Storage yard gridding illumination self-adaptive control method and system based on satellite positioning

Abstract

The invention relates to a storage yard gridding illumination self-adaptive control method and system based on satellite positioning. The method comprises the steps of storage yard gridding and illumination unit deployment; collecting multi-source information in real time; digital twinborn construction and real-time synchronization are carried out; carrying out multi-dimensional simulation and optimization decision making based on twin bodies; and strategy execution and twinborn self-learning are carried out. The system comprises a physical perception and acquisition layer which comprises a storage yard gridding deployment module, an operation unit module and an environment perception module; edge computing nodes; the cloud digital twinborn platform is connected with the edge computing node through a cloud edge collaborative architecture and comprises a digital twinborn body construction and management unit, a simulation rehearsal and decision engine and a model training and optimization unit; an illumination control execution module; and a communication network module. According to the invention, comprehensive upgrading of the port storage yard lighting system in the aspects of self-adaptability, predictability, collaboration, safety and intelligence is realized.

Description

Technical Field

[0001] This invention relates to the field of port automation and smart lighting technology, and in particular to a satellite positioning-based adaptive control method and system for gridded lighting in storage yards. Background Technology

[0002] As the core area for cargo loading, unloading, storage, and transshipment, port yard lighting systems are crucial for ensuring operational safety and efficiency at night or in low-visibility environments. Traditional yard lighting often employs a centralized arrangement of high-mast lights with overall start-stop control. While this meets basic lighting needs, it has significant shortcomings: Firstly, it cannot provide precise lighting based on the actual location of the equipment, often resulting in a "full on / off" phenomenon and significant energy waste. Secondly, its lighting control strategy is simplistic and cannot adapt to complex and changing operational scenarios and environmental conditions, impacting operational safety and equipment lifespan.

[0003] In recent years, with the development of the Internet of Things, satellite positioning, and edge computing technologies, yard lighting control has gradually evolved towards intelligence and precision. For example, Chinese invention patent application CN113395810A discloses a method and system for precise lighting in port yards based on satellite positioning. This solution divides the yard into grids and deploys lighting units with satellite positioning coordinates at the grid intersections. Combined with positioning devices and status acquisition devices installed on the work units, the system utilizes on-site edge computing equipment to analyze the location of the work units in real time and control the switching of lights in the corresponding grids, thereby achieving the basic goal of "lighting up wherever the work goes." This comparative patent, to a certain extent, solves the problem that traditional lighting systems cannot light up on demand, and also reduces the losses caused by frequent switching of lights through strategies such as delayed shutdown.

[0004] However, with the continuous improvement of automation and digitalization in port operations, the above-mentioned basic solutions still have several limitations:

[0005] The control strategy is relatively simple: it mainly relies on the real-time position of the work unit for switching control, and lacks the ability to comprehensively perceive and adaptively adjust multi-dimensional information such as work type, ambient light, and weather conditions.

[0006] Lack of predictive and pre-control capabilities: The system only responds to the current position of the work unit and cannot predict its movement trajectory, making it difficult to achieve "pre-illumination" to ensure continuous good visibility during equipment movement.

[0007] Insufficient coordination and optimization capabilities: When multiple work units are working on-site at the same time, the system lacks a mechanism for coordinated scheduling and optimized allocation of global lighting resources, which may cause lighting overlap or blind spots, affecting overall visibility and energy efficiency.

[0008] The system has limited scalability and intelligence: it relies on on-site edge computing devices for real-time processing and has weak capabilities in complex strategy simulation, historical data learning, and deep integration with other port systems (such as scheduling systems and security systems).

[0009] Therefore, it is urgent to further integrate advanced technologies such as digital twins, artificial intelligence prediction, and multi-dimensional simulation optimization on the basis of existing grid-based and location-based lighting control, in order to build a more adaptive, predictive, and collaborative intelligent lighting control system to achieve a more efficient, safe, and energy-saving port yard operation environment. Summary of the Invention

[0010] The present invention aims to address the shortcomings of the prior art by providing a satellite positioning-based adaptive control method and system for gridded lighting in storage yards.

[0011] To achieve the above objectives, the present invention employs the following technical solution: a satellite positioning-based adaptive control method for gridded lighting in a storage yard, comprising the following steps:

[0012] S1. Yard Gridding and Lighting Unit Deployment: Divide the yard area into multiple grids, deploy independently controllable lighting units at the intersection of each grid, and establish satellite positioning coordinates for each intersection to form a position coordinate matrix;

[0013] S2. Real-time acquisition of multi-source information: Real-time acquisition of satellite positioning data, operation type identification, and environmental perception data of work units entering the yard; environmental perception data includes at least one of visibility, illuminance, and meteorological information.

[0014] S3. Digital Twin Construction and Real-time Synchronization: Based on the gridded layout of the yard, the deployment location of lighting units, and the satellite positioning coordinate matrix, a yard lighting digital twin is constructed and maintained in the cloud or at the edge. The twin synchronizes in real time the satellite positioning data, operation type identification, environmental perception data, and status of each lighting unit of the work unit in the physical yard.

[0015] S4. Multi-dimensional simulation and optimization decision-making based on digital twins: In the digital twin, virtual simulation of lighting control strategies, including at least two of the following dimensions, is executed in parallel based on real-time synchronized data:

[0016] a. Job type adaptation dimension: Matching the preset lighting requirement parameters of the job unit;

[0017] b. Trajectory prediction and pre-illumination dimension: Based on the prediction of the motion trajectory of the work unit, the illumination pre-activation simulation of the target mesh in front is performed;

[0018] c. Environmental adaptive adjustment dimension: Dynamically adjust lighting brightness or color parameters based on environmental data;

[0019] Based on the parallel evaluation of illuminance distribution, energy consumption and safety indicators of each simulation strategy, a globally optimal lighting control strategy is generated.

[0020] S5. Strategy Execution and Digital Twin Self-Learning: The optimal strategy is converted into control commands for specific grid lighting units and issued for execution; at the same time, actual lighting effect data is collected and fed back to the digital twin to optimize the simulation model and decision-making algorithm.

[0021] Specifically, in step S4, the simulation of trajectory prediction and pre-illumination dimensions includes: using a time series prediction algorithm to predict the movement path of the work unit in the next 30 seconds to 2 minutes, and evaluating the effect of the gradient brightening pre-activation strategy on the lighting units of 1 to 3 grids ahead of the path in the simulation.

[0022] Specifically, in step S4, the simulation of the environmental adaptive adjustment dimension includes: establishing a mapping model between environmental parameters and lighting adjustment parameters; when the visibility in the simulation input data is lower than the first threshold or rain or snow is detected, evaluating multiple schemes in the simulation to increase the lighting brightness of the relevant area to the standard value of 120%-180%.

[0023] Specifically, step S4 also includes: simulating a multi-work unit collaborative scenario in the simulation of the digital twin; when the simulation detects that multiple work units enter adjacent or overlapping grids, automatically generating and evaluating a lighting area merging scheme to optimize overall energy consumption and avoid mutual interference.

[0024] Specifically, steps S3 and S5 are implemented using a cloud-edge collaborative architecture: the cloud is responsible for building and maintaining the master digital twin, and for periodically training models and optimizing strategies based on historical big data; the edge deploys lightweight twin replicas or receives optimized models from the cloud, and is responsible for performing real-time data synchronization, low-latency simulation and decision-making.

[0025] Specifically, the method also includes: in step S2, synchronously collecting real-time location data of personnel in the yard; in the digital twin simulation in step S4, simulating the scenario of personnel entering the work area, and prioritizing the evaluation and selection of lighting strategies that can simultaneously achieve work lighting and personnel safety warnings.

[0026] Specifically, the method also includes the following steps:

[0027] S6. Equipment Health Management: Establish a digital health profile for each lighting unit in the digital twin. By comparing the simulated expected operating parameters with the actual feedback operating parameters, perform fault prediction and location, and provide visual early warning in the twin interface.

[0028] A system for adaptive control of gridded lighting in a storage yard based on satellite positioning, comprising:

[0029] The physical sensing and acquisition layer includes:

[0030] Yard grid deployment module: used to divide the yard into multiple grids and deploy independently controllable lighting units at the intersection of each grid to form a position coordinate matrix based on satellite positioning coordinates;

[0031] Work unit module: Each work unit is equipped with a work status collector, which is used to collect and report its satellite positioning location, work status and work type identifier in real time;

[0032] Environmental sensing module: includes at least one of visibility sensors, illuminance sensors and meteorological sensors deployed at key locations in the yard, for real-time collection of yard environmental data;

[0033] Edge computing nodes: deployed on-site in the storage yard to receive and preprocess data from the physical sensing and acquisition layer in real time;

[0034] Cloud-based digital twin platform: Connected to edge computing nodes via a cloud-edge collaborative architecture, the platform includes:

[0035] Digital Twin Construction and Management Unit: Used to construct and maintain a digital twin of yard lighting based on the location coordinate matrix and lighting unit parameters, and to synchronize multi-source data from edge computing nodes in real time;

[0036] Simulation and decision engine: In a digital twin, based on real-time synchronized work unit data, environmental data, and lighting unit status, it performs parallel virtual simulations of lighting control strategies, including at least two dimensions such as work type adaptation, trajectory prediction pre-lighting, and environmental adaptive adjustment. Based on the evaluation of illuminance distribution, energy consumption, and safety indicators of the simulation results of each strategy, it generates the globally optimal lighting control strategy.

[0037] Model training and optimization unit: Based on historical data and feedback data, the simulation and decision-making models are continuously trained and updated;

[0038] Lighting control execution module: Used to receive control commands from the cloud digital twin platform or its authorized edge computing nodes, and to precisely control the switching, brightness and lighting angle of the lighting units in a specific grid;

[0039] Communication network module: Supports at least two of the following communication protocols: 5G, LoRaWAN, and NB-IoT. It is used to connect modules at each layer of the system and supports self-organizing network function between lighting unit nodes to achieve network redundancy.

[0040] In particular, the system also includes:

[0041] Personnel safety linkage module: including a positioning terminal that identifies the location of personnel in the yard through UWB or Bluetooth positioning technology, and an early warning unit that is linked with the lighting control execution module and the digital twin platform to trigger audible and visual warnings when personnel enter high-risk work grids;

[0042] Equipment health management module: Integrated into the lighting control execution module or digital twin platform, it is used to monitor the current, voltage and temperature data of each lighting unit in real time, and to perform performance analysis and life prediction through digital twin, so as to realize early warning and location of faults.

[0043] In particular, the system also includes an external system integration interface for:

[0044] By connecting with the port's operation scheduling system to obtain operation plans, the digital twin platform can simulate and optimize lighting plans in advance.

[0045] In conjunction with the port security monitoring system, real-time lighting status, work unit location, and early warning information are overlaid onto the monitoring video footage.

[0046] The beneficial effects of this invention are:

[0047] 1. This invention introduces a digital twin and a multi-dimensional parallel simulation decision-making mechanism. The system can not only respond to the current location, but also comprehensively consider the specific needs of the operation type, real-time environmental conditions (visibility, weather) and personnel safety factors, and dynamically adjust the brightness, color and even illumination angle of the lighting unit. This achieves a qualitative change from lighting based on location to optimizing lighting on demand, and greatly improves the accuracy of lighting and scene adaptability.

[0048] 2. This invention, through trajectory prediction and pre-lighting simulation, can predict the movement path of the work unit in the next 30 seconds to 2 minutes and pre-activate the lighting of 1 to 3 grids ahead with gradient brightening. This effectively overcomes the problem that the lighting response may lag behind the equipment's movement speed, ensuring that the work unit is always in a sufficient and abrupt lighting environment throughout the entire movement path, significantly improving the safety and continuity of operations during movement.

[0049] 3. To address the potential lighting overlap or conflict issues that may arise when multiple devices operate simultaneously, as discussed in the comparative patent, this invention specifically simulates collaborative scenarios involving multiple work units within a digital twin. The system can automatically generate and evaluate lighting area merging schemes. While meeting the illuminance requirements of all work units, it intelligently merges lighting areas, reducing the number of unnecessary lights turned on. This achieves optimal global energy consumption in complex work scenarios, resulting in more significant energy savings than single-device tracking lighting.

[0050] 4. This invention innovatively incorporates real-time personnel location data into the lighting control decision-making loop. When the digital twin simulation detects personnel entering or approaching a high-risk work grid, the system can prioritize and execute a lighting strategy that combines work lighting with personnel safety warnings (such as flashing specific light colors).

[0051] 5. This invention adopts a cloud-edge collaborative architecture. The cloud-based digital twin platform is responsible for complex model training, historical data analysis, and long-term strategy optimization, possessing powerful computing and learning capabilities. The edge side is responsible for real-time data synchronization and low-latency local simulation decision-making, ensuring real-time control. This architecture supports the operation of complex intelligent algorithms while ensuring rapid response of the system in critical control stages, resulting in higher scalability and intelligence.

[0052] 6. This invention establishes a digital health record for each lighting unit. By comparing the simulated expected parameters in the digital twin with the actual operational feedback data, it can achieve performance degradation analysis, fault prediction, and precise location of the lighting unit. This extends the system from simple control functions to predictive maintenance, enabling early warning of potential faults, scheduling maintenance, reducing the risk of sudden lighting outages, improving the reliability and availability of the entire lighting system, and reducing long-term operation and maintenance costs.

[0053] 7. This invention features a standardized external system integration interface, enabling seamless integration with existing port operation scheduling and security monitoring systems. This not only allows the lighting system to simulate and optimize lighting plans in advance based on operational schedules, improving planning efficiency, but also allows information such as lighting status, equipment location, and safety warnings to be overlaid on monitoring videos, achieving multi-dimensional information fusion display and unified command, thereby enhancing the overall intelligence level of port operation and management. Attached Figure Description

[0054] Figure 1 This is a flowchart of the method of the present invention;

[0055] Figure 2 This is a structural block diagram of the system of the present invention;

[0056] The following will describe in detail, with reference to the accompanying drawings, embodiments of the present invention. Detailed Implementation

[0057] The present invention will be further described below with reference to embodiments:

[0058] like Figure 1 As shown, a satellite positioning-based adaptive control method for gridded lighting in a storage yard includes the following steps:

[0059] S1. Yard Gridding and Lighting Unit Deployment: Divide the yard area into multiple grids, deploy independently controllable lighting units at the intersection of each grid, and establish satellite positioning coordinates for each intersection to form a position coordinate matrix;

[0060] In the specific implementation of grid division, the first step is to determine the yard boundaries, obstacles (such as fixed buildings and large equipment foundations), main passages, and different functional areas (such as bulk cargo storage areas, container areas, and roads) based on the overall layout plan of the yard and on-site measurements. Then, using the light distribution curves of the selected high-mast lights (lighting units), installation height, and illuminance calculation software (such as DIALux), the illuminance values ​​at the center and edges of the grid are simulated and calculated under different grid sizes. Through repeated simulations, a grid side length that meets the minimum illuminance standards, achieves good uniformity, and is economically reasonable (e.g., a common size between 30 and 50 meters) is determined. Finally, all divided grids are systematically and uniquely numbered (e.g., using a "row number-column number" format, such as G-05-12), and their physical extent is recorded. Each grid is considered an independent "lighting control area."

[0061] Each lighting unit is a smart node, comprising at least: a light source system, typically consisting of 12 or 24 LED lamps integrated on a single tower, each lamp having independent control capabilities, allowing for individual switching, dimming, and adjustment of the illumination angle (e.g., via motor drive) to achieve precise illumination of the grid in different directions; and a local controller, mounted on the lamp post, responsible for receiving and executing control commands from the upper-level system, driving the specific lamp actions, and may also integrate ambient light sensors, etc.

[0062] At a specific height on the mounting base or pole of each lighting unit, a high-precision GNSS receiver (such as RTK) is used to accurately measure its geographic coordinates (latitude, longitude, and elevation). These coordinates represent the satellite positioning coordinates of the grid intersection point, and their accuracy directly affects the accuracy of subsequent positioning matching. All coordinate data must be entered into the system database.

[0063] S2. Real-time acquisition of multi-source information: Real-time acquisition of satellite positioning data, operation type identification, and environmental perception data of work units entering the yard; environmental perception data includes at least one of visibility, illuminance, and meteorological information.

[0064] Specifically, regarding satellite positioning data: High-precision GNSS (Global Navigation Satellite System) receivers are installed on each work unit (such as bucket wheel stacker-reclaimers, container trucks, forklifts, etc.), supporting multiple systems such as BeiDou, GPS, and GLONASS. Models with RTK (Real-Time Kinematic) functionality are preferred to achieve sub-meter or even centimeter-level real-time positioning accuracy. The receivers can continuously and in real-time collect the three-dimensional geographic coordinates (longitude, latitude, elevation), speed, heading angle, and timestamp of the work unit. The positioning data is then packaged and uploaded to the edge computing node in real-time via an onboard communication terminal (supporting 5G, LTE-V2X, or Wi-Fi).

[0065] Regarding job type identification, it can originate from multiple automated or semi-automated sources: Manual / automatic selection via onboard terminal: The driver or operator selects the current job mode (e.g., "retrieving," "stacking," "horizontal transport") on the onboard human-machine interface (HMI). Integration with the Port Operations Scheduling System (TOS): Through the system interface, detailed work orders for the currently executing tasks of the job unit can be directly obtained from the TOS, thereby automatically identifying its job type and status. Automatic identification based on sensors or video: By using sensors (such as radar, weighing sensors) or AI cameras installed on the equipment, the status of the robotic arm and the loading situation can be automatically determined, indirectly inferring the job type. This identification is key to associating preset lighting requirement parameters (such as the basic illuminance and color temperature preference required for different jobs), enabling the lighting strategy to match specific operational activities. In addition to location and type, the engine status of the equipment and the on / off status of the lights can also be collected to help determine whether it is in active operation.

[0066] Collection of environmental perception data for the storage yard: Visibility data collection: Forward-scattering visibility meters are installed on high-mast lights at the intersections of four key grids around the storage yard (or according to the prevailing wind direction and terrain layout). This device quantitatively calculates meteorological optical range (MOR) by measuring the intensity of light scattering by suspended particles in the air. It provides the overall visibility status of the entire storage yard area or zone and is a core environmental parameter for determining whether to activate or enhance lighting and adjust lighting brightness levels. Illuminance data collection: Illuminance sensors are deployed near the centers of several representative grids or work surfaces within the storage yard. This allows for real-time monitoring of the combined illuminance levels of natural light (sunlight) and artificial light in various areas of the storage yard. This is used to verify lighting effectiveness and serves as a direct basis for triggering or adjusting artificial lighting at dusk, dawn, or when the weather changes rapidly, achieving "on-demand supplemental lighting". Meteorological information collection: Miniature weather stations are set up at appropriate locations in the storage yard. Monitoring parameters can include ambient temperature, humidity, wind speed, wind direction, rainfall / snowfall status, atmospheric pressure, etc. In severe weather conditions such as strong winds, heavy rain, and heavy snow, specific safety lighting and warning strategies can be directly triggered. In addition, combined with temperature and humidity data, the formation conditions of low visibility weather such as fog and haze can be more comprehensively assessed. Extreme weather information can also be used to trigger protective controls on the lighting equipment itself (such as adjusting the angle to prevent wind).

[0067] The edge computing nodes deployed on-site serve as data aggregation centers, responsible for: receiving raw data streams from GNSS terminals of all work units and various environmental sensors; assigning unified high-precision timestamps to all data from different sources to ensure consistency of data in the time dimension during subsequent analysis; performing preliminary data verification and removing obvious outliers; smoothing high-frequency positioning data to reduce jitter; quickly matching the raw coordinates of work units using the position coordinate matrix established by S1 and converting them into their logical grid numbers; and uploading the processed structured data (formatted as: {timestamp, device ID, grid number, work type, speed, visibility, illuminance, weather code...}) to the cloud-based digital twin platform via a high-speed, stable backhaul network (such as fiber optic, 5G), while also retaining a local copy for rapid decision-making.

[0068] This step integrates high-precision real-time positioning with gridded coordinates to provide core data support for precise "lights following vehicle movement" lighting; the introduction of job type identification and multi-dimensional environmental data enables the lighting strategy to evolve from a simple location response to intelligent decision-making that adapts to specific job requirements and environmental changes; continuous trajectory and environmental data streams make it possible to predict job unit paths, anticipate changes in lighting requirements, and achieve adaptive adjustment; multi-source information verifies and complements each other, and is uniformly processed through edge computing to build a reliable and consistent real-time data view, enhancing the overall robustness of the system.

[0069] S3. Digital Twin Construction and Real-time Synchronization: Based on the gridded layout of the yard, the deployment location of lighting units, and the satellite positioning coordinate matrix, a yard lighting digital twin is constructed and maintained in the cloud or at the edge. The twin synchronizes in real time the satellite positioning data, operation type identification, environmental perception data, and status of each lighting unit of the work unit in the physical yard.

[0070] Specifically, based on the gridded layout map, the deployment locations of lighting units, and their precise satellite positioning coordinate matrix generated in step S1, a high-fidelity 3D virtual model is constructed in the cloud or at the edge. This model not only includes the static geometric information of the yard geography and all lighting equipment, but also defines dynamically adjustable attributes (such as on / off status, brightness, and color temperature) for each lighting unit and each grid. This digital twin is synchronized with the real-time data stream converged by the edge computing nodes in step S2 via a communication network.

[0071] Synchronized data includes: Dynamic data: real-time location of all work units (matched to specific grids), work type identifier, and movement speed. Environmental data: visibility, illuminance, and current weather conditions in the storage yard. Equipment status: current on / off status, brightness, and health status (e.g., current, temperature) of each lamp in each lighting unit.

[0072] This step integrates the static spatial framework of S1 and the dynamic sensing data of S2 into a unified virtual platform by constructing and synchronizing a digital twin in real time. This platform is not only a real-time image of the physical system, but also the core computing environment for multi-dimensional simulation, optimization decision-making, and strategy pre-playing, providing a key carrier for the leap from "sensing" to "intelligent control".

[0073] S4. Multi-dimensional simulation and optimization decision-making based on digital twins: In the digital twin, virtual simulation of lighting control strategies, including at least two of the following dimensions, is executed in parallel based on real-time synchronized data:

[0074] a. Job Type Adaptation Dimension: Matching the preset lighting requirement parameters of the job unit; specifically, the system automatically matches the preset lighting requirement parameter library based on the "job type identifier" reported by the job unit (such as "forklift loading and unloading" or "stack-reclaimer stacking"). In the simulation, it evaluates the effect of providing the current grid with the most suitable illuminance value, color temperature, and even glare-free lighting scheme for the job.

[0075] b. Trajectory Prediction and Pre-illumination Dimension: Based on the predicted motion trajectory of the work unit, a pre-activation simulation of illumination for the target grid ahead is performed. Specifically, the simulation of trajectory prediction and pre-illumination includes: using a time-series prediction algorithm to predict the movement path of the work unit within the next 30 seconds to 2 minutes, and evaluating the effectiveness of a gradient pre-activation strategy for the illumination units of 1 to 3 grids ahead of the path in the simulation. For example, the brightness of the grid to be entered is increased by 30% in advance, and the brightness of the next grid is increased by 10% in advance to ensure a smooth transition, ensuring that the field of vision ahead remains bright as the equipment moves, and eliminating illumination lag.

[0076] c. Environmental Adaptive Adjustment Dimension: Dynamically adjusts lighting brightness or color parameters based on environmental data. The simulation of this dimension specifically includes: establishing a mapping model between environmental parameters and lighting adjustment parameters; when visibility in the simulation input data falls below a first threshold or rain / snow is detected, evaluating multiple schemes to increase the lighting brightness of the relevant area to 120%-180% of the standard value during simulation. Specifically, the system has a built-in mapping model between environmental parameters and lighting adjustment parameters (e.g., when visibility is <200 meters, the base brightness needs to be increased by X%). When the real-time synchronized visibility data falls below the threshold, or rain / snow is detected, the simulation engine automatically generates multiple brightness enhancement schemes (e.g., increasing the brightness of the relevant area to 120%, 150%, or 180% of the standard value) and evaluates their illuminance guarantee effect and energy consumption cost under different weather conditions.

[0077] Based on the parallel evaluation of illuminance distribution, energy consumption and safety indicators of each simulation strategy, a globally optimal lighting control strategy is generated.

[0078] Step S4 further includes: simulating a multi-work unit collaborative scenario in the digital twin simulation. When the simulation detects multiple work units entering adjacent or overlapping grids, it automatically generates and evaluates a lighting area merging scheme to optimize overall energy consumption and avoid mutual interference. Specifically, the system no longer controls the lights independently for each unit, but instead simulates and generates a lighting area merging scheme, i.e., using a slightly larger, continuous lighting area to cover all relevant work units. In the simulation, it compares the uniformity of illuminance distribution, total energy consumption, and the existence of mutual glare interference between the "independent lighting" and "merged lighting" strategies, thereby intelligently selecting the most energy-efficient and comfortable scheme while ensuring the safety of all operations.

[0079] All simulation schemes across the above dimensions will be executed in parallel at the same time, based on the same real-time data snapshot. After the simulation, the system will score each scheme according to a preset multi-objective evaluation system (the core of which includes: whether the illuminance meets the standard and is evenly distributed, whether the total energy consumption is minimized, and whether the safety redundancy is sufficient). Finally, the decision engine will integrate the scores and automatically generate a globally optimal lighting control strategy that simultaneously considers safety, efficiency, and energy saving, and prepare to issue it for execution.

[0080] This step, through parallel simulation, anticipates various possibilities (where the equipment will go, how the weather will change, and how multiple devices will cooperate), and through rigorous quantitative evaluation, proactively selects the best course of action, achieving a fundamental leap in lighting control from passive response to proactive, collaborative, and global optimization.

[0081] S5. Strategy Execution and Digital Twin Self-Learning: The optimal strategy is converted into control commands for specific grid lighting units and issued for execution; at the same time, actual lighting effect data is collected and fed back to the digital twin to optimize the simulation model and decision-making algorithm.

[0082] Specifically, the globally optimal lighting control strategy generated in step S4 based on the digital twin is converted into a specific, executable set of device control instructions. For example, the strategy "provide 150% standard brightness for grid G-05-12" is specifically broken down into: controlling the 12 lamps facing the grid on the four lighting units (numbered L23, L24, L35, and L36) serving the grid, adjusting the brightness of each lamp to 150% of the set value. The control instructions are issued through a cloud-edge collaborative architecture. Typically, the cloud-based digital twin platform sends the instructions to the edge computing node, which then accurately and reliably sends them to the local controller of the target lighting unit via communication networks such as LoRaWAN and 5G, driving the lamps to perform actions such as switching on and off and dimming. This process must ensure low latency and high reliability to achieve real-time response of lighting to the work scenario.

[0083] After the control command is executed, the system synchronously collects actual lighting effect data as feedback. This data includes not only the environmental perception data already in S2 (such as actual visibility and illuminance changes), but also data from newly added verification sensors, as well as the operating status reported back by the lighting unit itself (such as actual power consumption and luminaire temperature). This actual effect data is fed back to the digital twin in real time. The twin compares and analyzes the "actual effect" with the "predicted effect" from the previous simulation. If a persistent deviation is found (for example, the simulation predicts that increasing the brightness to 150% will meet the visibility requirement, but the actual feedback shows that 180% is still needed), the system will use this deviation data to automatically or semi-automatically optimize its core model and decision-making algorithm, namely: correcting the mapping model between environmental parameters and lighting adjustment, adjusting the parameters of the trajectory prediction algorithm to improve accuracy, and updating the preset lighting requirement parameter library for the job type.

[0084] Steps S3 and S5 are implemented using a cloud-edge collaborative architecture: the cloud is responsible for building and maintaining the master digital twin, and for periodically training models and optimizing strategies based on historical big data; the edge deploys lightweight twin replicas or receives optimized models from the cloud, and is responsible for performing real-time data synchronization, low-latency simulation and decision-making.

[0085] Specifically, the cloud leverages its powerful computing capabilities to perform deep learning and model optimization that do not require real-time processing, periodically (e.g., daily or weekly) deploying the trained, more intelligent models and policy libraries to the edge. The edge, using the latest models and real-time data, independently performs instantaneous decision-making and control in most situations, forming a low-latency, highly reliable autonomous control loop. While executing decisions locally, the edge continuously uploads operation logs and performance feedback data to the cloud, serving as fuel for the next round of model training, thus creating a continuous self-evolutionary closed loop of "edge execution - cloud learning - model deployment - edge upgrade."

[0086] Through a cloud-edge collaborative architecture, the key tasks of "building and maintaining" the digital twin in step S3 and "strategy execution and self-learning" in step S5 are scientifically allocated. Working together, they construct a high-efficiency lighting control system that possesses advanced intelligence and meets the stringent real-time requirements of industrial environments.

[0087] The method also includes: in step S2, synchronously collecting real-time location data of personnel in the yard; in the digital twin simulation in step S4, simulating the scenario of personnel entering the work area, and prioritizing the evaluation and selection of lighting strategies that can simultaneously achieve work lighting and personnel safety warnings.

[0088] Specifically, personal positioning terminals are provided to personnel entering the yard area (such as inspectors, maintenance workers, and dispatchers). These terminals can employ UWB (Ultra-Wideband), Bluetooth beacon, or fusion positioning technologies to achieve real-time, high-precision (typically sub-meter level) positioning of personnel within the yard's gridded coordinate system. Personnel location data (including grid number, movement speed, and identification) is aggregated in real-time to edge computing nodes via a dedicated network, along with work unit data and environmental perception data, and synchronously updated to the digital twin. This allows the digital twin to map not only the status of all equipment and the environment but also the real-time location of each person, achieving comprehensive digitization of the "person, machine, and environment" status.

[0089] During multi-dimensional simulations using a digital twin, the system continuously monitors the spatial relationship between personnel positions and the activity grid of the work unit. Once the simulation engine predicts or detects that personnel are entering or about to enter a "high-risk grid" where equipment is currently or will soon be active, it automatically triggers a dedicated "personnel safety linkage" simulation scenario. In this specific simulation, the system no longer simply optimizes illuminance and energy consumption, but prioritizes generating and evaluating lighting control strategies that simultaneously achieve two objectives: Objective 1 (Operational Assurance): Ensure that the illuminance of the work unit and the target grid meets its operational requirements. Objective 2 (Personnel Warning): Employ enhanced active visual warnings for grids where personnel are located, have entered, or are nearby. For example, the simulation evaluates the following options: controlling relevant lighting units to change the light directed towards the personnel area to a specific warning color (such as flashing amber), increasing the brightness of the area to create strong contrast, or manipulating the lights to generate dynamic light strips to indicate safe evacuation directions. The simulation engine selects the strategy with the most significant warning effect, the least operational interference, and acceptable energy consumption from all strategies that simultaneously meet both objectives, as the highest priority strategy output for this scenario.

[0090] By incorporating personnel location data in step S2 and prioritizing personnel safety strategies in the simulation decision-making process of step S4, this invention makes the intelligent lighting system an indispensable part of the proactive safety system for port yards. It not only illuminates the work area but also, through intelligent beams of light, delineates clear safety boundaries for personnel, achieving a deep integration of production assurance and personal safety protection.

[0091] The method also includes: S6, Equipment health management: A digital health profile is established for each lighting unit in the digital twin. By comparing the simulated expected operating parameters with the actual feedback operating parameters, fault prediction and location are performed, and a visual warning is given in the twin interface.

[0092] Specifically, an independent digital health profile is created for each physical lighting unit (and each lamp and controller within it). This profile not only records the static information of the equipment (such as model, installation date, and location), but more importantly, it continuously records the history of its dynamic operating parameters throughout its entire lifecycle, such as: electrical parameters (operating current, voltage, power factor), status parameters (number of switching cycles, cumulative runtime, dimming level history), and environmental and performance parameters (lamp housing temperature, actual light output (which can be indirectly fed back through nearby illuminance sensors), and communication response time).

[0093] Based on the digital model of the lighting unit, current control commands (such as "turn on and adjust to 80% brightness"), and historical normal operation data, the simulation engine in the digital twin predicts the theoretical operating parameters that the device should have under this command (e.g., the expected current should be A amperes, and the housing temperature should be approximately T degrees Celsius). Actual operating parameters are collected in real time by sensors or smart meters installed on the lighting unit. The system continuously compares the actual feedback values ​​with the simulated expected values. When persistent or trending deviations occur, potential problems can be diagnosed: Example 1: Expected current is A, actual current is consistently 1.2A. This may indicate aging of the drive power supply or minor leakage / poor contact in the wiring. Example 2: Expected temperature should be T, actual temperature is consistently higher. This may indicate a failure in the cooling system or severe light decay inside the luminaire. Example 3: The expected grid illuminance should reach X Lux under the expected light contribution, but the sensor feedback is only 0.7X Lux. This may indicate severe light decay or lens contamination. The system can use a rule engine or machine learning model to match these deviation patterns with a typical fault database to achieve preliminary prediction and location of the fault type.

[0094] All health status information is visualized in the digital twin's management interface: against a yard map background, the health level of each lighting unit is intuitively marked with different colors (green / yellow / red). When the system predicts a potential fault, a warning message pops up on the virtual model of the corresponding equipment, clearly indicating the possible fault type (e.g., "Light fixture No. 3 of lighthouse L23 is suspected of light decay, estimated remaining lifespan < 500 hours"), location, and suggested inspection items. The system can automatically generate preventative maintenance work orders and optimize the inspection routes for maintenance personnel, achieving a closed loop from "alarm" to "work order dispatch".

[0095] like Figure 2 As shown, a system for adaptive control of gridded lighting in a storage yard based on satellite positioning includes:

[0096] The physical sensing and acquisition layer includes:

[0097] Yard grid deployment module: used to divide the yard into multiple grids and deploy independently controllable lighting units at the intersection of each grid to form a position coordinate matrix based on satellite positioning coordinates;

[0098] Work unit module: Each work unit is equipped with a work status collector, which is used to collect and report its satellite positioning location, work status and work type identifier in real time;

[0099] Environmental sensing module: includes at least one of visibility sensors, illuminance sensors and meteorological sensors deployed at key locations in the yard, for real-time collection of yard environmental data;

[0100] Edge computing nodes: Deployed on-site in the storage yard, these nodes receive and preprocess data from the physical sensing and acquisition layer in real time. They are responsible for collecting, cleaning, and preprocessing massive amounts of real-time data (such as converting GPS coordinates to grid numbers), and running lightweight digital twins and low-latency decision-making algorithms to ensure millisecond-level response to emergencies (such as rapid equipment movement or personnel intrusion).

[0101] Cloud-based digital twin platform: Connected to edge computing nodes via a cloud-edge collaborative architecture, the platform includes:

[0102] Digital Twin Construction and Management Unit: Used to construct and maintain a digital twin of yard lighting based on the location coordinate matrix and lighting unit parameters, and to synchronize multi-source data from edge computing nodes in real time;

[0103] Simulation and decision engine: In a digital twin, based on real-time synchronized work unit data, environmental data, and lighting unit status, it performs parallel virtual simulations of lighting control strategies, including at least two dimensions such as work type adaptation, trajectory prediction pre-lighting, and environmental adaptive adjustment. Based on the evaluation of illuminance distribution, energy consumption, and safety indicators of the simulation results of each strategy, it generates the globally optimal lighting control strategy.

[0104] Model training and optimization unit: Based on historical data and feedback data, the simulation and decision-making models are continuously trained and updated;

[0105] Lighting control execution module: Used to receive control commands from the cloud digital twin platform or its authorized edge computing nodes, and to precisely control the switching, brightness and lighting angle of the lighting units in a specific grid;

[0106] Communication network module: Supports at least two of the following communication protocols: 5G, LoRaWAN, and NB-IoT. It is used to connect modules at each layer of the system and supports self-organizing network function between lighting unit nodes to achieve network redundancy.

[0107] The system also includes:

[0108] Personnel safety linkage module: including a positioning terminal that identifies the location of personnel in the yard through UWB or Bluetooth positioning technology, and an early warning unit that is linked with the lighting control execution module and the digital twin platform to trigger audible and visual warnings when personnel enter high-risk work grids;

[0109] Equipment health management module: Integrated into the lighting control execution module or digital twin platform, it is used to monitor the current, voltage and temperature data of each lighting unit in real time, and to perform performance analysis and life prediction through digital twin, so as to realize early warning and location of faults.

[0110] External system integration interface, used for:

[0111] By connecting with the port's operation scheduling system to obtain operation plans, the digital twin platform can simulate and optimize lighting plans in advance.

[0112] In conjunction with the port security monitoring system, real-time lighting status, work unit location, and early warning information are overlaid onto the monitoring video footage.

[0113] This invention achieves a comprehensive upgrade of port yard lighting systems in terms of adaptability, predictability, coordination, security, and intelligence by deeply integrating advanced technologies such as digital twins, artificial intelligence prediction, multi-source information fusion, and cloud-edge collaboration. It effectively solves the problems existing in the prior art, such as single control strategy, lack of prediction, difficulty in coordination, disconnect from personnel safety, and insufficient system scalability, and provides strong technical support for building safe, efficient, green, and intelligent modern ports.

[0114] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0115] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0116] The present invention has been described above by way of example. Obviously, the specific implementation of the present invention is not limited to the above-described manner. Any improvements made by adopting the inventive concept and technical solution of the present invention, or direct application to other occasions without modification, are all within the protection scope of the present invention.

Claims

1. A satellite positioning-based adaptive control method for gridded lighting in a storage yard, characterized in that, Includes the following steps: S1. Yard Gridding and Lighting Unit Deployment: Divide the yard area into multiple grids, deploy independently controllable lighting units at the intersection of each grid, and establish satellite positioning coordinates for each intersection to form a position coordinate matrix; S2. Real-time acquisition of multi-source information: Real-time acquisition of satellite positioning data, operation type identification, and environmental perception data of work units entering the yard; environmental perception data includes at least one of visibility, illuminance, and meteorological information. S3. Digital Twin Construction and Real-time Synchronization: Based on the gridded layout of the yard, the deployment location of lighting units, and the satellite positioning coordinate matrix, a yard lighting digital twin is constructed and maintained in the cloud or at the edge. The twin synchronizes in real time the satellite positioning data, operation type identification, environmental perception data, and status of each lighting unit of the work unit in the physical yard. S4. Multi-dimensional simulation and optimization decision-making based on digital twins: In the digital twin, virtual simulation of lighting control strategies, including at least two of the following dimensions, is executed in parallel based on real-time synchronized data: a. Job type adaptation dimension: Matching the preset lighting requirement parameters of the job unit; b. Trajectory prediction and pre-illumination dimension: Based on the prediction of the motion trajectory of the work unit, the illumination pre-activation simulation of the target mesh in front is performed; c. Environmental adaptive adjustment dimension: Dynamically adjust lighting brightness or color parameters based on environmental data; Based on the parallel evaluation of illuminance distribution, energy consumption and safety indicators of each simulation strategy, a globally optimal lighting control strategy is generated. S5. Strategy Execution and Digital Twin Self-Learning: The optimal strategy is converted into control commands for specific grid lighting units and issued for execution; at the same time, actual lighting effect data is collected and fed back to the digital twin to optimize the simulation model and decision-making algorithm.

2. The adaptive control method for gridded lighting in a storage yard based on satellite positioning according to claim 1, characterized in that, In step S4, the simulation of trajectory prediction and pre-illumination dimensions specifically includes: using a time series prediction algorithm to predict the movement path of the work unit in the next 30 seconds to 2 minutes, and evaluating the effect of the gradient brightening pre-activation strategy on the lighting units of 1 to 3 grids ahead of the path in the simulation.

3. The adaptive control method for gridded lighting in a storage yard based on satellite positioning according to claim 1, characterized in that, In step S4, the simulation of the environmental adaptive adjustment dimension specifically includes: establishing a mapping model between environmental parameters and lighting adjustment parameters; when the visibility in the simulation input data is lower than the first threshold or rain or snow is detected, evaluating multiple schemes in the simulation to increase the lighting brightness of the relevant area to the standard value of 120%-180%.

4. The adaptive control method for gridded lighting in a storage yard based on satellite positioning according to claim 1, characterized in that, Step S4 also includes: simulating a multi-work unit collaborative scenario in the digital twin simulation; when the simulation detects that multiple work units enter adjacent or overlapping grids, automatically generating and evaluating a lighting area merging scheme to optimize overall energy consumption and avoid mutual interference.

5. The adaptive control method for gridded lighting in a storage yard based on satellite positioning according to claim 1, characterized in that, Steps S3 and S5 are implemented using a cloud-edge collaborative architecture: the cloud is responsible for building and maintaining the master digital twin, and for periodically training models and optimizing strategies based on historical big data; the edge deploys lightweight twin replicas or receives optimized models from the cloud, and is responsible for performing real-time data synchronization, low-latency simulation and decision-making.

6. The adaptive control method for gridded lighting in a storage yard based on satellite positioning according to claim 1, characterized in that, The method also includes: in step S2, synchronously collecting real-time location data of personnel in the yard; in the digital twin simulation in step S4, simulating the scenario of personnel entering the work area, and prioritizing the evaluation and selection of lighting strategies that can simultaneously achieve work lighting and personnel safety warnings.

7. The adaptive control method for gridded lighting in a storage yard based on satellite positioning according to claim 1, characterized in that, The method also includes the following steps: S6. Equipment Health Management: Establish a digital health profile for each lighting unit in the digital twin. By comparing the simulated expected operating parameters with the actual feedback operating parameters, perform fault prediction and location, and provide visual early warning in the twin interface.

8. A system for implementing the satellite positioning-based adaptive control method for gridded lighting in a storage yard as described in any one of claims 1-7, characterized in that, include: The physical sensing and acquisition layer includes: Yard grid deployment module: used to divide the yard into multiple grids and deploy independently controllable lighting units at the intersection of each grid to form a position coordinate matrix based on satellite positioning coordinates; Work unit module: Each work unit is equipped with a work status collector, which is used to collect and report its satellite positioning location, work status and work type identifier in real time; Environmental sensing module: includes at least one of visibility sensors, illuminance sensors and meteorological sensors deployed at key locations in the yard, for real-time collection of yard environmental data; Edge computing nodes: Deployed at the yard site to receive and preprocess data from the physical sensing and acquisition layer in real time; Cloud-based digital twin platform: Connected to edge computing nodes via a cloud-edge collaborative architecture, the platform includes: Digital Twin Construction and Management Unit: Used to construct and maintain a digital twin of yard lighting based on the location coordinate matrix and lighting unit parameters, and to synchronize multi-source data from edge computing nodes in real time; Simulation and decision engine: In a digital twin, based on real-time synchronized work unit data, environmental data, and lighting unit status, it performs parallel virtual simulations of lighting control strategies, including at least two dimensions such as work type adaptation, trajectory prediction pre-lighting, and environmental adaptive adjustment. Based on the evaluation of illuminance distribution, energy consumption, and safety indicators of the simulation results of each strategy, it generates the globally optimal lighting control strategy. Model training and optimization unit: Based on historical data and feedback data, the simulation and decision-making models are continuously trained and updated; Lighting control execution module: Used to receive control commands from the cloud digital twin platform or its authorized edge computing nodes, and to precisely control the switching, brightness and lighting angle of the lighting units in a specific grid; Communication network module: Supports at least two of the following communication protocols: 5G, LoRaWAN, and NB-IoT. It is used to connect modules at each layer of the system and supports self-organizing network function between lighting unit nodes to achieve network redundancy.

9. The adaptive control system for gridded lighting in a storage yard based on satellite positioning according to claim 8, characterized in that, The system also includes: Personnel safety linkage module: including a positioning terminal that identifies the location of personnel in the yard through UWB or Bluetooth positioning technology, and an early warning unit that is linked with the lighting control execution module and the digital twin platform to trigger audible and visual warnings when personnel enter high-risk work grids; Equipment health management module: Integrated into the lighting control execution module or digital twin platform, it is used to monitor the current, voltage and temperature data of each lighting unit in real time, and to perform performance analysis and life prediction through digital twin, so as to realize early warning and location of faults.

10. The adaptive control system for gridded lighting in a storage yard based on satellite positioning according to claim 8, characterized in that, The system also includes an external system integration interface for: By connecting with the port's operation scheduling system to obtain operation plans, the digital twin platform can simulate and optimize lighting plans in advance. In conjunction with the port security monitoring system, real-time lighting status, work unit location, and early warning information are overlaid onto the monitoring video footage.

Images