A Smart Streetlight Intelligent Dimming Control System and Method Based on 5G Communication and Extreme Weather Prediction

Abstract

This invention discloses a smart street light intelligent dimming control system and method based on 5G communication and extreme weather prediction, relating to the fields of smart city and road lighting control technology. It collects local micro-meteorological data in real time and receives macro-meteorological forecast data via a 5G communication module; it aligns and fuses the two data in time to construct a fused feature vector containing macro-micro deviations and time change rates; it uses a time-series prediction model to predict the type, probability, and intensity level of local weather changes within the next K minutes; it dynamically generates lighting dimming commands based on the prediction results and sends them to the individual lamp controller via the 5G module for millisecond-level dimming; simultaneously, it transmits execution feedback and measured weather data back to the prediction model for closed-loop optimization. This invention achieves proactive lighting control before severe weather arrives, significantly improving road driving safety while also saving energy and reducing consumption, and possesses self-learning and multi-street light coordination capabilities.

Description

Technical Field

[0001] This invention relates to the field of smart city and road lighting control technology, and in particular to a smart street light intelligent dimming control system and method based on 5G communication and extreme weather prediction. Background Technology

[0002] Street lighting is a crucial component of urban infrastructure, consuming a significant proportion of urban public electricity. Traditional street lighting control methods primarily rely on fixed schedules (timed switching, segmented dimming) or simple ambient light sensors, which present the following problems:

[0003] Delayed response: Traditional methods cannot predict sudden weather changes (such as fog, rainstorms, strong winds, etc.) in advance. Lighting is often adjusted passively only after severe weather has occurred, which poses a safety hazard.

[0004] Lack of local perception: Urban weather forecasts have a large grid scale (kilometer level), which cannot reflect microclimate changes at the specific location of streetlights, such as localized fog or gusts of wind.

[0005] The strategy is too simplistic: most existing smart streetlights adjust their brightness based solely on ambient light intensity, without taking into account the combined effects of multiple meteorological factors such as visibility, rainfall intensity, wind speed, and icing risk.

[0006] Therefore, there is an urgent need for a control method that can predict sudden changes in local weather and dynamically optimize lighting strategies. To this end, a smart street light intelligent dimming control system and method based on 5G communication and extreme weather prediction is proposed. Summary of the Invention

[0007] The main objective of this invention is to provide a smart street light intelligent dimming control system and method based on 5G communication and extreme weather prediction. By integrating macro weather forecasts and local micro weather sensor data of street lights, a time-series prediction model is used to predict sudden changes in local weather, and refined lighting control commands are dynamically generated accordingly to improve the safety and optimize energy saving of road lighting.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] In a first aspect, the present invention provides a smart street light intelligent dimming control method based on 5G communication and extreme weather prediction, comprising the following:

[0010] Step S1: Real-time collection of local micro-meteorological data is achieved through local meteorological sensors deployed on streetlight poles. The local micro-meteorological data includes temperature, humidity, air pressure, rainfall, visibility, and wind speed. Macro-meteorological forecast data and early warning information provided by the city meteorological bureau or third-party commercial meteorological services are received through a 5G communication module. The local meteorological sensors include at least: temperature and humidity sensors, air pressure sensors, rainfall sensors, visibility sensors, and wind speed sensors.

[0011] Step S2: Align the local micro-meteorological data with the macro-meteorological forecast data along a unified time axis, calculate the macro-micro deviation characteristics and the time change rate characteristics of local meteorological elements, and form a fusion feature vector. The fusion feature vector includes at least the meteorological parameter vector collected by local sensors, the macro-forecast meteorological parameter vector, the macro-micro deviation characteristics, and the time change rate characteristics of local meteorological elements.

[0012] Step S3: Input the fused feature vector into the pre-trained time series prediction model to predict the type, probability, and intensity level of weather changes that will occur in the local area where the street light is located at time K in the future;

[0013] Step S4: Based on the predicted weather change, and combined with the current basic lighting strategy, generate a dynamic dimming command. The dynamic dimming command includes a target dimming value, on / off status, and transition time parameters.

[0014] Step S5: Send the dynamic dimming command to the controller of the single lamp through the 5G communication module and drive the LED lamp to execute it.

[0015] Preferably, the fused feature vector Represented as: , A vector of meteorological parameters collected by local sensors. For macroscopic forecast meteorological parameter vectors, This is a macro-micro deviation characteristic. This represents the temporal variation rate characteristics of local meteorological elements.

[0016] Preferably, the time-series prediction model uses a long short-term memory network or a gated recurrent unit, takes the fused feature vector of the past P time steps as input, and outputs the probability distribution of various weather emergencies in the next K time steps.

[0017] Preferably, the sudden weather change includes at least one or more of the following: a sudden drop in visibility, a sharp increase in rainfall intensity, a sudden increase in wind speed, and a sudden drop in temperature leading to a risk of road icing.

[0018] The sudden drop in visibility event is determined according to any of the following criteria:

[0019] Criterion 1: The predicted rate of decrease in visibility exceeds a set threshold;

[0020] Criterion 2: The absolute value of the predicted visibility is lower than the lower limit of safe visibility.

[0021] Preferably, the method for generating the dynamic dimming command includes:

[0022] Determine the reference dimming value ;

[0023] Calculate the comprehensive weather impact coefficient based on the weather change prediction results. : ,in Let j be the illumination demand correction function corresponding to the j-th type of weather change. These are the corresponding weighting coefficients;

[0024] The final dimming value is calculated based on the comprehensive weather impact coefficient. : ,in The maximum permissible dimming value, This is the energy-saving mode coefficient.

[0025] Preferably, the illuminance demand correction function includes:

[0026] Visibility correction function : , To predict visibility, This is a reference value for normal visibility.

[0027] Rainfall intensity correction function : , To predict rainfall intensity, for;

[0028] Wind speed correction function : , To predict wind speed, For safe wind speed;

[0029] Icing risk correction factor : , for, for, for, An additional factor is added to account for the risk of icing.

[0030] Preferably, during the generation of the dynamic dimming command, the dimming command actually sent to the individual lamp controller... First-order hysteresis filtering is used for processing: ,in For the gradual velocity coefficient, To control the cycle and avoid sudden flickering of lighting parameters.

[0031] Preferably, the status information of the lighting fixtures after execution and the actual weather data after execution are collected and fed back to the time series prediction model for model updates or retraining, specifically as follows:

[0032] The predicted weather change events are compared with the actual weather change events to calculate the prediction accuracy, false negative rate, and false positive rate. The prediction-actual sample pairs are added to the training dataset, and the time series prediction model is incrementally trained or learned online at fixed intervals.

[0033] Preferably, when the local weather change prediction result of a certain street light reaches the collaborative triggering threshold, the warning information is broadcast to the surrounding adjacent street lights through the 5G communication module, and the adjacent street lights adjust their lighting strategies in advance based on the received warning information.

[0034] Secondly, the present invention provides a smart street light intelligent dimming control system based on 5G communication and extreme weather prediction, used to implement the above method, including:

[0035] The sensing layer device includes:

[0036] Local meteorological sensor arrays deployed on streetlight poles are used to collect local micro-meteorological data in real time;

[0037] A data receiving interface that communicates with city meteorological bureaus or third-party commercial meteorological services to obtain macro-meteorological forecast data and early warning information;

[0038] The network layer device includes a 5G communication module integrated on the street light pole, used to realize low-latency cloud command delivery and status reporting, as well as collaborative information broadcasting between street lights;

[0039] Platform layer devices, including cloud servers or edge computing gateways, are deployed with:

[0040] The data fusion module is used to perform time alignment, anomaly removal, and feature extraction on the local micro-meteorological data and macro-meteorological forecast data to generate a fused feature vector.

[0041] The time-series prediction module is used to predict the type, probability, and intensity level of weather changes that will occur in the local area where the streetlight is located at time K in the future, based on the fused feature vector.

[0042] The lighting strategy generation module is used to generate dynamic dimming instructions based on the predicted results of the sudden weather change. The dynamic dimming instructions include target dimming value, on / off state and gradient time parameters.

[0043] The closed-loop feedback module is used to receive the status information of the lamps after execution and the actual weather data, and to update or retrain the time series prediction module.

[0044] Execution layer device, the execution layer device comprising:

[0045] The single-lamp controller receives the dynamic dimming command via a 5G communication module;

[0046] The intelligent LED driver module is electrically connected to the single lamp controller and performs dimming or switching actions according to instructions;

[0047] The perception layer devices, network layer devices, platform layer devices, and execution layer devices are connected in sequence to form a closed-loop control system.

[0048] Beneficial effects

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] This invention uses a time-series prediction model to anticipate sudden changes in local weather (such as fog, heavy rain, strong winds, etc.) and proactively adjusts lighting strategies before the weather deteriorates, significantly improving road safety.

[0051] This invention utilizes large-scale trend information from urban weather forecasts and combines it with real-time data from local micro-meteorological sensors on streetlights, effectively solving the problem of insufficient accuracy of macro forecasts at the local scale.

[0052] This invention comprehensively considers various weather factors such as visibility, rainfall intensity, wind speed, and icing risk, avoiding the one-sidedness of decision-making based on a single factor, and making the lighting strategy more in line with the actual environmental needs.

[0053] This invention maintains an energy-saving dimming mode when there are no sudden weather changes, and dynamically increases the illuminance when severe weather is predicted, achieving a dynamic optimal balance between safety and energy saving.

[0054] This invention continuously iterates and optimizes the prediction model by comparing the prediction results with the actual weather, making the system more and more accurate with use.

[0055] This invention employs 5G low-latency communication and first-order filter gradual control to ensure fast response while avoiding visual discomfort caused by sudden changes in lighting. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to the present invention.

[0057] Figure 2 This is a control logic block diagram of the method of the present invention;

[0058] Figure 3This is a network structure diagram of the time series prediction model of the present invention;

[0059] Figure 4 The decision flow diagram generated for the lighting strategy of this invention;

[0060] Figure 5 This is a schematic diagram of the actual appearance of the smart street light of the present invention;

[0061] Figure 6 This is a schematic diagram of the internal structure of the smart street light of the present invention.

[0062] In the diagram: 1. Local weather sensor array and 5G communication module; 2. Single lamp controller; 3. Power supply system; 4. LED lighting fixtures. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0064] See Figures 1-6 This invention provides a smart street light intelligent dimming control method based on 5G communication and extreme weather prediction, comprising:

[0065] Step S1: Multi-source data acquisition

[0066] Local micro-meteorological data is collected in real time by local meteorological sensors deployed on streetlight poles. The local micro-meteorological data includes temperature, humidity, air pressure, rainfall, visibility, and wind speed. At the same time, macro-meteorological forecast data and early warning information provided by the city meteorological bureau or third-party commercial meteorological services are received through a 5G communication module.

[0067] Step S2: Data Fusion and Feature Extraction

[0068] Local micrometeorological data and macro-meteorological forecast data are aligned along a unified time axis, and outliers are removed to construct a fused feature vector. The fused feature vector includes: local sensor raw values, macro-forecast values, macro-micro deviation values, and the temporal rate of change of local meteorological elements.

[0069] Step S3: Prediction of sudden changes in local weather

[0070] The fused feature vectors are input into the time series prediction model to predict the type, probability, and intensity level of weather changes that will occur in the local area where the streetlights are located within the next K minutes. Weather changes include at least one or more of the following: sudden drop in visibility, sudden increase in rainfall intensity, sudden increase in wind speed, and sudden drop in temperature leading to road icing risk.

[0071] Step S4: Lighting control strategy generation

[0072] Based on the forecast of sudden weather changes and combined with the current basic lighting strategy, a dynamic dimming instruction is generated. The dynamic dimming instruction includes the target dimming value, on / off status, and transition time parameters.

[0073] Step S5: Issuance and Execution of Instructions

[0074] The 5G communication module sends dynamic dimming commands to the individual lamp controllers of the streetlights. The individual lamp controllers drive the intelligent drive modules of the LED lamps to perform millisecond-level dimming or switching actions according to the commands.

[0075] Step S6: Closed-loop feedback

[0076] Collect the status information (current, voltage, power, actual dimming value) of the lighting fixtures after execution, as well as the actual weather data that occurred after execution, and feed it back to the time series prediction model for model update or retraining.

[0077] Example 1: Single Lamp Independent Control Scenario

[0078] Scenario setting: A smart street light on a main road in a city is equipped with five types of miniature meteorological sensors, namely temperature and humidity, air pressure, rainfall, visibility and wind speed, and is connected to the city meteorological bureau's API through a 5G module.

[0079] Step 1, Data Collection

[0080] The streetlight's local sensor collects data every 10 seconds: current visibility 800 meters, wind speed 3 m / s, no rain. Simultaneously, a grid forecast from the meteorological bureau indicates visibility will remain above 500 meters for the next hour, with no warnings.

[0081] Step 2, Feature Extraction

[0082] Construct a feature window from the past 30 minutes, calculate the visibility decline rate and the macro-micro visibility deviation (macro forecast value is 50 meters lower than local value), and use these as model inputs.

[0083] Step 3, Mutation Prediction

[0084] LSTM model output: The probability of a sudden drop in visibility within the next 15 minutes is 85%, and the visibility is predicted to drop to 120 meters (fog formation).

[0085] Step 4, Strategy Generation

[0086] Base dimming value is 60% (energy-saving mode). Visibility correction function ϕ vis =1−120 / 500=0.76; Take K weather =0.8 × 0.76 = 0.61. Final dimming value L final=60%×(1+0.61)=96.6%, upper limit 100%, rounded to 97%.

[0087] Step 5, Instruction Issuance

[0088] Using 5G downlink commands: dimming value 97%, gradient time 30 seconds. The single-lamp controller smoothly increases the brightness from 60% to 97% within 30 seconds.

[0089] Step 6, Control Feedback

[0090] Fifteen minutes later, the local visibility sensor measured 115 meters, close to the prediction. The system recorded this prediction as correct and used it as a positive sample for subsequent model training.

[0091] Example 2: Multi-light coordinated control scenario

[0092] Scene setting: Streetlights A, B, and C are continuously installed along the road, with a spacing of 40 meters.

[0093] The local wind speed sensor of street light A suddenly increased (from 3 m / s to 12 m / s), and its LSTM model predicted that the wind speed would continue to rise to 18 m / s within the next 5 minutes, reaching the collaborative triggering threshold (>15 m / s).

[0094] Streetlight A broadcasts a "strong wind warning" message (including the forecast time window and intensity level) to adjacent streetlights B and C via a 5G module.

[0095] After receiving the warning, streetlights B and C reduced their dimming values ​​by 20% in advance (to reduce the impact of wind load on the light poles and auxiliary equipment) and entered safety monitoring mode.

[0096] After the strong winds pass, all streetlights will automatically return to normal lighting conditions.

[0097] Example 3: Online Model Update

[0098] The system performs the following operations daily at 2:00 AM (during low load):

[0099] Extract all “predicted-actual” sample pairs from the past 24 hours.

[0100] Calculate the daily prediction accuracy (sudden drop in visibility event: 85%).

[0101] Add the missed and false positive samples to the retraining dataset.

[0102] Perform incremental training on the LSTM model for 3 epochs and update the weight parameters.

[0103] Verify the performance of the updated model on the data from the most recent 6 hours. If the accuracy improves, deploy the new model; otherwise, roll back.

[0104] Example 4:

[0105] This invention also provides a smart street light intelligent dimming control system based on 5G communication and extreme weather prediction, used to implement the above method, including:

[0106] Sensing layer devices, including:

[0107] Local meteorological sensor arrays deployed on streetlight poles are used to collect local micro-meteorological data in real time;

[0108] A data receiving interface that communicates with city meteorological bureaus or third-party commercial meteorological services to obtain macro-meteorological forecast data and early warning information;

[0109] Network layer devices include 5G communication modules integrated on street light poles, used to enable low-latency cloud command delivery and status reporting, as well as collaborative information broadcasting between street lights;

[0110] Platform layer devices, including cloud servers or edge computing gateways, are deployed with:

[0111] The data fusion module is used to perform time alignment, anomaly removal, and feature extraction on local micro-meteorological data and macro-meteorological forecast data to generate a fused feature vector.

[0112] The time-series prediction module is used to predict the type, probability, and intensity level of weather changes that will occur in the local area where the streetlight is located at time K in the future, based on the fused feature vector.

[0113] The lighting strategy generation module is used to generate dynamic dimming instructions based on the forecast of sudden weather changes. The dynamic dimming instructions include the target dimming value, on / off status and gradient time parameters.

[0114] The closed-loop feedback module is used to receive the status information of the lamps after execution and the actual weather data, and to update or retrain the time series prediction module.

[0115] Execution layer devices, including:

[0116] The single-lamp controller receives dynamic dimming commands via a 5G communication module;

[0117] The intelligent LED driver module is electrically connected to the single lamp controller and performs dimming or switching actions according to instructions.

[0118] The perception layer devices, network layer devices, platform layer devices, and execution layer devices are connected in sequence to form a closed-loop control system.

[0119] Industrial applicability

[0120] The intelligent street lighting strategy control method proposed in this invention can be deployed in newly built intelligent streetlights or for upgrading existing LED streetlights (by adding weather sensors and 5G communication modules). Its hardware and software architecture is clear, and the prediction model can run on a cloud platform or edge computing gateway. It is suitable for road sections prone to sudden local weather changes, such as urban expressways, main roads, bridges, tunnel entrances and exits, and mountain roads, and has good market promotion value.

[0121] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A smart street light intelligent dimming control method based on 5G communication and extreme weather prediction, characterized in that, Includes the following steps: S1: Real-time collection of local micro-meteorological data through local meteorological sensors deployed on streetlight poles, and reception of macro-meteorological forecast data and early warning information provided by the city meteorological bureau or third-party commercial meteorological services through a 5G communication module; S2: Align the local micro-meteorological data with the macro-meteorological forecast data along a unified time axis, calculate the macro-micro deviation characteristics and the temporal change rate characteristics of local meteorological elements, and form a fused feature vector; S3: Input the fused feature vector into the pre-trained time series prediction model to predict the type, probability, and intensity level of weather changes that will occur in the local area where the street light is located at time K in the future; S4: Based on the predicted results of the sudden weather change and combined with the basic lighting strategy for the current period, generate a dynamic dimming instruction, which includes a target dimming value, on / off status and transition time parameters. S5: The dynamic dimming command is sent to the controller of the single lamp via the 5G communication module, and the LED lamp is driven to execute.

2. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 1, characterized in that, The fused feature vector At least include meteorological parameter vectors collected by local sensors Macro-forecast meteorological parameter vector Macro-micro deviation characteristics and temporal variation characteristics of local meteorological elements , is represented as: .

3. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 1, characterized in that, The time-series prediction model uses a long short-term memory network or a gated recurrent unit, taking the fused feature vectors of the past P time steps as input, and outputting the probability distribution of various weather events in the next K time steps.

4. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 1, characterized in that, The sudden weather changes include at least one or more of the following: a sudden drop in visibility, a sharp increase in rainfall intensity, a sudden increase in wind speed, and a sudden drop in temperature leading to the risk of road icing. The sudden drop in visibility event is determined according to any of the following criteria: Criterion 1: The predicted rate of decrease in visibility exceeds a set threshold; Criterion 2: The absolute value of the predicted visibility is lower than the lower limit of safe visibility.

5. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 1, characterized in that, The method for generating the dynamic dimming command includes: Determine the reference dimming value ; Calculate the comprehensive weather impact coefficient based on the weather change prediction results. : ,in Let j be the illumination demand correction function corresponding to the j-th type of weather change. These are the corresponding weighting coefficients; The final dimming value is calculated based on the comprehensive weather impact coefficient. : ,in The maximum permissible dimming value, This is the energy-saving mode coefficient.

6. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 5, characterized in that, The illuminance demand correction function includes: Visibility correction function : , To predict visibility, This is a reference value for normal visibility. Rainfall intensity correction function : , To predict rainfall intensity, for; Wind speed correction function : , To predict wind speed, For safe wind speed; Icing risk correction factor : , for, for, for, An additional factor is added to account for the risk of icing.

7. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 1, characterized in that, During the generation of the dynamic dimming command, the dimming command actually sent to the individual lamp controller... First-order hysteresis filtering is used for processing: ,in For the gradual velocity coefficient, To control the cycle and avoid sudden flickering of lighting parameters.

8. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 1, characterized in that, The local meteorological sensors include at least: a temperature and humidity sensor, a barometric pressure sensor, a rainfall sensor, a visibility sensor, and a wind speed sensor.

9. The intelligent dimming control method for smart streetlights based on 5G communication and extreme weather prediction according to claim 1, characterized in that, Also includes: S6: Collect the status information of the lighting fixtures after execution and the actual weather data that occurred after execution, and feed it back to the time series prediction model for model updates or retraining. Specifically: The predicted weather change events are compared with the actual weather change events to calculate the prediction accuracy, false negative rate and false positive rate. The prediction-actual sample pairs are added to the training dataset, and the time series prediction model is incrementally trained or learned online at fixed intervals. S7: When the local weather change prediction result of a certain street light reaches the collaborative triggering threshold, the warning information is broadcast to the surrounding adjacent street lights through the 5G communication module. The adjacent street lights will adjust their lighting strategies in advance based on the received warning information.

10. A smart street light intelligent dimming control system based on 5G communication and extreme weather prediction, used to implement the method described in any one of claims 1-9, comprising: The sensing layer device includes: Local meteorological sensor arrays deployed on streetlight poles are used to collect local micro-meteorological data in real time; A data receiving interface that communicates with city meteorological bureaus or third-party commercial meteorological services to obtain macro-meteorological forecast data and early warning information; The network layer device includes a 5G communication module integrated on the street light pole, used to realize low-latency cloud command delivery and status reporting, as well as collaborative information broadcasting between street lights; Platform layer devices, including cloud servers or edge computing gateways, are deployed with: The data fusion module is used to perform time alignment, anomaly removal, and feature extraction on the local micro-meteorological data and macro-meteorological forecast data to generate a fused feature vector. The time-series prediction module is used to predict the type, probability, and intensity level of weather changes that will occur in the local area where the streetlight is located at time K in the future, based on the fused feature vector. The lighting strategy generation module is used to generate dynamic dimming instructions based on the predicted results of the sudden weather change. The dynamic dimming instructions include target dimming value, on / off state and gradient time parameters. The closed-loop feedback module is used to receive the status information of the lamps after execution and the actual weather data, and to update or retrain the time series prediction module. Execution layer device, the execution layer device comprising: The single-lamp controller receives the dynamic dimming command via a 5G communication module; The intelligent LED driver module is electrically connected to the single lamp controller and performs dimming or switching actions according to instructions; The perception layer devices, network layer devices, platform layer devices, and execution layer devices are connected in sequence to form a closed-loop control system.

Images