Urban lighting adaptive adjustment method and system under digital twinning

Abstract

The invention discloses an adaptive adjustment method and system for urban lighting under digital twinning, and relates to the technical field of urban intelligent lighting, and the method comprises the steps: carrying out the twinning reconstruction of a lighting equipment network in an urban region, and determining a lighting twinning body; through urban distributed sensing, state initialization is carried out on a lighting twinborn body, lamp control constraint elements are introduced, each lighting device is used as a light environment intelligent agent, local game decision is executed, a lighting strategy is determined, and the lamp control constraint elements at least comprise comfort income, energy-saving income and glare punishment; and performing multi-thread instruction generation according to the lighting strategy, and performing lighting management and control on the lighting equipment network. According to the invention, the technical problems that the traditional urban illumination regulation and control are difficult to accurately match the actual demand and cannot balance the multi-aspect targets are solved, and the technical effects of adaptively adjusting the urban illumination, balancing the multi-aspect targets and improving the accuracy and effectiveness of illumination management and control are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of urban intelligent lighting, in particular to a city lighting adaptive adjustment method and system under digital twinning. BACKGROUND

[0002] City lighting is a key infrastructure for ensuring city night operation and improving resident activity experience. The regulation quality is directly related to light environment comfort, energy utilization efficiency and glare safety risk. In the prior art, city lighting relies on traditional means such as unified switch and fixed period adjustment, which can meet the basic lighting demand, but has obvious limitations in complex city scenes. The traditional regulation cannot realize differentiated adaptive adjustment for city lighting device network, and it is difficult to balance multiple targets, resulting in low matching degree of obtained lighting state data and actual demand, lack of precision of regulation strategy, and inability to meet the demand of precise evaluation and efficient management of city lighting. SUMMARY

[0003] The present application provides a city lighting adaptive adjustment method and system under digital twinning, which solves the technical problem that traditional city lighting regulation cannot accurately match actual demand and balance multiple targets.

[0004] In a first aspect, the present application provides a city lighting adaptive adjustment method under digital twinning, which comprises: performing twinning reconstruction on a lighting device network of a city area to determine a lighting twin; initializing the state of the lighting twin through city distributed sensing, introducing a light control constraint element, taking each lighting device as a light environment agent, executing local game decision to determine a lighting strategy, wherein the light control constraint element at least includes comfort benefit, energy saving benefit and glare penalty; and generating multi-thread instructions according to the lighting strategy to perform lighting management and control on the lighting device network.

[0005] In a second aspect, the present application provides a city lighting adaptive adjustment system under digital twinning, which comprises: a lighting twin acquisition module for performing twinning reconstruction on a lighting device network of a city area to determine a lighting twin; a lighting strategy acquisition module for initializing the state of the lighting twin through city distributed sensing, introducing a light control constraint element, taking each lighting device as a light environment agent, executing local game decision to determine a lighting strategy, wherein the light control constraint element at least includes comfort benefit, energy saving benefit and glare penalty; and a lighting management and control execution module for generating multi-thread instructions according to the lighting strategy to perform lighting management and control on the lighting device network.

[0006] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0007] The application carries out twin reconstruction on the urban lighting device network and determines the lighting twin, initializes the state of the twin by means of distributed sensing, introduces related constraint elements, executes local game decision with each lighting device as the main body to determine the lighting strategy, generates multi-thread instructions according to the strategy, and controls the urban lighting device network, so as to realize adaptive adjustment of urban lighting and balance of multiple demands, make the urban lighting control more suitable for the actual scene, improve the accuracy and effectiveness of lighting control, and achieve the technical effects of adaptive adjustment of urban lighting, balance of multiple targets, and improvement of the accuracy and effectiveness of lighting control. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0009] Figure 1 is a flowchart of the urban lighting adaptive adjustment method under digital twin provided by the embodiments of the present application.

[0010] Figure 2 is a structural schematic diagram of the urban lighting adaptive adjustment system under digital twin provided by the embodiments of the present application.

[0011] The figure mark explanation: the lighting twin acquisition module 1, the lighting strategy acquisition module 2, and the lighting control execution module 3. DETAILED DESCRIPTION

[0012] The present application provides an urban lighting adaptive adjustment method and system under digital twin, which solves the technical problems that the traditional urban lighting regulation and control are difficult to accurately match the actual demand and cannot balance multiple targets.

[0013] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0014] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices.

[0015] Embodiment one, as shown in the city lighting adaptive adjustment method under the digital twin, wherein the method comprises: Figure 1

[0016] The lighting device network of the urban area is twin reconstructed to determine the lighting twin.

[0017] In the embodiments of the present application, the lighting device network is a set of physical devices arranged in a certain layout in the urban area, including various lighting devices such as street lamps and landscape lamps, and has physical connection and control link.

[0018] Specifically, first, the basic data collection of the lighting device network of the urban area is carried out, and the collection content covers the specific position information of all lighting devices in the area, the device model parameters and the network connection topological relationship between devices. The collection method combines field survey means to record the actual installation position of the device and the surrounding environment related information, calls the factory account of the lighting device, obtains the technical parameters such as the model corresponding power, brightness adjustment range, and at the same time through network communication test, sorts out the data transmission link and control level relationship between devices.

[0019] Then, the collected basic data is processed. First, the data is cleaned to remove redundant or erroneous data entries, supplement missing key information, and ensure the integrity and accuracy of the data. Then, the processed device position data is matched with the city geographic information, and the device parameter data is standardized converted according to a unified data format, so that the parameters of different types of lighting devices have comparability and compatibility.

[0020] Then, based on the processed standardized data, a professional 3D modeling software is used to construct a digital model, and the specific process is as follows:

[0021] ​Step a: Perform geographic information mapping, associate the on-site location coordinates of each lighting device with urban geographic spatial data according to the actual physical structure of the lighting device network, accurately restore the spatial distribution of the devices in the digital environment, and at the same time, through network topology modeling, reproduce the communication connection link and control level relationship between devices, to ensure that the device connection logic in the digital model is consistent with the physical network.

[0022] Step b: Perform parameter attribute mapping, match and assign the technical parameters of each lighting device after standardization processing, such as power, brightness adjustment range, color parameter, etc., to the corresponding device unit in the digital model one by one, so that the digital unit has consistent basic attribute characteristics with the physical device.

[0023] Step c: Assign each device unit in the digital model with the corresponding functional characteristics of the physical lighting device, such as brightness adjustment function, color switching function, etc., and set the same operating parameter range as the physical device, such as brightness upper and lower limits, color adjustment interval, etc., to ensure that the digital model not only matches the physical device in structure and attributes, but also simulates the core operating functions of the physical device, so that the constructed digital model fully reflects the basic characteristics of the physical lighting device.

[0024] After that, the constructed digital model is verified and optimized. Compare the device distribution and network structure presented by the digital model with the on-site survey results, check whether there is a position deviation or connection relationship error. Test whether the digital model unit can normally respond to the simulation instruction by simulating the simple operating state of the lighting device, adjust and optimize the digital model according to the problems found in the verification process, correct the parameter setting or structure layout, and ensure that the digital model is consistent with the physical lighting device network in structure and basic functions.

[0025] The final verified and optimized digital model is the lighting twin, which fully retains the physical structure characteristics and basic attributes of the urban area lighting device network, and can be used as a digital carrier for subsequent state initialization and game decision.

[0026] Through city distributed sensing, the state of the lighting twin is initialized, and light control constraint elements are introduced, taking each lighting device as a light environment intelligent agent, performing local game decision, and determining the lighting strategy, wherein the light control constraint elements at least include comfort benefit, energy saving benefit and glare penalty.

[0027] In the embodiments of the present application, the light environment intelligent agent is an intelligent control unit corresponding to a single lighting device in the urban lighting device network, which has light control brightness, light control color and light control logic attributes, and can participate in collaborative game decision.

[0028] Optionally, first, the intelligent lighting management and control platform and the city distributed sensing end are established to communicate and interact, then the city light environment data are collected by the distributed sensing end in real time and returned to the data port of the intelligent lighting management and control platform, and finally, according to the collected light environment data, the initialization operation of the lighting twin is completed.

[0029] Then, taking the lighting twin as the basis, the light environment agent is constructed for each lighting device, the matching mapping of the light environment agent and the lighting twin is established, and the collaborative game training between the light environment agents is performed by the lamp control constraint element, so as to constitute the lighting regulation and control module, and the lighting regulation and control module is an embedded plug-in of the intelligent lighting management and control platform.

[0030] Finally, the first macro lighting demand is determined, which can be an objective lighting demand generated based on the light environment data, or a subjective lighting demand uploaded by the user end, then the lighting regulation and control module is driven to carry out the lamp control decision-making under the guidance of the first macro lighting demand, and finally the lighting strategy is generated.

[0031] According to the lighting strategy, the multi-thread instruction generation is performed, and the lighting management and control are performed on the lighting device network.

[0032] In an embodiment of the present application, each lighting device in the lighting device network is taken as a minimum control unit, the multi-thread instruction based on the previously determined lighting strategy is generated, then the multi-thread instruction is issued by the intelligent lighting management and control platform, and thus the control and management of the entire lighting device network are realized.

[0033] Further, the method provided in the embodiments of the present application comprises:

[0034] The communication interaction between the intelligent lighting management and control platform and the city distributed sensing end is established, the city light environment data are collected in real time by the distributed sensing end and returned to the data port of the intelligent lighting management and control platform, and according to the light environment data, the initialization of the lighting twin is performed.

[0035] In the embodiments of the present application, the intelligent lighting management and control platform is a city lighting comprehensive management and control system which can establish communication interaction with the city distributed sensing end, receive light environment data, initialize the lighting twin, drive the lighting regulation and control module to make decisions and issue multi-thread instructions. The city distributed sensing end is a set of light environment sensing devices distributed in the city area, which can collect light environment data in real time and return the data to the data port of the intelligent lighting management and control platform.

[0036] Specifically, first, the communication interaction between the intelligent lighting management and control platform and the city distributed sensing end is established, the communication module of the intelligent lighting management and control platform and the sensing node of the city distributed sensing end are configured with communication parameters in a manner of Internet of Things communication protocol adaptation and link verification, the configuration content covers IP address, port number, transmission rate and verification method, then the adaptive Internet of Things communication protocol such as MQTT protocol or LoRaWAN protocol is selected according to the communication coverage range and data transmission demand of the city area, to ensure that the protocol stack version of the platform and the sensing end is consistent, then the link connectivity test is performed, the intelligent lighting management and control platform sends a preset test data packet to the city distributed sensing end, the city distributed sensing end returns a response data packet after receiving, the intelligent lighting management and control platform verifies the integrity and transmission delay of the response data packet, if the verification result meets the preset threshold, the communication interaction is established, if not, the communication parameters are adjusted again and the link connectivity test is repeated until a stable communication link is established.

[0037] Then the light environment detection equipment deployed by the city distributed sensing end is started, including the illuminance sensor and the color temperature sensor, the collection period of each sensor is set to 30 seconds / time, the sensor detects the light environment data such as illuminance value and color temperature value of different monitoring points in the city area in real time according to the collection period, after the collection is completed, the sensing end encapsulates the light environment data according to the preset data format such as JSON format by using data standardization encapsulation technology, adds metadata such as collection timestamp and collection point coordinate, forms a standard data frame, and then transmits the standard data frame to the data port of the intelligent lighting management and control platform in a continuous data stream transmission manner through the established Internet of Things communication link. The data port of the intelligent lighting management and control platform receives the data frame and parses out the light environment data and metadata, and stores them in the platform database.

[0038] Finally, the lighting twin is initialized according to the light environment data, the intelligent lighting management and control platform identifies the abnormal value in the light environment data by using the 3σ criterion, supplements the missing value by using linear interpolation, and obtains clean effective light environment data. Then the effective light environment data is associated with the corresponding attribute field of each lighting device model in the lighting twin, for example, the illuminance value of a monitoring point is associated with the real-time environment illuminance attribute of the corresponding lighting device model of the point, the color temperature value is associated with the real-time environment color temperature attribute, etc. After the association is completed, the effective light environment data is written into the corresponding attribute field of the lighting twin, and the state parameters of the lighting twin are updated synchronously, so that the light environment state presented by the lighting twin is completely matched with the actual light environment state of the city, and the initialization of the lighting twin is completed.

[0039] Further, the method provided by the embodiment of the application comprises:

[0040] The comfort benefit is positively correlated with the overall regional illumination uniformity and color temperature coordination; the energy saving benefit is negatively correlated with the lighting device's own power consumption; the glare penalty is positively correlated with the caused glare risk.

[0041] Optionally, first, a light environment parameter monitoring system is deployed in a grid layout in a city area. The system includes a multi-channel illumination sensor and a high-resolution color temperature sensor. The instantaneous illumination values and color temperature values of different monitoring points are collected in real time through the sensor array. The spatial interpolation processing is performed on the discrete point data by using the Kriging interpolation method. The overall regional illumination uniformity, i.e., the ratio of the maximum illumination value to the minimum illumination value in the region, is calculated. At the same time, the color temperature deviation of each point is analyzed by comparing the color temperature value with the preset standard color temperature interval. The overall regional color temperature coordination is determined. Then, the power consumption monitoring module built in the lighting device is started. The module collects the device working current and voltage data in real time. The power consumption of each lighting device is converted by using the power calculation formula: power = voltage x current.

[0042] Then, a glare risk quantitative evaluation model is constructed. The CIE 117-1995 "Discomfort Glare in Interior Lighting" standard is taken as the core basis. The unified glare value (UGR) formula is determined as the core calculation logic of the model. The effective light source area calculation rules of non-uniform brightness lamps are supplemented according to the CIE 232:2019 standard. The spatial relative relationship calculation module of the light source and the observer is introduced according to the parameter dimension adaptation of the city lighting scene, and the basic model framework is formed. In the training process, first, the sample data of multiple scenes are collected. The physical parameters such as the average brightness of the light source and the background brightness are obtained by the light environment parameter monitoring system. The human physiological response data under different glare conditions are collected by using the eye tracker. The various glare image annotation information in the Flare7K data set is introduced. The training data set covering multiple scenes such as roads and squares is constructed. Then, the gradient boosting tree algorithm is used to fit and train the data set. The deviation between the measured glare index and the predicted value of the model is taken as the objective function. The weight coefficients of the Guth position index and the solid angle in the model are adjusted through cross-validation. The 3σ criterion is introduced to eliminate abnormal data. The fitting accuracy of the model is optimized. The deviation between the predicted value and the measured value is stable within ±5%. The input of the glare risk quantitative evaluation model includes the light intensity of the lighting device, the average brightness of the light source in the observer direction, the environmental background brightness, and other light parameters, as well as the distance between the light source and the observer, the line-of-sight deviation angle, and other spatial parameters. The output is the quantitative glare risk index, i.e., the UGR value directly reflecting different glare levels. Then, the glare risk quantitative evaluation model is used to calculate the glare index by combining the light intensity data of the lighting device and the environmental background light intensity data collected by the light environment parameter monitoring system. The glare risk caused is determined. The corresponding relationship between the comfort benefit, the energy saving benefit, and the glare penalty is determined.

[0043] Then, the Pearson correlation analysis and the multiple linear regression modeling method are used to establish the quantitative correlation between the light control constraint elements and the associated indicators. The collected regional overall illuminance uniformity and color temperature coordination data and the preset comfort benefit score sample set are input into the data analysis platform. The Pearson correlation analysis algorithm is used to calculate the correlation coefficient of the two. If the correlation coefficient is positive and the absolute value is greater than 0.7, it is determined that there is a strong positive correlation. Then, the multiple linear regression model is used to construct the comfort benefit quantitative function: comfort benefit value = a x regional overall illuminance uniformity + β x regional overall color temperature coordination, where a and β are the weight coefficients obtained by least squares fitting. The lighting device power consumption data and the energy saving benefit score sample set are input into the platform. The Pearson correlation analysis is used to obtain the negative correlation coefficient. The linear regression model is used to construct the energy saving benefit quantitative function: energy saving benefit value = γ - δ x lighting device power consumption, where γ is the baseline energy saving benefit value, and δ is the power consumption influence coefficient, which are determined by data fitting. The glare risk data and the glare penalty score sample set are input into the platform. After the Pearson analysis confirms the positive correlation, the linear regression model is used to construct the glare penalty quantitative function: glare penalty value = ε x glare risk, where ε is the glare risk influence coefficient, which is determined by data fitting. The quantitative correlation between the light control constraint elements and the associated indicators is completed.

[0044] Finally, the constraint condition coding embedding method is used to integrate the quantified light control constraint elements into the light control logic of the light environment agent. The control firmware of the light environment agent is called through the light control logic programming interface. The comfort benefit, energy saving benefit and glare penalty quantitative functions constructed above are written into the decision algorithm module of the control firmware in the form of code. At the same time, a data calling interface is set in the control firmware to ensure that the light environment agent can obtain the latest data of the regional overall illuminance uniformity, color temperature coordination, lighting device power consumption and glare risk in real time through the interface when making decisions. Then, the control firmware is debugged to simulate different light environment scenes, such as high illuminance scene, high power consumption scene and high glare risk scene, to verify whether the light environment agent can accurately calculate the values of the light control constraint elements through the quantitative functions. If the deviation between the calculation results and the preset theoretical values is within ±5%, it is determined that the embedding is successful. Finally, the light control constraint elements become the core evaluation basis for the light environment agent to make local game decisions, ensuring that the comfort, energy saving and glare prevention goals are considered in the decision-making process.

[0045] Further, the method provided by the embodiment of the present application comprises:

[0046] Based on the lighting twin, the light environment agent is constructed for each lighting device; the matching mapping of the light environment agent and the lighting twin is established, the collaborative game training between the light environment agents is performed by the light control constraint element, and a lighting regulation module is constituted, wherein the lighting regulation module is an embedded plug-in of the intelligent lighting management and control platform.

[0047] Specifically, for any lighting device in the lighting device network, the lighting device is taken as a first lighting device, and a corresponding light environment agent, i.e., a first light environment agent, is constructed by using the light control brightness, the light control chrominance and the light control logic. Meanwhile, the upper and lower limits of the light control brightness and the upper and lower limits of the light control chrominance are taken as the light environment boundary conditions and are written into the light control logic. This step is described in detail in the subsequent content.

[0048] Then, the matching mapping of the light environment agent and the lighting twin is established. First, the spatial coordinates of longitude and latitude and the installation height of each device in the lighting twin are obtained through GPS positioning data or a city geographic information system, and a unique device identification ID is assigned to each light environment agent. A mapping relationship table is created using an SQL database to store the light environment agent ID, the lighting twin device ID and the spatial coordinates in association, so as to realize the matching in the spatial position. Then, the core data fields of the light environment agent, such as the real-time brightness value, the real-time chrominance value and the running state, and the data fields of the corresponding devices in the lighting twin are sorted out. The field mapping rules are defined by using the JSON data format, and a data synchronization script is written to ensure that the state change of the light environment agent can be synchronized to the lighting twin in real time, and the device state update of the lighting twin can also be fed back to the light environment agent in a timely manner.

[0049] Then, the collaborative game training between the light environment agents is performed by the light control constraint element. First, a training environment is built in the simulation module of the intelligent lighting management and control platform, and the quantitative calculation functions of the comfort benefit, the energy-saving benefit and the glare penalty obtained in the foregoing steps are imported. Each light environment agent is set as an individual agent in the reinforcement learning, and the training target is set as the maximization of the comprehensive benefit of “comfort benefit + energy-saving benefit - glare penalty”. The Q-learning algorithm is used to initialize the decision Q table of each agent. In the training process, different light environment scenes are randomly generated, such as overcast evening, sunny noon and night peak, so that the light environment agent can adjust the brightness and chrominance of the light control strategy according to the scene data. After each round of strategy execution, the comprehensive benefit value is calculated, the decision weight in the decision Q table is updated according to the comprehensive benefit value, and the training is repeated for 500-1000 rounds until the comprehensive benefit value of each light environment agent is stabilized in a preset optimal interval, so as to realize the improvement of the collaborative game ability between the agents.

[0050] Finally, the trained light environment agent cluster, matching mapping logic, and collaborative game training algorithm are packaged as independent functional modules, encapsulated using Docker containerization technology, and a portable container image is generated to ensure stable operation of the modules in different hardware environments. Then, referring to the plug-in development specification provided by the intelligent lighting management platform, plug-in registration code is written to register the containerized functional modules as embedded plug-ins of the platform, load them through the platform's plug-in management interface, test the interaction functions of the modules and the platform, including receiving macro lighting requirements from the platform, uploading game decision results, and responding to state query instructions, and after confirming that all functions of the modules are normal, the lighting control module is constructed.

[0051] By the methods of digital twin model instantiation, spatial coordinate association and data field mapping, multi-agent reinforcement learning training, and modular encapsulation and plug-in integration, a lighting control module that can be embedded in the intelligent lighting management platform is constructed, providing a functional complete carrier for subsequent local game decision making by each lighting device as a light environment agent.

[0052] Further, the method provided by the embodiments of the present application comprises:

[0053] For the first lighting device, a first light environment agent is constructed based on lamp control brightness, lamp control chrominance, and lamp control logic, wherein the first lighting device is any lighting device in the lighting device network; wherein the upper and lower limits of the lamp control brightness and the upper and lower limits of the lamp control chrominance are used to deploy the light environment boundary conditions, and the lamp control logic is written.

[0054] Specifically, first, a first lighting device is determined from the lighting device network in the urban area, any lighting device is selected as the first lighting device through device number retrieval or physical location positioning, and basic information such as the model specification and installation location of the device is collected to determine its association in the lighting device network, which determines the target object for subsequent light environment agent construction.

[0055] Then, the lamp control brightness and lamp control chrominance parameters of the first lighting device are obtained, the adjustable range of the lamp control brightness, i.e., the upper and lower limits, is determined by real-time detection of the actual luminous brightness value of the device at different adjustment gears using a luminance meter, the upper and lower limits of the lamp control chrominance are determined by measuring the color temperature and tone data of the device in different working states using a colorimeter, and the existing control logic framework of the device is sorted out to determine the basic structure of the lamp control logic, thereby preparing the core elements for constructing the light environment agent.

[0056] Then deploy the light environment boundary condition and write the lamp control logic, based on the upper and lower limits of lamp control brightness and lamp control chrominance obtained in the above steps, convert these parameters into constraint code through a logic programming tool, embed the constraint code into the corresponding module of the lamp control logic according to the timing nodes of the device control flow, ensure that the lamp control logic can automatically identify and follow the preset boundary conditions when calling the brightness and chrominance adjustment functions, and avoid exceeding the reasonable adjustment range.

[0057] Finally, the determined lamp control brightness parameters, lamp control chrominance parameters, and the lamp control logic written with the light environment boundary conditions are imported into the agent development platform, and the three are integrated into a functional unit with independent decision-making ability through the built-in module packaging tool of the platform. Debug and verify the functional unit to ensure that it can respond to external control instructions and realize the autonomous adjustment of brightness and chrominance according to the lamp control logic, and finally complete the construction of the first light environment agent.

[0058] Further, the method provided by the embodiment of the application comprises:

[0059] determining a first macroscopic lighting demand, wherein the first macroscopic lighting demand is an objective lighting demand generated based on the light environment data, or a subjective lighting demand uploaded by a user end; and driving the lighting regulation module to make a lamp control decision and generate the lighting strategy in the light of the first macroscopic lighting demand.

[0060] Specifically, first, the light environment data is collected and compared with the standard by the city distributed sensing end to generate the objective lighting demand. The illuminance sensor and the color temperature sensor of the city distributed sensing end are started, and the illuminance value and the color temperature value of each region in the city are collected in real time at a frequency of 30 seconds / time. After collection, the original data collected for 5 times in succession are calculated by using the moving average method to obtain stable effective light environment data, so as to avoid the judgment deviation caused by single data fluctuation. Then, the city lighting standard data stored in the intelligent lighting management and control platform is called, the data including the lighting index requirements of different city regions such as city trunk roads, commercial squares, and residential areas in GB 50034-2013 “Building Lighting Design Standard”, and the effective light environment data is compared with the standard data of the corresponding region. If the actual illuminance value of a region is lower than the lower limit of the standard, the demand of “improving the lighting brightness of the region to the lower limit of the standard” is generated, and if the actual color temperature value deviates from the standard interval, the demand of “adjusting the lighting color temperature of the region to the standard interval” is generated, so as to obtain the objective lighting demand based on the light environment data.

[0061] Then, through user interaction and demand verification, an effective subjective lighting demand is obtained. A lightweight user interaction interface is built on the intelligent lighting management and control platform. The interface sets a demand area selection bar for listing the city road sections, squares and other areas managed and controlled by the platform; a demand type option, such as enhanced lighting, reduced lighting, and special scene lighting; and a demand time period input item. The user can complete the filling according to the actual demand and submit it. The submitted demand data is transmitted to the platform data receiving module through the MQTT communication protocol. The module calls the management and control area list and the regular lighting control period data to verify the demand, and marks it as an effective subjective lighting demand after confirming that the demand area is within the control range and the time period meets the regular control rules.

[0062] Then, through demand priority judgment, the first macro lighting demand is determined. The demand priority rules are set: the highest priority is the emergency subjective demand, the second is the regular subjective demand, and the lowest priority is the objective lighting demand. The intelligent lighting management and control platform first searches whether there is an effective subjective demand. If there is a high-priority emergency subjective demand, it takes the demand as the core and adjusts the parameters in combination with the corresponding regional objective lighting demand, for example, the user requires the square to be brightened to 30 lux after 1 hour, and the objective lighting demand is 20 lux. Then, 30 lux is taken as the target. If there is only a regular subjective demand or no subjective lighting demand, the objective lighting demand is directly used. Finally, the first macro lighting demand is integrated to form.

[0063] Finally, the first macro lighting demand is read, and each light environment agent determines the local lighting demand through lighting demand matching based on spatial phase. Then, each local lighting demand is transmitted to the matched light environment agent, and multiple local lighting strategies are determined through lighting decision. Finally, the local lighting strategies are adjusted through mutual game, and the final lighting strategy is generated. This step is described in detail in the subsequent content.

[0064] Further, the method provided by the embodiment of the present application comprises:

[0065] The first macro lighting demand is read, and each light environment agent determines the local lighting demand through lighting demand matching based on spatial phase. Then, each local lighting demand is transmitted to the matched light environment agent, and multiple local lighting strategies are determined through lighting decision. Finally, the local lighting strategies are adjusted through mutual game, and the final lighting strategy is generated.

[0066] In one embodiment, first, the intelligent lighting management platform calls the stored first macro lighting demand data, which contains key information such as target area range, lighting brightness target value, color temperature target interval, etc., adopts JSON format to parse the data, and extracts structured parameters that can be directly used for calculation. Subsequently, the space phase matching process is started. Each light environment agent has its own space coordinate information, i.e. longitude and latitude and installation height, pre-stored. The platform divides the area covered by the first macro lighting demand into grid units with an accuracy of 5m x 5m, calculates the space phase angle of each grid unit center coordinate and the corresponding light environment agent, including the horizontal phase angle and the vertical phase angle. The horizontal phase angle is calculated by the longitude difference between two points, and the vertical phase angle is calculated by the tangent value of the height difference and the horizontal distance between two points. When the calculated phase angle is within the lighting coverage angle range of the light environment agent, the lighting demand of the grid unit, i.e. the brightness and color temperature targets based on the macro demand, is assigned to the light environment agent, so as to determine the corresponding local lighting demand of each light environment agent.

[0067] Then the intelligent lighting management platform adopts the MQTT protocol (Message Queue Telemetry Transport protocol) to encapsulate the local lighting demand corresponding to each light environment agent into a data frame, which contains fields such as agent unique identifier ID, local demand brightness value, local demand color temperature value, etc. The platform sends the data frame to the communication receiving end of the corresponding light environment agent at a preset communication frequency of 1 second / time. After receiving the data frame, the light environment agent calls the built-in lighting parameter decision module, which adopts a rule matching algorithm to read the pre-set light environment boundary conditions in the light control logic, i.e. the upper and lower limits of light control brightness and light control chroma, compare the brightness value in the local lighting demand with the upper and lower limits of brightness, and if the demand brightness value is within the upper and lower limit range, the brightness value is directly taken as the candidate parameter; if it exceeds the upper limit, the upper limit of brightness is taken, and if it is lower than the lower limit, the lower limit of brightness is taken. The processing logic of the color temperature value is consistent with that of the brightness value. Subsequently, combined with the basic control rules in the light control logic, such as adjusting the color temperature to 3000K-4000K during the evening period, the local lighting strategy of the light environment agent is finally determined, including the specific brightness adjustment value, color temperature adjustment value and execution timing, such as immediate execution or delayed execution for 5 seconds.

[0068] Finally, the first lighting game group is determined to be composed of the first lighting device and related lighting devices including physically adjacent lighting devices and light environment interaction lighting devices. The in-group local lighting strategy of the first lighting game group is adjusted and the result is added to the lighting strategy. Finally, the in-group local lighting strategy game adjustment of each lighting game group is carried out in parallel, and the final lighting strategy is formed by integrating these adjustment results. This step is described in detail in the subsequent content.

[0069] Further, the method provided by the embodiments of the present application comprises:

[0070] determining a first lighting game group, wherein the first lighting game group is composed of a first lighting device and related lighting devices, the related lighting devices including physically adjacent lighting devices and light environment interaction lighting devices; performing in-group local lighting strategy game adjustment on the first lighting game group, adding the lighting strategy; and performing in-group local lighting strategy game adjustment on each lighting game group in parallel, and integrating to form the lighting strategy.

[0071] Optionally, first, all lighting device spatial coordinate data stored in the lighting twin, i.e. latitude, longitude and installation height, is called by the intelligent lighting management and control platform, a clustering radius, such as 5 meters, is set by using the DBSCAN density clustering algorithm, the clustering radius can be determined according to the conventional spacing of urban lighting devices, and the devices with a physical location distance less than the clustering radius are divided into a set of potential physically adjacent lighting devices. Then, a light environment simulation module built in the platform is called, the module is based on the Radiance software simplified algorithm, inputs the rated luminous intensity, beam angle and other parameters of each device, simulates the light intensity superposition effect of the surrounding device lighting area when a single device is turned on, and if the illumination contribution of a device to the lighting area of another device exceeds 5% after being turned on, it is determined as the threshold of light environment interaction, then the two are classified into a set of light environment interaction lighting devices.

[0072] Subsequently, an optional lighting device in the lighting device network is selected as the first lighting device, which is combined with the lighting devices in the set of physically adjacent lighting devices and the set of light environment interaction lighting devices to form the first lighting game group. At the same time, each light environment agent of the corresponding lighting device perceives the light environment data of the area where it is located in real time through the sensor carried by itself, including illumination and color temperature, combines the local lighting demand to propose an initial brightness strategy, and obtains the initial brightness strategy of the agent corresponding to other lighting devices in the lighting game group through the MQTT protocol, and uses the weighted summation method to calculate the influence of neighbor strategies on the self light control constraint elements, wherein the weight is set according to the distance between lighting devices, for example, the brightness of adjacent lighting devices is too high, which may lead to an increase in the risk of glare in the area, and thus increase the glare penalty.

[0073] Next, a greedy decision-making and Nash equilibrium iterative method is used to adjust the local lighting strategy within the first lighting game group. Each lighting agent first executes a greedy decision based on its own needs: based on its own local lighting needs for brightness and color temperature, combined with the quantification functions of comfort benefits, energy-saving benefits, and glare penalties of the lighting control constraints, the comprehensive benefit value under different brightness strategies is calculated, where comprehensive benefit = comfort benefit + energy-saving benefit - glare penalty. The top 3 strategies with the highest comprehensive benefits are selected as the minimum strategy set. Subsequently, the higher-level coordination game optimization adjustment stage begins. Each lighting agent in the first lighting game group interacts with the strategies in the minimum strategy set through the distributed communication protocol ZigBee. After each round of interaction, each agent recalculates its own comprehensive benefit value under the combination of neighboring strategies. If the comprehensive benefit value of the current strategy is lower than that of the adjusted strategy, for example, reducing its own brightness can reduce the glare impact on neighbors, while increasing its own energy-saving benefits. Then update the strategy and enter the next round of interaction until the agent adjusts the strategy in three consecutive rounds in the dark environment. It is determined that the corresponding strategy has converged to a stable configuration. At this time, the stable configuration is used as the adjusted local lighting strategy of the first lighting game group and added to the overall lighting strategy.

[0074] Then, the smart lighting management platform calls Python's ThreadPoolExecutor multithreading framework to create a corresponding number of threads based on the number of lighting game groups. For example, 10 lighting game groups correspond to 10 threads. Each thread independently executes the greedy decision-making and Nash equilibrium iterative adjustment process described above, avoiding mutual blocking between the adjustment processes of different groups. After each thread completes the adjustment within its group, it uploads the adjusted local lighting strategy to the smart lighting management platform. Then, a data fusion algorithm is used, such as strategy merging based on regional coverage weights, where the core area has a higher weight than the edge area, to integrate the local lighting strategies of all lighting game groups into a unified lighting strategy covering the entire city's lighting equipment network. At the same time, the consistency of the strategy is verified, such as whether the brightness and color temperature parameters of the boundary areas of adjacent game groups are continuous, to avoid obvious abrupt changes in the light environment. If inconsistencies exist, local fine-tuning is performed, such as controlling the brightness difference in the boundary areas to within 50 lux.

[0075] The lighting game groups are determined by DBSCAN clustering and light environment simulation. The convergence of strategies within the groups is achieved by combining greedy decision-making and Nash equilibrium iteration. Then, the strategy integration is completed by multi-threaded parallel processing and data fusion. Finally, the effect of transforming multiple local lighting strategies into an overall lighting strategy that takes into account both device synergy and light environment stability is achieved.

[0076] Furthermore, the method provided in this application embodiment includes:

[0077] The lighting device network is taken as a minimum control unit to generate multi-thread instructions based on the lighting strategy; and the multi-thread instructions are issued by the intelligent lighting management and control platform to control and manage the lighting device network.

[0078] In one embodiment, the determined lighting strategy data is first read by the intelligent lighting management and control platform, which contains the target brightness value, target chrominance value and other adjustment parameters of each lighting device. The strategy data is split into independent device adjustment tasks according to the device ID, and each task corresponds to a lighting device. Then, the Python threading multi-thread library is called to create threads matching the number of lighting devices, and each thread is responsible for converting the adjustment parameters of a single device into a standardized instruction format. In the conversion process, the parameters are mapped to the register address and value recognizable by the device by referring to the device hardware protocol Modbus RTU protocol, to generate multi-thread instructions containing device ID, instruction type and adjustment parameters, ensuring that each instruction only corresponds to the adjustment requirements of a single lighting device.

[0079] Then the intelligent lighting management and control platform establishes a stable connection with each lighting device through the MQTT communication protocol. During the connection process, device identity authentication is performed, such as device unique key verification, to confirm the legality and online status of the device. Then the multi-thread instruction distribution process is started, and each thread corresponds to a lighting device. The generated multi-thread instructions are accurately sent to the corresponding lighting device according to the device ID. The lighting device returns an acknowledgement response signal after receiving the instruction, and the intelligent lighting management and control platform records the instruction execution status after receiving the response. At the same time, the intelligent lighting management and control platform receives the running state data returned by each device in real time, such as the current brightness and power consumption, and compares it with the target parameters in the lighting strategy. If there is a deviation, the instruction fine-tuning process is triggered, and the corrected multi-thread instructions are regenerated and issued, to realize the continuous control and management of the lighting device network.

[0080] In summary, the urban lighting adaptive adjustment method based on digital twin provided by the embodiments of the present application has the following technical effects:

[0081] The present application determines the lighting twin by twin reconstruction of the urban area lighting device network, initializes the twin through city distributed sensing, introduces light control constraint elements containing comfort benefits, energy saving benefits and glare penalties, determines the lighting strategy by executing local game decision as a light environment agent for each lighting device, generates multi-thread instructions and issues them through the intelligent lighting management and control platform, realizes adaptive management and control of the lighting device network, improves the accuracy and efficiency of lighting adjustment, and achieves the technical effects of adaptive adjustment of urban lighting, balancing multiple goals, and improving the accuracy and effectiveness of lighting management and control.

[0082] Embodiment two, as Figure 2As shown, based on the same inventive concept as the preceding embodiment one, the embodiment of the present application provides a city lighting adaptive adjustment system under digital twinning, which comprises:

[0083] A lighting twin acquisition module 1, configured to perform twinning reconstruction on a lighting device network of a city area, and determine a lighting twin.

[0084] A lighting strategy acquisition module 2, configured to perform state initialization on the lighting twin through city distributed sensing, introduce light control constraint elements, take each lighting device as a light environment agent, perform local game decision, and determine a lighting strategy, wherein the light control constraint elements at least include comfort benefits, energy saving benefits, and glare penalties.

[0085] A lighting management and control execution module 3, configured to perform multi-thread instruction generation according to the lighting strategy, and perform lighting management and control on the lighting device network.

[0086] Further, the lighting strategy acquisition module 2 is configured to perform the following steps:

[0087] For each lighting device, a light environment agent is constructed based on the lighting twin; a matching mapping between the light environment agent and the lighting twin is established; and a collaborative game training between light environment agents is performed according to the light control constraint elements, to form a lighting regulation module, wherein the lighting regulation module is an embedded plug-in of a smart lighting management and control platform.

[0088] Further, the lighting strategy acquisition module 2 is configured to perform the following steps:

[0089] For a first lighting device, a first light environment agent is constructed according to light control brightness, light control chrominance, and light control logic, wherein the first lighting device is any lighting device in the lighting device network; and light environment boundary conditions are deployed according to upper and lower limits of the light control brightness and upper and lower limits of the light control chrominance, and the light control logic is written.

[0090] Further, the lighting strategy acquisition module 2 is configured to perform the following steps:

[0091] The comfort benefits are positively correlated with overall uniformity of area illumination and color temperature coordination; the energy saving benefits are negatively correlated with self-power consumption of the lighting device; and the glare penalties are positively correlated with glare risks caused.

[0092] Further, the lighting strategy acquisition module 2 is configured to perform the following steps:

[0093] Establish communication interaction of the intelligent lighting management and control platform with the city distributed sensing end; through the distributed sensing end, real-time collection of city light environment data is carried out, and the data is returned to the data port of the intelligent lighting management and control platform; according to the light environment data, the lighting twin is initialized.

[0094] Further, the lighting strategy acquisition module 2 is used to execute the following steps:

[0095] Determine the first macroscopic lighting demand, wherein the first macroscopic lighting demand is an objective lighting demand generated based on the light environment data or a subjective lighting demand uploaded by the user end; drive the lighting regulation module to make lamp control decisions based on the first macroscopic lighting demand, and generate the lighting strategy.

[0096] Further, the lighting strategy acquisition module 2 is used to execute the following steps:

[0097] Read the first macroscopic lighting demand, and each light environment agent performs lighting demand matching based on spatial phase to determine local lighting demand; transmit each local lighting demand to the matched light environment agent to make lighting decisions and determine multiple local lighting strategies; and mutual game adjustment is performed on the multiple local lighting strategies to generate the lighting strategy.

[0098] Further, the lighting strategy acquisition module 2 is used to execute the following steps:

[0099] Determine the first lighting game group, wherein the first lighting game group is composed of a first lighting device and related lighting devices, and the related lighting devices include physically adjacent lighting devices and light environment mutual influence lighting devices; game adjustment of the local lighting strategy in the first lighting game group is performed to add to the lighting strategy; and game adjustment of the local lighting strategy in each lighting game group is performed in parallel to integrate to form the lighting strategy.

[0100] Further, the lighting management and control execution module 3 is used to execute the following steps:

[0101] Each lighting device in the lighting device network is taken as a minimum control unit to generate a multi-thread instruction based on the lighting strategy; and the multi-thread instruction is issued by the intelligent lighting management and control platform to control and manage the lighting device network.

[0102] The city lighting adaptive adjustment system under the digital twin provided by the embodiment of the application can execute the city lighting adaptive adjustment method under the digital twin provided by any embodiment of the application, has the corresponding function modules and beneficial effects of the execution method.

[0103] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, the various units and modules are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific name of each functional unit is only for the convenience of mutual differentiation, and does not serve to limit the protection scope of the present application.

[0104] The above detailed description does not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application. In some cases, the actions or steps described in the present application can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

Claims

1. A method for adaptive adjustment of urban lighting under digital twin, characterized in that, The method includes: Perform a twin reconstruction of the urban area's lighting equipment network to determine the lighting twin; By using urban distributed sensing, the state of the lighting twin is initialized, lighting control constraints are introduced, and each lighting device is used as a light environment intelligent agent to perform local game decision-making to determine the lighting strategy. The lighting control constraints include at least comfort benefits, energy-saving benefits, and glare penalties. Multi-threaded instructions are generated based on the lighting strategy to control the lighting of the lighting device network.

2. The adaptive adjustment method for urban lighting under digital twin as described in claim 1, characterized in that, Before executing local game-theoretic decisions, each lighting device is treated as an intelligent agent for the light environment. A lighting control module is constructed, including: Based on the aforementioned lighting twin, an intelligent agent for the light environment is constructed for each lighting device; A matching mapping is established between the light environment intelligent agent and the lighting twin. The collaborative game training between the light environment intelligent agents is performed using the lighting control constraint elements to form a lighting control module, wherein the lighting control module is an embedded plug-in of the smart lighting management and control platform.

3. The adaptive adjustment method for urban lighting under digital twin as described in claim 2, characterized in that, For each lighting device, construct a smart agent for the light environment, including: For the first lighting device, a first light environment intelligent agent is constructed based on the lighting control brightness, lighting control chromaticity and lighting control logic, wherein the first lighting device is any lighting device in the lighting device network; Specifically, the upper and lower limits of the light control brightness and the upper and lower limits of the light control chromaticity are used to deploy the light environment boundary conditions, which are then written into the light control logic.

4. The adaptive adjustment method for urban lighting under digital twin as described in claim 3, characterized in that, The aforementioned comfort benefits are positively correlated with the overall uniformity of regional illuminance and color temperature coordination; The energy-saving benefits are negatively correlated with the power consumption of the lighting equipment itself; The glare penalty is positively correlated with the glare risk caused.

5. The adaptive adjustment method for urban lighting under digital twin as described in claim 4, characterized in that, The lighting twin is initialized in state using city-wide distributed sensing, including: Establish communication and interaction between the smart lighting management and control platform and the city's distributed sensing terminals; The distributed sensing terminal collects urban light environment data in real time and transmits it back to the data port of the smart lighting management and control platform. The lighting twin is initialized based on the light environment data.

6. The adaptive adjustment method for urban lighting under digital twin as described in claim 5, characterized in that, Each lighting device acts as an intelligent agent for the light environment, performing local game-theoretic decision-making to determine lighting strategies, including: Determine the first macro lighting requirement, wherein the first macro lighting requirement is an objective lighting requirement generated based on the light environment data, or a subjective lighting requirement uploaded by the user. Guided by the first macroscopic lighting demand, the lighting control module is driven to make lighting control decisions and generate the lighting strategy.

7. The adaptive adjustment method for urban lighting under digital twin as described in claim 6, characterized in that, Driving the lighting control module to make lighting control decisions includes: Read the first macroscopic lighting requirements, and each lighting environment agent performs lighting requirement matching based on spatial phase to determine the local lighting requirements; Transmit the lighting requirements of each local area to the matching light environment intelligent agent to make lighting decisions and determine multiple local lighting strategies. The multiple local lighting strategies are adjusted through mutual game theory to generate the lighting strategy.

8. The adaptive adjustment method for urban lighting under digital twin as described in claim 7, characterized in that, The lighting strategy is generated by performing game-theoretic adjustments on the multiple local lighting strategies, including: A first lighting game group is determined, wherein the first lighting game group consists of a first lighting device and related lighting devices, and the related lighting devices include physically adjacent lighting devices and lighting devices that interact with the light environment; The game adjustment of the local lighting strategy within the first lighting game group is carried out, and the lighting strategy is added. The game-theoretic adjustments of the local lighting strategies within each lighting game group are performed in parallel, and the resulting strategies are integrated to form the lighting strategy.

9. The adaptive adjustment method for urban lighting under digital twin as described in claim 1, characterized in that, Using each lighting device in the lighting device network as the smallest control unit, multi-threaded instructions based on the lighting strategy are generated. The lighting equipment network is controlled and managed by issuing multi-threaded commands through the intelligent lighting management and control platform.

10. A digital twin-based adaptive urban lighting control system, characterized in that, For implementing the urban lighting adaptive adjustment method under digital twins as described in any one of claims 1-9, the system comprises: The lighting twin acquisition module is used to perform twin reconstruction of the lighting equipment network in urban areas and determine the lighting twin; The lighting strategy acquisition module is used to initialize the state of the lighting twin through urban distributed sensing, introduce lighting control constraints, and perform local game decision-making with each lighting device as a light environment intelligent agent to determine the lighting strategy. The lighting control constraints include at least comfort benefits, energy-saving benefits and glare penalties. The lighting control execution module is used to generate multi-threaded instructions based on the lighting strategy and to control the lighting of the lighting equipment network.

Images