cd / m²) of the j-th lamp in the i-th basic lighting unit; For tunnels The base brightness standard value for the location; For vehicle model coefficient; This is the vehicle speed coefficient; This refers to the ambient illuminance coefficient. This is the tunnel location coefficient.

[0018] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lighting coordination provided by this invention may also have the following features: wherein, the energy storage charging and discharging strategy in S4 includes: charging strategy: when the electricity price is in a valley and the state of charge of the energy storage system is lower than a first preset threshold, it is charged at rated power; when the grid load is lower than a preset load threshold and the state of charge of the energy storage system is lower than a second preset threshold, it is charged at half power; discharging strategy: when the electricity price is in a peak period and the state of charge of the energy storage system is higher than a third preset threshold, the energy storage system supplies power, and the power supply is matched with the lighting load corresponding to the lighting control command; when the grid load exceeds the preset load threshold, the energy storage system discharges to supplement the power supply.

[0019] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lighting coordination provided by the present invention may also have the following features: wherein, the energy storage charging and discharging strategy in S4 further includes: autonomous operation strategy: when a power grid outage or power quality exceeding the standard is detected, the power grid connection is disconnected, and the energy storage system independently supplies power to maintain the preset minimum safe lighting in the tunnel. When the power grid returns to normal, the power grid connection is restored, and the energy storage system is charged.

[0020] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination provided by this invention may also have the following features: The guidance modes in S5 include: normal driving mode: corresponding to constant white light; forward deceleration warning mode: corresponding to gradually flashing yellow light, with a flashing frequency of a first preset frequency and brightness gradually changing within a preset range; accident or construction warning mode: corresponding to alternating flashing red light, with a flashing frequency of a second preset frequency, the second preset frequency being higher than the first preset frequency, and adjacent lights alternately illuminating to form a dynamic light strip guiding lane changes; abnormal vehicle warning mode: corresponding to pulsed blue light, with a flashing frequency of a third preset frequency, the third preset frequency being higher than the second preset frequency, and brightness suddenly increasing and decreasing; congestion and slow traffic mode: corresponding to constant orange light, with brightness higher than the brightness corresponding to the normal driving mode. The priority of the various guidance modes, from highest to lowest, is: accident or construction warning mode, abnormal vehicle warning mode, congestion and slow traffic mode, forward deceleration warning mode, and normal driving mode.

[0021] The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lighting coordination provided by the present invention may also have the following features: wherein, the method for continuously optimizing subsequent instructions in S6 is: based on a deep reinforcement learning algorithm, the decision logic for generating lighting control instructions, energy storage scheduling instructions and safety guidance instructions is continuously optimized using execution effect data.

[0022] This invention also provides an adaptive energy-saving lighting and safety guidance system for tunnel vehicle-light coordination, characterized by: a multi-source data acquisition and fusion module, which synchronously acquires multi-source data through sensing devices deployed in the tunnel, and performs time alignment and preprocessing on the acquired multi-source data to obtain a standardized time-series dataset; the multi-source data includes vehicle data, environmental data, energy data, and equipment data; a dynamic lighting matrix management module, which divides the tunnel into multiple basic lighting units at preset intervals, dynamically divides the tunnel into multiple lighting areas based on the basic lighting units and vehicle data, according to real-time vehicle distribution, and identifies vehicle driving status based on vehicle data; and an adaptive lighting control module, which calculates the lighting parameters for each lighting area based on vehicle data, vehicle driving status, environmental data, and tunnel location. The system generates lighting control commands based on the lighting parameters of each lamp in the illuminated area. The energy storage microgrid scheduling module generates energy storage charging and discharging strategies and corresponding energy storage scheduling commands based on real-time electricity prices, grid load status, the state of charge of the tunnel's energy storage system, and lighting load predictions determined by the lighting control commands. The optical coding safety guidance module converts safety events within the tunnel into optical codes representing lighting color, brightness, and flickering patterns, generating multiple guidance modes and making decisions based on preset priority rules to generate safety guidance commands that integrate lighting and safety guidance. The collaborative execution control module distributes lighting control commands, energy storage scheduling commands, and safety guidance commands to the corresponding devices for execution and collects execution effect data in real time for feedback, continuously optimizing subsequent commands.

[0023] Compared with the prior art, the present invention has the following advantages:

[0024] This invention addresses core issues in existing tunnel lighting systems, such as low vehicle-light coordination, poor energy efficiency, separation of lighting and safety guidance, and lack of integration with energy storage systems for optimized power distribution. It proposes an adaptive energy-saving lighting and safety guidance method and system for vehicle-light coordination in tunnels. This invention constructs a four-in-one collaborative architecture integrating vehicle perception, lighting control, energy storage scheduling, and safety guidance. It proposes a dynamic lighting matrix with spatiotemporal matching of light, electricity, and vehicle, an autonomous operation mode for energy storage and lighting microgrids, and multimodal visual guidance optical coding technology. This achieves refined lighting at the single-vehicle level, intelligent scheduling of distributed energy storage, and integrated lighting and safety guidance, ultimately achieving the triple goals of maximizing energy saving, optimizing safety, and minimizing cost, providing core technological support for intelligent lighting and low-carbon operation in urban tunnels. Attached Figure Description

[0025] Figure 1 This is a flowchart of an adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in an embodiment of the present invention.

[0026] Figure 2This is a timing flowchart of the adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in an embodiment of the present invention.

[0027] Figure 3 This is a schematic diagram of the dynamic lighting matrix in an embodiment of the present invention.

[0028] Figure 4 This is an adaptive lighting control logic block diagram in an embodiment of the present invention.

[0029] Figure 5 This is a diagram of the energy storage microgrid architecture in an embodiment of the present invention.

[0030] Figure 6 This is a block diagram of the charging and discharging scheduling logic of the energy storage system in an embodiment of the present invention.

[0031] Figure 7 This is a schematic diagram of the adaptive energy-saving lighting and safety guidance system architecture that enables tunnel vehicle lights to work in tandem, as described in an embodiment of the present invention. Detailed Implementation

[0032] To make the technical means, creative features, objectives and effects of the present invention easy to understand, the following embodiments, in conjunction with the accompanying drawings, specifically illustrate the adaptive energy-saving lighting and safety guidance method and system for tunnel vehicle lights to work together.

[0033] This embodiment provides an adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights to work together, including the following steps:

[0034] Figure 1 This is a flowchart of an adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in an embodiment of the present invention. Figure 2 This is a timing flowchart of the adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in an embodiment of the present invention.

[0035] like Figures 1-2 As shown, step S1 is the multi-source data acquisition and fusion step. Multi-source data is simultaneously acquired using sensing devices deployed within the tunnel. The acquired multi-source data is then time-aligned and preprocessed to obtain a standardized time-series dataset. Specifically:

[0036] By deploying RSU devices every 50m in the tunnel, 77GHz millimeter-wave radar every 100m, high-precision illuminance sensors at tunnel entrances and exits and in the middle section, power quality monitoring devices in the power distribution room, and BMS systems in the energy storage cabinet, four types of core data—vehicle data, environmental data, energy data, and equipment data—are collected simultaneously.

[0037] (1) Vehicle data: The vehicle ID, precise location (error ≤ 1m), speed, acceleration, vehicle type (small car / medium car / large car / special vehicle), light status, and driving direction are obtained by receiving BSM messages broadcast by vehicles through the RSU device. The location and speed of vehicles not connected to the network are supplemented by millimeter-wave radar to achieve full vehicle coverage.

[0038] (2) Environmental data: Real-time illuminance inside and outside the tunnel is collected by illuminance sensor, temperature and humidity inside the tunnel are collected by temperature and humidity sensor, visibility inside the tunnel is collected by visibility sensor, and dry or wet or icy road surface status is collected by road surface status sensor.

[0039] (3) Energy data: Real-time voltage, current, power and frequency of the power grid are collected through power quality monitoring devices, real-time time-of-use electricity price is obtained through the power grid interface, SOC (state of charge), SOH (state of health), single cell voltage, temperature and charge and discharge power of the energy storage system are collected through BMS, and DC bus voltage is collected through DC bus voltage sensor.

[0040] (4) Equipment data: Real-time brightness, color temperature, operating status and fault information of each LED lamp are collected through the LED intelligent controller, the status of each branch switch is collected through the power distribution switch controller, and the working mode, input and output power and efficiency of the bidirectional AC / DC converter are collected through the converter controller.

[0041] The edge controller, a computing device deployed locally in the tunnel, receives multi-source data collected by sensing devices via a built-in edge gateway and performs microsecond-level synchronization of the multi-source data based on the PTP (Precise Time Protocol). After synchronization, the edge gateway continues to preprocess the multi-source data, specifically including: firstly, removing outliers (such as abrupt changes caused by sensor malfunctions) based on the 3σ principle; filling data with ≤3 consecutive missing sampling points using linear interpolation; and marking data with >3 consecutive missing sampling points as invalid and replacing them with historical data. Then, sensor data from different sampling frequencies are uniformly resampled to 10Hz, completing the spatiotemporal alignment and feature fusion of the multi-source data, and outputting a standardized time-series dataset.

[0042] Step S2 is the dynamic lighting matrix management step, which divides the tunnel into multiple basic lighting units according to preset intervals. Based on the basic lighting units and vehicle data, the tunnel is dynamically divided into multiple lighting areas according to real-time vehicle distribution, and the vehicle driving status is identified based on vehicle data. Specifically:

[0043] Figure 3 This is a schematic diagram of the dynamic lighting matrix in an embodiment of the present invention.

[0044] like Figure 3As shown, the tunnel is divided into multiple basic lighting units at 50m intervals. Each unit contains 16 independently controllable LED lights (8 lights per side for a one-way three-lane tunnel), improving control granularity. The lighting areas are dynamically divided and adjusted according to the real-time vehicle distribution, forming a dynamic lighting matrix composed of basic lighting units. The dynamic lighting matrix can change according to the vehicle's position, achieving lights turning on when a vehicle approaches, turning off when a vehicle moves, and lights following the vehicle's movement.

[0045] The division of lighting zones follows these rules:

[0046] (1) Taking the vehicle as the center, the tunnel is divided into a core lighting zone (one core basic lighting unit at the front and one at the back, totaling 150m with 100% brightness) and a transition lighting zone (one transition basic lighting unit outside the core zone, totaling 100m with brightness decreasing linearly from 70% to 30% away from the vehicle). The remaining area is a basic safety lighting zone (brightness 0.5cd / m²) to meet monitoring and emergency needs.

[0047] (2) Multi-vehicle scenario: When the distance between adjacent vehicles is less than 150m, they are automatically merged into a multi-vehicle merged lighting area, that is, the core lighting area and transition lighting area of ​​the adjacent vehicles are merged into a continuous 100% high brightness lighting area to avoid lighting flicker and alternation of light and dark.

[0048] (3) No vehicle scenario: All basic lighting units in the tunnel are set as basic safety lighting zones (brightness 0.5cd / m²) to meet monitoring and emergency needs.

[0049] Simultaneously, the vehicle's driving status is identified based on changes in vehicle speed, acceleration, and position, and is divided into five categories: normal driving (absolute acceleration < 2m / s²), acceleration (acceleration ≥ 2m / s²), deceleration (acceleration ≤ -2m / s²), stopping (speed < 5km / h and lasting ≥ 3s), and abnormal driving (sudden braking: acceleration ≤ -5m / s²; frequent lane changes: ≥ 2 lane changes within 10s; driving in the wrong direction: driving in the opposite direction to the tunnel's designated direction).

[0050] Figure 4 This is an adaptive lighting control logic block diagram in an embodiment of the present invention.

[0051] Step S3 is the adaptive lighting control step, such as... Figure 4 As shown, based on vehicle data, vehicle driving status, environmental data, and tunnel location, the lighting parameters of each LED light in each lighting area are calculated, and lighting control instructions are generated, specifically:

[0052] Based on the collected input features such as vehicle type, speed, ambient illuminance, and tunnel location, the system calculates the lighting parameters for each LED in each lighting area suitable for the current scene, including target brightness, target color temperature, and target emission angle, through brightness benchmark calculation and multi-factor correction. To avoid visual discomfort caused by sudden brightness changes, the calculated target brightness is also smoothed, and finally, precise LED lighting control commands are generated.

[0053] (1) Target brightness adjustment: The formula for calculating target brightness is:

[0054]

[0055] In the formula, Let be the luminance (cd / m²) of the j-th lamp in the i-th basic lighting unit;

[0056] For tunnels The basic luminance standard value for the location (determined according to the "Detailed Rules for Lighting Design of Highway Tunnels" JTG / TD70 / 2-01-2014);

[0057] For vehicle type coefficients (small cars 1.0, medium cars 1.2, large cars 1.5, special vehicles 2.0);

[0058] The speed coefficient is 1.2 when v < 40 km / h, 1.0 when 40 km / h ≤ v ≤ 80 km / h, and 0.9 when v > 80 km / h.

[0059] The ambient illuminance coefficient ( =1000 / ,in The external illuminance of the tunnel is expressed in lux. When the value is >10000 lux, take 0.1).

[0060] This is the tunnel location coefficient (1.2 for entrance and exit sections, and 1.0 for the middle section).

[0061] Basic brightness corresponding to small cars ( The brightness is 1.0 times that of a standard vehicle, 1.2 times that of a medium-sized vehicle, 1.5 times that of a large vehicle, and 2.0 times that of a special vehicle. Brightness increases by 20% when vehicle speed v < 40 km / h and decreases by 10% when v > 80 km / h. The brightness of the transition sections at entrances and exits (50m before and 200m after the entrance, and 200m before and 50m after the exit) transitions smoothly according to an exponential curve to avoid abrupt changes in brightness.

[0062] (2) Target color temperature control: 4000K neutral white light is used during normal driving; 5000K cool white light is used in the transition section at the entrance and exit to alleviate the light and dark adaptation; 3000K warm white light is used during the peak period of fatigue driving from 2:00 to 5:00 am.

[0063] (3) Target light emission angle adjustment: The LED light illumination angle of the large vehicle in the lighting area is automatically adjusted to shift downward by 5° to avoid glare.

[0064] Step S4 is the energy storage microgrid dispatching step. Based on real-time electricity prices, grid load status, the state of charge of the energy storage system in the tunnel, and the lighting load forecast determined based on lighting control instructions, an energy storage charging and discharging strategy and corresponding energy storage dispatching instructions are generated, specifically as follows:

[0065] Figure 5 This is a diagram of the energy storage microgrid architecture in an embodiment of the present invention.

[0066] like Figure 5 The diagram shows a three-level DC microgrid architecture consisting of a power grid, energy storage, and lighting. All loads and energy storage are directly connected to the DC bus, avoiding energy loss from multiple AC-DC conversions.

[0067] Figure 6 This is a block diagram of the charging and discharging scheduling logic of the energy storage system in an embodiment of the present invention.

[0068] The charging and discharging scheduling logic of the tunnel energy storage system is as follows: Figure 6 As shown, the tunnel energy storage system includes a prediction module, a condition monitoring module, a charge / discharge optimization calculation module, a charge / discharge control module, and a protection module. The specific workflow is as follows:

[0069] The prediction module acquires real-time grid price information, grid load prediction data, and lighting load prediction values ​​determined based on the lighting control instructions generated in step S3 for future scheduling cycles, and outputs the aforementioned prediction time-series data to the charge-discharge optimization calculation module. The status monitoring module acquires the current state of charge, health status, battery temperature, and DC bus voltage of the energy storage system in real time, and outputs these parameters to the charge-discharge optimization calculation module and the protection module.

[0070] The charge / discharge optimization calculation module receives the predicted time-series data output by the prediction module and the operating parameters output by the state monitoring module. With the goal of minimizing total electricity cost, and under the constraints of upper and lower limits of state of charge, charging and discharging power limits, and grid power change rate limits, it generates an energy storage charging and discharging strategy based on the following optimization objective function:

[0071]

[0072]

[0073] Constraints:

[0074]

[0075]

[0076]

[0077]

[0078] In the formula, C: total electricity cost; Power supplied by the power grid at time t; Real-time electricity price at time t; Average power consumption of the power grid within the dispatch period T; : Power grid load stabilization weighting coefficient; : Energy storage battery loss cost at time t; Purchase cost of energy storage batteries; : Battery cycle life; : Rated capacity of energy storage battery Energy storage battery charging power; : Discharge power of energy storage battery; Rated power of the bidirectional AC / DC converter; : Rate of change of power grid.

[0079] The charge and discharge control module receives the energy storage charge and discharge strategy generated by the charge and discharge optimization calculation module, converts it into specific energy storage scheduling instructions, and sends them to the execution mechanism (BMS, bidirectional AC / DC converter, and battery cluster composed of multiple energy storage batteries) for execution.

[0080] Energy storage charging and discharging strategies include:

[0081] (1) Charging strategy: When the grid electricity price is in a low-price period (taking Shanghai as an example, 22:00 - 6:00 the next day) and the state of charge of the energy storage system is lower than the first preset threshold (energy storage SOC < 90%), it is charged at rated power. When the grid load is lower than the threshold (such as grid load factor < 50%) and the state of charge of the energy storage system is lower than the second preset threshold (energy storage SOC < 80%), it is charged at half power to absorb the surplus power of the grid.

[0082] (2) Discharge strategy: When the electricity price is at its peak (taking Shanghai as an example, 8:00-22:00) and the state of charge of the energy storage system is higher than the third preset threshold (energy storage SOC>20%), the energy storage system will be used to supply power first, and the power supply will be matched with the lighting load corresponding to the lighting control command in real time. When the grid load exceeds the preset load threshold (such as grid load factor>80%), the energy storage system will discharge to supplement the power supply.

[0083] The protection module receives the operating parameters output by the status monitoring module and executes corresponding protection actions according to the different abnormal states detected: when the energy storage system is detected to have over-temperature, over-voltage, or over-current, it sends a power reduction or shutdown protection signal to the charge and discharge control module; when the grid is detected to be out of power or the power quality is seriously out of standard (e.g., voltage deviation exceeds ±10%, frequency deviation exceeds ±0.5Hz), it sends a grid fault protection signal to the charge and discharge control module, triggering the following autonomous operation strategy.

[0084] (3) Autonomous operation strategy: After receiving the grid fault protection signal sent by the protection module, the charging and discharging control module immediately disconnects the grid connection switch, and the energy storage system automatically switches to independent power supply mode to maintain the preset minimum safe lighting (0.5cd / m²) in the tunnel for a duration of ≥2 hours. When the grid returns to normal, the energy storage system automatically switches back to grid-connected operation mode, restores grid connection, and recharges the energy storage battery.

[0085] Step S5 is the optical coding safety guidance step, which converts safety events in the tunnel into optical codes represented by lighting color, brightness and flashing mode, generates multiple guidance modes, and makes decisions on multiple guidance modes according to preset priority rules, generating a safety guidance command that integrates lighting and safety guidance, so that lighting is guidance and no additional guidance equipment needs to be installed.

[0086] The boot mode includes 5 preset basic boot modes and supports custom extensions. Different boot modes have preset priorities, with higher priority boot modes overriding lower priority boot modes, including:

[0087] Normal driving mode: Provides basic lighting with 4000K constant white light.

[0088] Forward deceleration warning mode: The LED lights in the upstream 200m area turn yellow and flash gradually (frequency is the first preset frequency: 1Hz, duty cycle 50%, brightness gradually changes from 100% to 50% and then back to 100%) to guide vehicles to decelerate in advance.

[0089] Accident or construction warning mode: The LED lights 300m upstream of the accident area turn red and flash alternately (frequency is the second preset frequency: 2Hz). Adjacent LED lights light up alternately to form a dynamic light strip to guide vehicles to change lanes to unaffected lanes.

[0090] Abnormal Vehicle Warning Mode: The LED lights in the area where the abnormal vehicle is located and within 100m in front and behind it turn into blue light pulses flashing (frequency is the third preset frequency: 3Hz, duty cycle 30%, brightness suddenly increases from 0% to 100% and then suddenly drops to 0%), warning surrounding vehicles to pay attention and avoid the area.

[0091] Congestion Mode: LED lights in congested areas turn orange and remain on, increasing brightness by 20% to remind drivers to maintain a safe following distance.

[0092] The priority of the various guidance modes, from highest to lowest, is as follows: accident or construction warning mode, abnormal vehicle warning mode, congestion and slow traffic mode, forward deceleration warning mode, and normal driving mode. High-priority modes can cover low-priority modes.

[0093] Step S6 is the collaborative execution control step, which sends lighting control commands, energy storage scheduling commands, and safety guidance commands to the corresponding devices for execution, and collects execution effect data in real time for feedback to continuously optimize subsequent commands. Specifically:

[0094] The edge controller sends optimized lighting control commands (brightness, color temperature, light emission angle), energy storage scheduling commands (charging and discharging power, working mode), and safety guidance commands (light color, brightness, flashing mode) to the LED smart controller and inverter controller via the CAN bus. The LED light response time is ≤100ms.

[0095] Real-time data on the actual brightness and color temperature of LED lights, and the actual charging and discharging power of the energy storage system are collected and fed back. The algorithm model set (hereinafter referred to as the edge model) in each control step (steps S3-S5) deployed in the edge controller is continuously optimized through deep reinforcement learning algorithm (DQN) to update the decision logic used to generate lighting control commands, energy storage scheduling commands, and safety guidance commands, thereby continuously improving control accuracy and adaptability.

[0096] Figure 7 This is a schematic diagram of the adaptive energy-saving lighting and safety guidance system architecture that enables tunnel vehicle lights to work in tandem, as described in an embodiment of the present invention.

[0097] This embodiment also provides an adaptive energy-saving lighting and safety guidance system (hereinafter referred to as the guidance system) that can coordinate with tunnel vehicle lights, such as Figure 7 The diagram shows the layered architecture of this guidance system. The modules communicate bidirectionally via Ethernet and 5G networks to collaboratively achieve integrated control of the vehicle, lighting, and energy systems. The guidance system includes a multi-source data perception layer, an edge computing decision layer, a collaborative execution control layer, and a cloud management and iteration layer.

[0098] The system comprises three layers: a multi-source data perception layer for real-time acquisition and fusion of multi-source data; an edge computing decision layer for adaptive lighting control, dynamic energy storage scheduling, and safety guidance strategy generation via edge computing; a collaborative execution control layer for coordinated control of LED lights and energy storage systems; and a cloud management and iteration layer for remote centralized management and continuous iterative optimization of edge models. All layers interact via a unified vehicle-lighting-energy data exchange bus, enabling the uploading of multi-source data, the issuance of commands, and feedback of execution results.

[0099] The multi-source data perception layer includes a multi-source data acquisition and fusion module, which is used to implement step S1, namely: synchronously acquiring multi-source data through sensing devices deployed in the tunnel, and performing time alignment and preprocessing on the acquired multi-source data to obtain a standardized time-series dataset; the multi-source data includes vehicle data, environmental data, energy data and equipment data.

[0100] This module serves as the foundational data module for the guidance system, responsible for the acquisition, synchronization, and preprocessing of all sensed data. Its hardware comprises an RSU device, a 77GHz millimeter-wave radar, a high-precision illuminance sensor, as well as temperature and humidity sensors, visibility sensors, a power quality monitor, an energy storage BMS, and an LED intelligent controller. Its core functions include PTP time synchronization, data cleaning, outlier removal, missing value imputation, multi-source feature fusion, and data standardization. The input is raw sensor data with a sampling frequency of 1Hz-100Hz, and the output is a standardized 10Hz time-series fusion dataset.

[0101] The edge computing decision layer includes a dynamic lighting matrix management module, an adaptive lighting control module, an energy storage microgrid scheduling module, and an optically encoded security guidance module.

[0102] The dynamic lighting matrix management module is used to implement step S2, namely: dividing the tunnel into multiple basic lighting units according to a preset interval, dynamically dividing the tunnel into multiple lighting areas based on the basic lighting units and vehicle data, and identifying the vehicle driving status based on the vehicle data.

[0103] This module undertakes the core function of dynamic division and refined management of tunnel lighting areas. It has built-in vehicle clustering algorithm and dynamic merging algorithm based on DBSCAN, which can complete basic lighting unit management; vehicle driving status recognition; automatic division of core, transition and basic safety lighting areas; intelligent merging of multi-vehicle lighting areas, etc. The input is vehicle position, speed and vehicle type data, and the output is accurate lighting area division results and vehicle driving status recognition results.

[0104] The adaptive lighting control module is used to implement step S3, namely: based on vehicle data, vehicle driving status, environmental data and tunnel location, calculate the lighting parameters of each lamp in each lighting area and generate lighting control instructions.

[0105] This module, as the core control module of the guidance system, is responsible for generating personalized lighting parameters adapted to different scenarios. It incorporates a lighting demand prediction model based on deep learning algorithms and a multi-objective optimization algorithm, enabling multi-dimensional adaptive adjustment of brightness, color temperature, and light emission angle; precise vehicle-following lighting control; smooth optimization of lighting at entrances and exits; and smoothing processing to suppress brightness abrupt changes. The module's inputs are vehicle characteristics, environmental data, and lighting area information; its outputs are independent brightness, color temperature, and flicker mode control commands for each LED.

[0106] The energy storage microgrid dispatch module is used to implement step S4, namely: generating energy storage charging and discharging strategies and corresponding energy storage dispatch instructions based on real-time electricity prices, grid load status, the state of charge of the energy storage system in the tunnel, and the lighting load forecast determined based on lighting control instructions.

[0107] This module, serving as the core of the energy management system, is responsible for the coordinated scheduling between the energy storage system, the power grid, and lighting loads. It incorporates a lighting load prediction model based on deep learning algorithms and a multi-objective optimization algorithm. Its core functions include: real-time grid price and load status monitoring, energy storage system state of charge and health management, lighting load prediction, charging and discharging strategy optimization calculation, and autonomous operation control during grid faults. The module's inputs are real-time grid price, grid load status, energy storage system operating parameters, and lighting load data corresponding to lighting control commands. Its outputs are energy storage charging and discharging strategies and corresponding energy storage scheduling commands.

[0108] The optical coding safety guidance module is used to implement step S5, namely: converting safety events in the tunnel into optical codes represented by lighting color, brightness and flashing mode, generating multiple guidance modes, and making decisions on multiple guidance modes according to preset priority rules to generate a safety guidance instruction that integrates lighting and safety guidance.

[0109] This module integrates lighting and safety guidance control without requiring additional guidance equipment. Its core functions include: analysis and classification of various safety events, optical encoding conversion of guidance modes, priority management of multiple guidance modes, generation of safety guidance instructions, and real-time evaluation of guidance effectiveness. The module's inputs are information on safety events such as accidents, construction, and congestion within the tunnel, as well as real-time vehicle distribution data. Its output is safety guidance instructions for the corresponding lighting areas.

[0110] The collaborative execution control layer includes a collaborative execution control module, which is used to implement step S6, namely: sending lighting control instructions, energy storage scheduling instructions and safety guidance instructions to the corresponding devices for execution, and collecting execution effect data in real time for feedback, so as to continuously optimize subsequent instructions.

[0111] This module is responsible for converting abstract decision commands issued by the edge controller into control signals that can be executed by the equipment. The hardware components of this module include an LED intelligent controller, a bidirectional AC / DC converter controller, and a power distribution switch controller. The core functions of this module include: command parsing and timing synchronization, real-time equipment status feedback, automatic fault detection and emergency handling. The input is the control commands issued by the edge controller, and the output is the equipment execution signal. Simultaneously, it uploads equipment operating status and fault information to the upper layer.

[0112] The cloud management and iteration layer includes a cloud management and iteration module, which enables remote centralized management of the boot system and continuous iterative optimization of the edge model. Its core functions include: real-time monitoring of the entire boot system device status, multi-dimensional energy consumption statistics and analysis, remote updating of edge model parameters, device fault alarms and remote diagnostics, and historical data backup and recovery. The module's inputs are the operating data and status information uploaded by the edge controller deployed in the tunnel, and its outputs are edge model update parameters and remote management commands.

[0113] The role and effect of the embodiments

[0114] The adaptive energy-saving lighting and safety guidance method and system for tunnel vehicle lights in coordination with the present invention has the following beneficial effects:

[0115] (1) Vehicle-light-energy deep collaborative architecture: This invention breaks through the limitations of the existing technology of vehicle-light separation and light-energy disconnection, deeply integrates vehicle perception, lighting control and energy scheduling, realizes the collaborative optimization of the three, and solves the core contradiction of energy saving but not safety and safety but not energy saving in the existing technology.

[0116] (2) Significantly improved control granularity: In step S2 of the present invention, a basic lighting unit with a preset spacing (e.g., 50m) is used for division, and based on the basic lighting unit, a fine division of the core lighting area, transition lighting area and basic safety lighting area with decreasing brightness centered on the vehicle is realized, which effectively solves the problems of lighting blind spots and over-lighting in the prior art.

[0117] (3) Multi-dimensional adaptive lighting control: In step S3 of the present invention, multi-dimensional features such as vehicle type, vehicle speed, vehicle driving status, ambient illuminance and tunnel location are integrated to independently control and smooth the brightness, color temperature and light output angle of each lamp, taking into account both energy saving effect and driver visual comfort.

[0118] (4) Constructing an energy storage-lighting microgrid: In step S4 of this invention, an energy storage charging and discharging strategy is generated based on real-time electricity price, grid load status and lighting load prediction, which realizes grid load smoothing, peak-valley electricity price arbitrage and islanded operation, significantly reducing electricity costs and improving power supply reliability.

[0119] (5) Optical coding safety guidance technology: In step S5 of the present invention, safety events in the tunnel are converted into optical codes represented by lighting color, brightness and flashing mode, generating multiple guidance modes, and generating safety guidance instructions that integrate lighting and safety guidance based on the guidance modes. No additional guidance equipment is required, reducing construction and maintenance costs. At the same time, the guidance effect is improved through a multi-mode priority decision mechanism.

[0120] (6) Fast response speed and high control precision: In step S6 of the present invention, the lighting control command, energy storage scheduling command and safety guidance command are sent to the corresponding equipment for execution. During the command sending and execution process, the lamp response time is ≤100ms, which is better than the response time of more than 1s of the existing system, thus avoiding lighting flicker and lag.

[0121] (7) Highly applicable and low cost of transformation: The method of this invention is fully compatible with existing LED lighting systems, vehicle-road cooperative equipment and energy storage systems, without the need for large-scale replacement of infrastructure; the modular architecture facilitates functional expansion and maintenance.

[0122] (8) Wide adaptability: The method of the present invention can be adapted to urban tunnels and highway tunnels of different lengths and traffic volumes, and can also be extended to lighting scenarios such as underground parking lots and urban expressways, with a wide range of applications.

[0123] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for adaptive energy-saving lighting and safety guidance that coordinates with vehicle lights in tunnels, characterized in that, Includes the following steps: S1: Multi-source data acquisition and fusion step, which involves synchronously acquiring multi-source data through sensing devices deployed in the tunnel, and performing time alignment and preprocessing on the acquired multi-source data to obtain a standardized time-series dataset; the multi-source data includes vehicle data, environmental data, energy data and equipment data; S2: Dynamic lighting matrix management step, the tunnel is divided into multiple basic lighting units according to a preset interval, and based on the basic lighting units and the vehicle data, the tunnel is dynamically divided into multiple lighting areas according to the real-time vehicle distribution, and the vehicle driving status is identified according to the vehicle data; S3: Adaptive lighting control step, based on the vehicle data, the vehicle driving status, the environmental data and the tunnel location, calculates the lighting parameters of each lamp in each of the lighting areas and generates a lighting control command; S4: Energy storage microgrid scheduling steps: Based on real-time electricity price, grid load status, state of charge of the energy storage system in the tunnel, and lighting load prediction determined based on the lighting control command, generate energy storage charging and discharging strategies and corresponding energy storage scheduling commands. S5: The optical coding safety guidance step converts safety events in the tunnel into optical codes represented by lighting color, brightness, and flashing mode, generates multiple guidance modes, and makes decisions on the multiple guidance modes according to preset priority rules to generate a safety guidance instruction that integrates lighting and safety guidance. S6: Coordinated execution control steps, issuing the lighting control command, the energy storage scheduling command, and the safety guidance command to the corresponding devices for execution, and collecting execution effect data in real time for feedback to continuously optimize subsequent commands. The division of the multiple lighting areas in S2 follows the following rules: Single vehicle scenario: Centered on the vehicle, the core lighting area is divided along the tunnel length according to the number of preset basic lighting units, and the transition lighting area is divided upstream and downstream of the core lighting area. The brightness of the transition lighting area decreases linearly with a gradient away from the vehicle. The remaining area is the basic safety lighting area, and the brightness of the basic safety lighting area is lower than that of the transition lighting area. Multi-vehicle scenario: When the distance between adjacent vehicles is less than a preset distance threshold, the overlapping core lighting area and transition lighting area of ​​the adjacent vehicles are merged into a continuous high-brightness lighting area, and the brightness of the high-brightness lighting area is the same as that of the core lighting area. In a car-free scenario: all basic lighting units inside the tunnel are set up as basic safety lighting zones.

2. The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination with those of claim 1, characterized in that: in, The method for time alignment of the multi-source data in S1 is as follows: multi-source data is synchronized at the microsecond level based on the PTP (Precise Time Protocol). The method for preprocessing multi-source data is as follows: outliers are removed based on the 3σ principle through the edge gateway; linear interpolation is used to fill multi-source data with ≤3 consecutive missing sampling points; multi-source data with >3 consecutive missing sampling points are marked as invalid and replaced with historical multi-source data; multi-source data with different sampling frequencies are uniformly resampled to the same frequency to complete spatiotemporal alignment and feature fusion.

3. The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination with the method described in claim 1. Its features are: The lighting parameters in S3 include: The target brightness is calculated based on the vehicle type and speed from the vehicle data, the ambient illuminance from the environmental data, and the tunnel location. The target color temperature is adjusted based on the vehicle's driving status and the current time of day. The target beam angle is adjusted based on the vehicle model data to avoid glare.

4. The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination with those of claim 3, characterized in that: in, The formula for calculating the target brightness in S3 is as follows: ， In the formula, Let be the luminance (cd / m²) of the j-th lamp in the i-th basic lighting unit; For tunnels The base brightness standard value for the location; For vehicle model coefficient; This is the vehicle speed coefficient; This refers to the ambient illuminance coefficient. This is the tunnel location coefficient.

5. The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination with claim 1, Its features are: The energy storage charging and discharging strategy in S4 includes: Charging strategy: When the electricity price is in a low-price period and the state of charge of the energy storage system is lower than the first preset threshold, it is charged at the rated power. When the grid load is lower than the preset load threshold and the state of charge of the energy storage system is lower than the second preset threshold, it is charged at half power. Discharge strategy: When the electricity price is at its peak and the state of charge of the energy storage system is higher than the third preset threshold, the energy storage system supplies power, and the power supply is matched with the lighting load corresponding to the lighting control command. When the grid load exceeds the preset load threshold, the energy storage system discharges to supplement the power supply.

6. The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination with those of claim 5, characterized in that: in, The energy storage charging and discharging strategy in S4 also includes: Autonomous operation strategy: When a power grid outage or power quality failure is detected, the grid connection is disconnected and the energy storage system supplies power independently to maintain the preset minimum safe lighting in the tunnel. When the grid is restored to normal, the grid connection is restored and the energy storage system is charged.

7. The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination with claim 1, Its features are: The boot mode in S5 includes: Normal driving mode: corresponds to constant white light; Forward deceleration warning mode: The corresponding yellow light gradually flashes, with the flashing frequency being the first preset frequency and the brightness gradually changing within the preset range; Accident or construction warning mode: The corresponding red light flashes alternately at a second preset frequency, which is higher than the first preset frequency. Adjacent lights are lit alternately to form a dynamic light strip to guide lane changes. Abnormal vehicle warning mode: corresponds to blue light pulse flashing, the flashing frequency is the third preset frequency, the third preset frequency is higher than the second preset frequency, and the brightness suddenly increases and decreases; Traffic congestion mode: The corresponding orange light remains constantly on, with a brightness higher than that of normal driving mode. The priority of the various guidance modes, from highest to lowest, is as follows: accident or construction warning mode, abnormal vehicle warning mode, congestion and slow traffic mode, forward deceleration warning mode, and normal driving mode.

8. The adaptive energy-saving lighting and safety guidance method for tunnel vehicle lights in coordination with those of claim 1, characterized in that: in, The method for continuously optimizing subsequent instructions in S6 is as follows: Based on deep reinforcement learning algorithms, the decision logic used to generate the lighting control commands, energy storage scheduling commands, and safety guidance commands is continuously optimized using execution performance data.

9. A tunnel vehicle-light coordinated adaptive energy-saving lighting and safety guidance system, used to execute the tunnel vehicle-light coordinated adaptive energy-saving lighting and safety guidance method as described in any one of claims 1-8, characterized in that, include: The multi-source data acquisition and fusion module synchronously acquires multi-source data through sensing devices deployed in the tunnel, and performs time alignment and preprocessing on the acquired multi-source data to obtain a standardized time-series dataset; the multi-source data includes vehicle data, environmental data, energy data, and equipment data; The dynamic lighting matrix management module divides the tunnel into multiple basic lighting units at preset intervals. Based on the basic lighting units and the vehicle data, it dynamically divides the tunnel into multiple lighting areas according to the real-time vehicle distribution and identifies the vehicle driving status based on the vehicle data. The adaptive lighting control module calculates the lighting parameters of each lamp in each of the lighting areas based on the vehicle data, the vehicle driving status, the environmental data, and the tunnel location, and generates lighting control instructions. The energy storage microgrid dispatch module generates energy storage charging and discharging strategies and corresponding energy storage dispatch instructions based on real-time electricity prices, grid load status, the state of charge of the energy storage system in the tunnel, and the lighting load forecast determined based on the lighting control instructions. The optical coding safety guidance module converts safety events in the tunnel into optical codes represented by lighting color, brightness, and flashing mode, generates multiple guidance modes, and makes decisions on the multiple guidance modes according to preset priority rules to generate a safety guidance command that integrates lighting and safety guidance. The collaborative execution control module sends the lighting control command, the energy storage scheduling command, and the safety guidance command to the corresponding devices for execution, and collects execution effect data in real time for feedback, so as to continuously optimize subsequent commands.

Images