Intelligent lighting dynamic operation and maintenance management system and method based on digital twinning

Abstract

The invention discloses an intelligent lighting dynamic operation and maintenance management system and method based on digital twinning, relates to the technical field of public safety monitoring and intelligent analysis, and solves the technical problem of lack of construction of a digital twinning model for outputting an operation and maintenance decision by fusing environment, operation and safety features. Comprising a data acquisition module, a data processing module, a data storage module, a data analysis module and an intelligent lighting digital twinning dynamic operation and maintenance decision module. Environment, operation state and image safety data are collected and transmitted to a cloud database through preprocessing and encryption, then an illumination environment comprehensive influence evaluation value is calculated by analyzing environment data, an operation state comprehensive influence evaluation value is calculated by analyzing operation state data, and an illumination safety comprehensive influence evaluation value is calculated by analyzing image safety data. And finally, constructing a digital twin model based on multiple evaluation values, outputting an operation and maintenance decision result after training, and dynamically adjusting a strategy according to real-time data.

Description

Technical Field

[0001] This invention belongs to the field of public safety monitoring and intelligent analysis technology, specifically relating to a smart lighting dynamic operation and maintenance management system and method based on digital twins. Background Technology

[0002] With the acceleration of urbanization, the number of urban lighting facilities has increased dramatically, and traditional manual inspection and maintenance methods can no longer meet the needs of efficient and accurate operation and maintenance. Existing systems often suffer from problems such as data silos, delayed maintenance response, and low operation and maintenance efficiency, making it difficult to achieve full lifecycle management and preventive maintenance of street light assets. Therefore, developing a smart lighting operation and maintenance management system that integrates multiple technologies is of great significance for improving the operation and maintenance management level of urban lighting facilities.

[0003] Defects and shortcomings of existing technology: Traditional lighting operation and maintenance management has the following problems: Passive maintenance: relies on manual inspections or fault reporting, resulting in delayed response and a high rate of sudden failures; Data silos: Asset information and operation and maintenance records are scattered and lack a unified analysis platform; Safety hazards: Problems such as equipment damage and cable aging are difficult to monitor in a timely manner.

[0004] In existing technologies, a single Internet of Things (IoT) cannot achieve data fusion and predictive maintenance, and there is an urgent need for an integrated and intelligent solution. Summary of the Invention

[0005] This invention aims to solve at least one of the technical problems existing in the prior art; to this end, this invention proposes a smart lighting dynamic operation and maintenance management system and method based on digital twins, used to solve or improve one of the following technical problems: Defects and shortcomings of existing technology: Traditional lighting operation and maintenance management has the following problems: Passive maintenance relies on manual inspections or fault reporting, resulting in delayed response and a high rate of sudden failures. Data silos: Asset information and operation and maintenance records are scattered and lack a unified analysis platform; Safety hazards: Problems such as equipment damage and cable aging are difficult to monitor in a timely manner.

[0006] To address the aforementioned problems, a first aspect of the present invention provides a smart lighting dynamic operation and maintenance management system based on digital twins, comprising the following modules: Data acquisition module: Collects environmental data from various sensors distributed on urban lighting facilities, as well as the operating status data of the lighting facilities. At the same time, it uses cameras to capture images of the lighting facilities to obtain image safety data of the lighting facilities. Data processing module: preprocesses multi-source data and transmits the data to the cloud via encrypted communication; Data storage module: Establishes a database in the cloud to store multi-source data in a categorized manner; Data Analysis Module: Analyzes environmental data and calculates the comprehensive impact assessment value of the lighting environment; analyzes operational status data and calculates the comprehensive impact assessment value of the operational status; analyzes image safety data and calculates the comprehensive impact assessment value of lighting safety. The smart lighting digital twin dynamic operation and maintenance decision module calculates the comprehensive impact evaluation value of lighting environment, operation status and safety based on data, which serves as the input features of the digital twin model. The digital twin model is constructed, trained using the training set and optimized through the validation set, and the model outputs operation and maintenance decision results.

[0007] Preferably, the data acquisition module includes: Multiple environmental sensors are installed on urban lighting facilities to collect data on temperature, humidity, light intensity, air quality, and precipitation; multiple operating status sensors are installed on the control units of the lighting facilities to collect operating status data, including current, voltage, power, and switch status data. At the same time, cameras are installed on the lighting facilities to collect camera data, including security data of the lighting facilities and their surroundings; A data acquisition terminal is deployed in the power supply box of the lighting facility to receive and process data from sensors and cameras. The communication module supports wireless or wired data transmission and sets the data acquisition frequency.

[0008] Preferably, the data processing module includes: The data acquisition terminal reads environmental and operational status data daily and performs preprocessing, including filtering, calibration, and formatting. The system collects image data of the lighting facilities and their surroundings using cameras, and then compresses, labels, and timestamps the images. The collected environmental data, operational status data, and image security data are transmitted to the cloud via a communication module, and encryption technology is used to encrypt them during the transmission process.

[0009] Preferably, the data storage module is used to establish a database in the cloud, classify and store environmental data, operating status data and image security data, index and back up the stored data, and perform desensitization processing on the image security data.

[0010] Preferably, the analysis of environmental data and the calculation of the comprehensive impact assessment value of the lighting environment includes: Multiple parameter values ​​in the environmental data are mapped to a standardized range. Specifically, the precipitation parameter value is assigned a value of 1 if there is precipitation on that day, and 0 otherwise. The standardized parameter values ​​are combined with their corresponding weights to obtain the comprehensive impact assessment value of the lighting environment, specifically:

[0011] in, The comprehensive impact assessment value of the lighting environment. , , , and The standardized values ​​for temperature, humidity, light intensity, air quality, and precipitation. , , , and These are the corresponding weighting coefficients.

[0012] Preferably, the step of analyzing the operating status data and calculating the comprehensive impact evaluation value of the operating status includes: Based on historical data, normal ranges were set for the parameter values ​​in the operational status data, and these parameter values ​​were mapped to standardized ranges. Weights were assigned to each parameter, and the weighted summation yielded the comprehensive operational status impact evaluation value. Specifically:

[0013] in, The comprehensive impact evaluation value of the operating status. , , and These are the standardized current value, voltage value, power value, and switch status value, respectively. , , and The corresponding weighting coefficients; The switch status value is obtained by dividing the actual number of switches by the maximum allowed number of switches for the lighting device.

[0014] Preferably, the analysis of image security data includes: The acquired image data is preprocessed, including noise reduction, contrast enhancement, and region segmentation. The region segmentation is based on the light intensity to segment the sky, buildings, roads, and lighting equipment areas. Features are extracted from the image based on three dimensions, and the corresponding illumination assessment value, energy consumption assessment value, and damage assessment value are calculated respectively. The lighting evaluation value is obtained by calculating the mean image brightness, standard deviation of brightness, and color temperature deviation, and then summing them after standardization and weighting; wherein, the color temperature deviation is obtained by dividing the actual color temperature by the target color temperature; The energy consumption assessment value is obtained by combining image analysis with lighting equipment tags to obtain the actual power, calculating the energy efficiency ratio, and then adding them together after standardization and weighting. The damage assessment value is obtained by standardizing and weighting the degree of equipment wear, equipment aging, and equipment contamination by identifying surface cracks and deformations of the lighting equipment using the Canny algorithm, color decay analysis, texture analysis, and standardizing the degree of equipment wear, equipment aging, and equipment contamination. The comprehensive impact assessment value of lighting safety is obtained by weighting and adding the illumination assessment value, energy consumption assessment value, and damage assessment value.

[0015] Preferably, the comprehensive impact assessment value for lighting safety includes: The comprehensive impact assessment value for lighting safety is as follows:

[0016] in, The comprehensive impact assessment value for lighting safety. , and These are the standardized illumination assessment value, energy consumption assessment value, and damage assessment value, respectively. , and These are the corresponding weighting coefficients.

[0017] Preferably, the smart lighting digital twin dynamic operation and maintenance decision-making module includes: Historical data is collected, and annotation rules are formulated, including maintenance time and energy consumption optimization coefficients. After labeling the maintenance time and energy consumption optimization coefficients, the data is cleaned and augmented, and then divided into training set, validation set and test set. Based on the collected data, the comprehensive impact assessment values ​​of the lighting environment, operation status, and lighting safety are calculated and used as input features for the digital twin model. A digital twin model is constructed, the training set is input into the digital twin model for training, and the model parameters are optimized through the validation set. The digital twin model outputs maintenance time and energy consumption optimization coefficients as operation and maintenance decision results. The trained model is deployed to the system, the evaluation value is updated based on real-time data, and operation and maintenance decisions are dynamically generated and sent to physical devices. Collect feedback data after physical equipment executes operation and maintenance decisions, and optimize the digital twin model based on the feedback data.

[0018] A second aspect of the present invention provides a method for dynamic operation and maintenance management of smart lighting based on digital twins, comprising the following steps: S1: Collect environmental data from various sensors distributed on urban lighting facilities and operational status data of the lighting facilities. At the same time, use cameras to collect images of the lighting facilities to obtain image safety data of the lighting facilities. S2: Preprocess the multi-source data separately and transmit the data to the cloud via encrypted communication; S3: Establish a database in the cloud to categorize and store multi-source data; S4: Analyze environmental data and calculate the comprehensive impact assessment value of the lighting environment; analyze operational status data and calculate the comprehensive impact assessment value of the operational status; analyze image safety data and calculate the comprehensive impact assessment value of lighting safety. S5: Based on data calculation, the comprehensive impact evaluation value of lighting environment, operation status and safety is used as the input features of the digital twin model. The digital twin model is constructed, trained using the training set and optimized through the validation set. The model outputs operation and maintenance decision results.

[0019] The beneficial effects of this invention are: This invention integrates environmental sensors, operational status sensors, and image security data to achieve comprehensive monitoring of multi-dimensional characteristics such as temperature, humidity, light intensity, equipment status, and safety hazards, thereby improving data integrity and accuracy. This invention integrates lighting environment, operating status, and safety comprehensive impact evaluation values ​​through a digital twin model, outputting decision results for maintenance time and energy consumption optimization coefficients. This reduces manual intervention, improves operation and maintenance response speed and resource utilization, and adopts a data closed-loop mechanism to iteratively optimize the digital twin model using feedback data. This adapts to complex scenarios such as equipment aging and environmental changes, improving the model's generalization ability and long-term stability. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the module flow of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 As shown, this invention is a smart lighting dynamic operation and maintenance management system based on digital twins, comprising the following modules: Data acquisition module: Collects environmental data from various sensors distributed on urban lighting facilities, as well as the operating status data of the lighting facilities. At the same time, it uses cameras to capture images of the lighting facilities to obtain image safety data of the lighting facilities. Data processing module: preprocesses multi-source data and transmits the data to the cloud via encrypted communication; Data storage module: Establishes a database in the cloud to store multi-source data in a categorized manner; Data Analysis Module: Analyzes environmental data and calculates the comprehensive impact assessment value of the lighting environment; analyzes operational status data and calculates the comprehensive impact assessment value of the operational status; analyzes image safety data and calculates the comprehensive impact assessment value of lighting safety. The smart lighting digital twin dynamic operation and maintenance decision module calculates the comprehensive impact evaluation value of lighting environment, operation status and safety based on data, which serves as the input features of the digital twin model. The digital twin model is constructed, trained using the training set and optimized through the validation set, and the model outputs operation and maintenance decision results.

[0023] In one embodiment of the present invention, the data acquisition module includes: Multiple environmental sensors are installed on urban lighting facilities to collect data on temperature, humidity, light intensity, air quality, and precipitation; multiple operating status sensors are installed on the control units of the lighting facilities to collect operating status data, including current, voltage, power, and switch status data. At the same time, cameras are installed on the lighting facilities to collect camera data, including security data of the lighting facilities and their surroundings; A data acquisition terminal is deployed in the power supply box of the lighting facility to receive and process data from sensors and cameras. The communication module supports wireless or wired data transmission and sets the data acquisition frequency.

[0024] Specifically, select suitable temperature, humidity, light intensity, air quality (PM2.5 / CO2), and precipitation sensors, and install them on or around representative locations on the lighting facility pole. Set temperature, humidity, and light intensity to be collected once per minute, and air quality and precipitation to be collected once per hour. Integrate current, voltage, and power sensors into the power supply line of the lighting facility. Install the on / off status sensor in the control circuit of the lighting equipment. The current, voltage, and power sensors collect data once per second, and the on / off status is triggered to collect data each time the status changes. Fix the camera on the top or side of the lighting facility pole, covering the lamps, surrounding roads, and area. Set the resolution to 1080P or higher, the frame rate to 1 frame / second, and the shooting range to include the lamp surface, the surrounding environment, and the area where pedestrians / vehicles are active. Install a data acquisition terminal in the power supply box of the lighting facility, connect all sensors and cameras, and support 4G / 5G wireless transmission or wired Ethernet. Select the appropriate terminal based on site conditions, set the acquisition frequency of each sensor through the terminal software, and set a timed snapshot (take a picture every 5 minutes).

[0025] In one embodiment of the present invention, the data processing module includes: The data acquisition terminal reads environmental and operational status data daily and performs preprocessing, including filtering, calibration, and formatting. The system collects image data of the lighting facilities and their surroundings using cameras, and then compresses, labels, and timestamps the images. The collected environmental data, operational status data, and image security data are transmitted to the cloud via a communication module, and encryption technology is used to encrypt them during the transmission process.

[0026] Specifically, the data acquisition terminal automatically reads data from environmental and operational status sensors at midnight every day, continuously collects and caches data at a set frequency, and performs preprocessing operations, including removing outliers, calibrating the data according to the sensor manual, and standardizing the data format to JSON or CSV, adding fields such as device ID, sensor type, timestamp, and calibrated values; the camera captures images of the lighting facilities and their surroundings at a set frequency, each shot covering the lamps, poles, and the surrounding environment, including roads and pedestrian areas. The raw images are compressed into JPEG format, the resolution is adjusted to 1280*720, and image processing tools are used to achieve [further processing]. Batch compression is performed, and original data tags are added, including device ID, geographic coordinates, shooting time, and image content description. A precise timestamp is added to each image for subsequent analysis. Environmental data and operational status data are grouped by device ID and timestamp, packaged into data packets, and grouped every 10 minutes. Image security data is packaged separately and associated with corresponding annotation information. Data is transmitted via communication module (4G / 5G or WiFi) using MQTT, HTTPS, or CoAP protocols. Data packets are encrypted using TLS / SSL protocols to ensure transmission security. Real-time data (such as switch status and alarm signals) is transmitted first, while non-real-time data (such as images) is uploaded at a set frequency.

[0027] In one embodiment of the present invention, the data storage module is used to establish a database in the cloud, classify and store environmental data, operating status data and image security data, index and back up the stored data, and perform desensitization processing on the image security data.

[0028] Specifically, time-series or relational databases are used to store environment and operational status data, object storage is used to store raw images, and the database stores image metadata. Virtual machines are created on the cloud platform or managed database services are used directly, and database parameters are configured. Data is inserted into the time-series database according to device ID and timestamp to ensure time-series continuity, and partitioned by month or device ID to improve query efficiency. Compressed image files are uploaded to object storage, and image metadata is stored in MongoDB or a relational database. A composite primary key index is created by device_id and timestamp to support fast queries. Indexes are built by device_id and timestamp to accelerate image retrieval. For environment and operational status data, the time-series database is incrementally backed up daily to a remote server or cloud storage. For image security data, cross-region replication of object storage is enabled, and metadata backups are exported periodically. The de-identification process involves using image recognition algorithms to detect sensitive targets, such as faces and license plates, applying Gaussian blur or pixelation to sensitive areas, sampling and checking the de-identified images to ensure sensitive information has been removed, overwriting the original files with the processed images or saving them as new files, and updating the image path in the metadata to the de-identified file path.

[0029] In one embodiment of the present invention, the step of analyzing environmental data and calculating the comprehensive impact assessment value of the lighting environment includes: Multiple parameter values ​​in the environmental data are mapped to a standardized range. Specifically, the precipitation parameter value is assigned a value of 1 if there is precipitation on that day, and 0 otherwise. The standardized parameter values ​​are combined with their corresponding weights to obtain the comprehensive impact assessment value of the lighting environment, specifically:

[0030] in, The comprehensive impact assessment value of the lighting environment. , , , and The standardized values ​​for temperature, humidity, light intensity, air quality, and precipitation. , , , and These are the corresponding weighting coefficients.

[0031] Specifically, the standardization range is defined as mapping all data to a range of 0 to 1. Temperature, humidity, and light intensity are linearly normalized, and air quality is directionally normalized to obtain values ​​within the standardization range. Precipitation parameters are assigned values ​​based on whether precipitation occurs: if precipitation > 0 mm, a value of 1 is assigned; otherwise, a value of 0 is assigned. Corresponding weighting coefficients are assigned based on the degree of impact of environmental data on lighting equipment. , , , and The corresponding weights are 0.2, 0.15, 0.3, 0.25 and 0.1, respectively. The calculated comprehensive impact assessment values ​​of the lighting environment are stored in the database for subsequent analysis. If a parameter is missing, the historical average is used instead, and the weight is dynamically adjusted according to the scenario. For example, during the rainy season, the weight of precipitation is increased.

[0032] In one embodiment of the present invention, the step of analyzing the operating status data and calculating the comprehensive impact evaluation value of the operating status includes: Based on historical data, normal ranges were set for the parameter values ​​in the operational status data, and these parameter values ​​were mapped to standardized ranges. Weights were assigned to each parameter, and the weighted summation yielded the comprehensive operational status impact evaluation value. Specifically:

[0033] in, The comprehensive impact evaluation value of the operating status. , , and These are the standardized current value, voltage value, power value, and switch status value, respectively. , , and The corresponding weighting coefficients; The switch status value is obtained by dividing the actual number of switches by the maximum allowed number of switches for the lighting device.

[0034] Specifically, based on historical data, the normal ranges for each operating status parameter are defined, including the switch status values, which are as follows: ,in, This represents the actual number of times the device was switched on and off. The maximum allowable number of switching operations for this lighting equipment is defined; based on the degree of influence of the operating status parameters on the equipment, corresponding weights are assigned. , , and The corresponding values ​​are 0.3, 0.25, 0.25 and 0.2 respectively; the comprehensive impact evaluation value of the operating status is calculated, with a value range from 0 to 1. The larger the value, the closer the operating status is to normal.

[0035] In one embodiment of the present invention, the analysis of image security data includes: The acquired image data is preprocessed, including noise reduction, contrast enhancement, and region segmentation. The region segmentation is based on the light intensity to segment the sky, buildings, roads, and lighting equipment areas. Features are extracted from the image based on three dimensions, and the corresponding illumination assessment value, energy consumption assessment value, and damage assessment value are calculated respectively. The lighting evaluation value is obtained by calculating the mean image brightness, standard deviation of brightness, and color temperature deviation, and then summing them after standardization and weighting; wherein, the color temperature deviation is obtained by dividing the actual color temperature by the target color temperature; The energy consumption assessment value is obtained by combining image analysis with lighting equipment tags to obtain the actual power, calculating the energy efficiency ratio, and then adding them together after standardization and weighting. The damage assessment value is obtained by standardizing and weighting the degree of equipment wear, equipment aging, and equipment contamination by identifying surface cracks and deformations of the lighting equipment using the Canny algorithm, color decay analysis, texture analysis, and standardizing the degree of equipment wear, equipment aging, and equipment contamination. The comprehensive impact assessment value of lighting safety is obtained by weighting and adding the illumination assessment value, energy consumption assessment value, and damage assessment value.

[0036] Specifically, image preprocessing includes denoising via Gaussian filtering or median filtering, contrast enhancement via histogram equalization or adaptive histogram equalization, and region segmentation using thresholding or clustering algorithms based on illumination intensity to segment the sky, buildings, roads, and lighting equipment regions. The image brightness channel's mean, standard deviation, and color temperature deviation are calculated. The color temperature deviation is calculated as the ratio of the actual color temperature to the target color temperature, standardized, and weighted to obtain the illumination evaluation value. Actual power is obtained through image analysis combined with lighting equipment labels, and the energy efficiency ratio is obtained by comparing the actual power to the standard power. The actual power and energy efficiency ratio are standardized and weighted to obtain the energy efficiency evaluation value. The color image is converted to grayscale, Gaussian filtering is used to remove noise, and low and high thresholds are adjusted. To adapt to crack and deformation detection, an expansion operation is performed to fill edge gaps and connect broken cracks. cv2.findContours is used to extract edge contours, and cracks and deformations are filtered out based on features such as contour length and width ratio. Detected cracks and deformations are marked on the original image. The area ratio is obtained by the ratio of the total area of ​​crack and deformation regions to the total image area, and the density is obtained by the ratio of the number of edge pixels of cracks and deformations to the total number of pixels in the image. The device wear level is obtained by weighted sum of the area ratio and density. The device aging degree is quantified by analyzing the color changes in the RGB channels and using a color decay factor. Device aging usually leads to differences in the RGB channels between the image and the reference image. The specific calculation of device aging degree is as follows: ,in, , and These represent the attenuation rates for the red, green, and blue channels, respectively. , and These are the corresponding weight coefficients, each with a value of 0.33. The degree of equipment contamination is quantified by calculating the image contrast. Contamination will reduce the image contrast. The degree of equipment contamination is obtained by subtracting the ratio of the current image contrast to the reference image contrast from 1. The damage assessment value is obtained by standardizing the degree of equipment wear, the degree of equipment aging, and the degree of equipment contamination, and then adding them together with weights.

[0037] In one embodiment of the present invention, the comprehensive impact assessment value for lighting safety includes: The comprehensive impact assessment value for lighting safety is as follows:

[0038] in, The comprehensive impact assessment value for lighting safety. , and These are the standardized illumination assessment value, energy consumption assessment value, and damage assessment value, respectively. , and These are the corresponding weighting coefficients.

[0039] Specifically, weights are assigned based on the importance of factors affecting lighting safety, including illumination assessment values, energy consumption assessment values, and damage assessment values. If the illumination assessment value has the greatest impact on safety, it is given a higher weight. , and These are the corresponding weighting coefficients, with values ​​of 0.4, 0.3, and 0.3 respectively.

[0040] In one embodiment of the present invention, the smart lighting digital twin dynamic operation and maintenance decision module includes: Historical data is collected, and annotation rules are formulated, including maintenance time and energy consumption optimization coefficients. After labeling the maintenance time and energy consumption optimization coefficients, the data is cleaned and augmented, and then divided into training set, validation set and test set. Based on the collected data, the comprehensive impact assessment values ​​of the lighting environment, operation status, and lighting safety are calculated and used as input features for the digital twin model. A digital twin model is constructed, the training set is input into the digital twin model for training, and the model parameters are optimized through the validation set. The digital twin model outputs maintenance time and energy consumption optimization coefficients as operation and maintenance decision results. The trained model is deployed to the system, the evaluation value is updated based on real-time data, and operation and maintenance decisions are dynamically generated and sent to physical devices. Collect feedback data after physical equipment executes operation and maintenance decisions, and optimize the digital twin model based on the feedback data. Specifically, data collected from various devices and operation and maintenance records are acquired. The annotation content includes maintenance time and energy consumption optimization coefficient. The time dimension records the specific time of each maintenance session, and the energy consumption optimization coefficient is scored based on the energy consumption reduction ratio, with a value between 0 and 1. Experts or operation and maintenance personnel annotate the data according to historical records and annotation rules, dividing the processed data into training, validation, and validation sets. A digital twin model is constructed using neural networks or ensemble models. Input features include the comprehensive impact assessment value of the lighting environment, the comprehensive impact assessment value of the operating status, and the comprehensive impact assessment value of lighting safety. The output is maintenance time and energy consumption optimization coefficient. The energy consumption optimization coefficient uses the energy consumption value recorded for a period before optimization (e.g., one month) as a baseline value, and the monthly energy consumption after optimization as the energy consumption value. The energy consumption reduction ratio is obtained by dividing the difference between the baseline value and the energy consumption value by the baseline value. This ratio is mapped to a value between 0 and 1. The difference between the energy consumption reduction ratio and 1 minus the proportional cost is weighted and added together to obtain the energy efficiency optimization coefficient. The higher the energy efficiency optimization coefficient, the more significant the optimization effect.

[0041] Please see Figure 2 As shown, this invention is a smart lighting dynamic operation and maintenance management solution based on digital twins, including the following steps: S1: Collect environmental data from various sensors distributed on urban lighting facilities and operational status data of the lighting facilities. At the same time, use cameras to collect images of the lighting facilities to obtain image safety data of the lighting facilities. S2: Preprocess the multi-source data separately and transmit the data to the cloud via encrypted communication; S3: Establish a database in the cloud to categorize and store multi-source data; S4: Analyze environmental data and calculate the comprehensive impact assessment value of the lighting environment; analyze operational status data and calculate the comprehensive impact assessment value of the operational status; analyze image safety data and calculate the comprehensive impact assessment value of lighting safety. S5: Based on data calculation, the comprehensive impact evaluation value of lighting environment, operation status and safety is used as the input features of the digital twin model. The digital twin model is constructed, trained using the training set and optimized through the validation set. The model outputs operation and maintenance decision results.

[0042] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A smart lighting dynamic operation and maintenance management system based on digital twins, characterized in that, Includes the following modules: Data acquisition module: Collects environmental data from various sensors distributed on urban lighting facilities, as well as the operating status data of the lighting facilities. At the same time, it uses cameras to capture images of the lighting facilities to obtain image safety data of the lighting facilities. Data processing module: preprocesses multi-source data and transmits the data to the cloud via encrypted communication; Data storage module: Establishes a database in the cloud to store multi-source data in a categorized manner; Data Analysis Module: Analyzes environmental data and calculates the comprehensive impact assessment value of the lighting environment; analyzes operational status data and calculates the comprehensive impact assessment value of the operational status; analyzes image safety data and calculates the comprehensive impact assessment value of lighting safety. The smart lighting digital twin dynamic operation and maintenance decision module calculates the comprehensive impact evaluation value of lighting environment, operation status and safety based on data, which serves as the input features of the digital twin model. The digital twin model is constructed, trained using the training set and optimized through the validation set, and the model outputs operation and maintenance decision results.

2. The intelligent lighting dynamic operation and maintenance management system based on digital twins according to claim 1, characterized in that, The data acquisition module includes: Multiple environmental sensors are installed on urban lighting facilities to collect data on temperature, humidity, light intensity, air quality, and precipitation; multiple operating status sensors are installed on the control units of the lighting facilities to collect operating status data, including current, voltage, power, and switch status data. At the same time, cameras are installed on the lighting facilities to collect camera data, including security data of the lighting facilities and their surroundings; A data acquisition terminal is deployed in the power supply box of the lighting facility to receive and process data from sensors and cameras. The communication module supports wireless or wired data transmission and sets the data acquisition frequency.

3. The intelligent lighting dynamic operation and maintenance management system based on digital twins according to claim 1, characterized in that, The data processing module includes: The data acquisition terminal reads environmental and operational status data daily and performs preprocessing, including filtering, calibration, and formatting. The system collects image data of the lighting facilities and their surroundings using cameras, and then compresses, labels, and timestamps the images. The collected environmental data, operational status data, and image security data are transmitted to the cloud via a communication module, and encryption technology is used to encrypt them during the transmission process.

4. The intelligent lighting dynamic operation and maintenance management system based on digital twins according to claim 1, characterized in that, The data storage module is used to establish a database in the cloud, classify and store environmental data, operational status data and image security data, index and back up the stored data, and perform desensitization processing on the image security data.

5. The intelligent lighting dynamic operation and maintenance management system based on digital twins according to claim 1, characterized in that, The analysis of environmental data and the calculation of the comprehensive impact assessment value of the lighting environment include: Multiple parameter values ​​in the environmental data are mapped to a standardized range. Specifically, the precipitation parameter value is assigned a value of 1 if there is precipitation on that day, and 0 otherwise. The standardized parameter values ​​are combined with their corresponding weights to obtain the comprehensive impact assessment value of the lighting environment, specifically: in, The comprehensive impact assessment value of the lighting environment. , , , and The standardized values ​​for temperature, humidity, light intensity, air quality, and precipitation. , , , and These are the corresponding weighting coefficients.

6. The intelligent lighting dynamic operation and maintenance management system based on digital twins according to claim 1, characterized in that, The analysis of operational status data and the calculation of the comprehensive impact evaluation value of operational status include: Based on historical data, normal ranges were set for the parameter values ​​in the operational status data, and these parameter values ​​were mapped to standardized ranges. Weights were assigned to each parameter, and the weighted summation yielded the comprehensive operational status impact evaluation value. Specifically: in, The comprehensive impact evaluation value of the operating status. , , and These are the standardized current value, voltage value, power value, and switch status value, respectively. , , and The corresponding weighting coefficients; The switch status value is obtained by dividing the actual number of switches by the maximum allowed number of switches for the lighting equipment.

7. The intelligent lighting dynamic operation and maintenance management system based on digital twins according to claim 1, characterized in that, The analysis of image security data includes: The acquired image data is preprocessed, including noise reduction, contrast enhancement, and region segmentation. The region segmentation is based on the light intensity to segment the sky, buildings, roads, and lighting equipment areas. Features are extracted from the image based on three dimensions, and the corresponding illumination assessment value, energy consumption assessment value, and damage assessment value are calculated respectively. The lighting evaluation value is obtained by calculating the mean image brightness, standard deviation of brightness, and color temperature deviation, and then summing them after standardization and weighting; wherein, the color temperature deviation is obtained by dividing the actual color temperature by the target color temperature; The energy consumption assessment value is obtained by combining image analysis with lighting equipment tags to obtain the actual power, calculating the energy efficiency ratio, and then adding them together after standardization and weighting. The damage assessment value is obtained by standardizing and weighting the degree of equipment wear, equipment aging, and equipment contamination by identifying surface cracks and deformations of the lighting equipment using the Canny algorithm, color decay analysis, texture analysis, and standardizing the degree of equipment wear, equipment aging, and equipment contamination. The comprehensive impact assessment value of lighting safety is obtained by weighting and adding the illumination assessment value, energy consumption assessment value, and damage assessment value.

8. A smart lighting dynamic operation and maintenance management system based on digital twins according to claim 7, characterized in that, The comprehensive impact assessment value for lighting safety includes: The comprehensive impact assessment value for lighting safety is as follows: in, The comprehensive impact assessment value for lighting safety. , and These are the standardized illumination assessment value, energy consumption assessment value, and damage assessment value, respectively. , and These are the corresponding weighting coefficients.

9. A smart lighting dynamic operation and maintenance management system based on digital twins according to claim 1, characterized in that, The smart lighting digital twin dynamic operation and maintenance decision-making module includes: Historical data is collected, and annotation rules are formulated, including maintenance time and energy consumption optimization coefficients. After labeling the maintenance time and energy consumption optimization coefficients, the data is cleaned and augmented, and then divided into training set, validation set and test set. Based on the collected data, the comprehensive impact assessment values ​​of the lighting environment, operation status, and lighting safety are calculated and used as input features for the digital twin model. A digital twin model is constructed, the training set is input into the digital twin model for training, and the model parameters are optimized through the validation set. The digital twin model outputs maintenance time and energy consumption optimization coefficients as operation and maintenance decision results. The trained model is deployed to the system, the evaluation value is updated based on real-time data, and operation and maintenance decisions are dynamically generated and sent to physical devices. Collect feedback data after physical equipment executes operation and maintenance decisions, and optimize the digital twin model based on the feedback data.

10. A method for dynamic operation and maintenance management of smart lighting based on digital twins, characterized in that, Includes the following steps: S1: Collect environmental data from various sensors distributed on urban lighting facilities and operational status data of the lighting facilities. At the same time, use cameras to collect images of the lighting facilities to obtain image safety data of the lighting facilities. S2: Preprocess the multi-source data separately and transmit the data to the cloud via encrypted communication; S3: Establish a database in the cloud to categorize and store multi-source data; S4: Analyze environmental data and calculate the comprehensive impact assessment value of the lighting environment; analyze operational status data and calculate the comprehensive impact assessment value of the operational status; analyze image safety data and calculate the comprehensive impact assessment value of lighting safety. S5: Based on data calculation, the comprehensive impact evaluation value of lighting environment, operation status and safety is used as the input features of the digital twin model. The digital twin model is constructed, trained using the training set and optimized through the validation set. The model outputs operation and maintenance decision results.

Images