A method and system for controlling urban smart street lights

Abstract

This invention discloses a method and system for controlling smart urban streetlights. The method includes collecting road environment parameters (PM2.5 concentration, humidity, ambient illumination, SO2 concentration) and traffic status parameters (vehicle volume, pedestrian volume, average lane speed, and lane occupancy); constructing brightness and color temperature adjustment models, weighting them using a combination of entropy weighting and analytic hierarchy process (AHP), introducing a dynamic weight correction factor, and integrating real-time wind speed data to compensate for SO2 concentration diffusion, generating brightness and color temperature judgment values; and, based on the judgment values, assigning brightness and color temperature adjustment levels to dynamically adjust streetlight brightness and color temperature. This method achieves refined regulation of streetlight brightness and color temperature, improving the scientific nature, accuracy, real-time nature, and adaptability of the regulation, ensuring that streetlight lighting meets road lighting requirements and can adapt to varying traffic and weather conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of road traffic lighting, and in particular to a method and system for controlling urban smart street lamps. Background Art

[0002] In modern urban traffic management, streetlights, as a crucial component of urban infrastructure, not only provide illumination for nighttime vehicles and pedestrians but also play a vital role in improving road safety, energy conservation and emission reduction, and shaping the city's image. With the acceleration of urbanization, urban traffic pressure is increasing. Traditional streetlight control systems are no longer able to meet the growing demand for intelligent and refined management. Traditional streetlight control systems often rely on timed switches or simple photosensitive controls, lacking flexibility and intelligence. They are unable to dynamically adjust lighting intensity and duration based on actual traffic flow and environmental changes, resulting in energy waste and inefficient lighting.

[0003] In recent years, with the development and widespread adoption of IoT technology, the concept of smart streetlights has emerged. By integrating sensors, wireless communication modules, and intelligent control algorithms, smart streetlights enable remote monitoring and intelligent management of streetlights. For example, by using vehicle detectors and pedestrian sensors installed on streetlights, the lighting system can collect real-time traffic flow data and dynamically adjust lighting intensity and on / off times based on this data, thereby achieving energy conservation and emission reduction. Furthermore, smart streetlights can be integrated with other urban management systems, such as traffic signal control systems and video surveillance systems, to enable information sharing and collaborative work, further enhancing the intelligent level of urban traffic management.

[0004] Patent publication number CN118317484B discloses a smart streetlight lighting control method and system based on IoT technology. For each streetlight in the smart streetlight system, the current lighting parameter configuration scheme is obtained, and the streetlight lighting data corresponding to this scheme is collected using IoT technology. Next, based on the collected streetlight lighting data, the energy consumption generated by the scheme is calculated. If the energy consumption exceeds a preset energy consumption threshold, the lighting parameters in the scheme are converted into a gene string structure. Using a preset objective function, the gene string is iteratively optimized to obtain the optimized lighting parameter configuration scheme.

[0005] However, existing smart streetlight systems still have some limitations. For example, most systems lack data processing and decision support capabilities, making them unable to implement lighting control strategies in complex scenarios. Furthermore, due to a lack of effective data analysis and predictive models, streetlight systems lack adaptability. Summary of the Invention

[0006] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide an urban smart street light control method and system, aiming to solve how to achieve fine adjustment of street light brightness and color temperature by comprehensively collecting road environment parameters and traffic status parameters to ensure that street light lighting meets road lighting needs.

[0007] The object of the present invention is achieved through the following technical solutions:

[0008] A method for controlling urban smart streetlights, comprising the following:

[0009] S1. Collect road environmental parameters and traffic status parameters, including PM2.5 concentration, humidity, ambient illumination, SO2 concentration, and traffic status parameters including traffic flow, pedestrian flow, average lane speed, and lane occupancy;

[0010] S2. Construct a brightness adjustment model and a color temperature adjustment model, calculating the weights of each parameter using a combined weighting algorithm. The brightness adjustment model uses vehicle flow, pedestrian flow, PM2.5 concentration, humidity, and ambient illumination as input parameters, and uses a combined entropy weighting method and the analytic hierarchy process to assign weights. A dynamic weight correction factor is introduced to generate a brightness judgment value. The color temperature adjustment model uses PM2.5 concentration, SO2 concentration, and humidity as input parameters, and uses a combined entropy weighting method and the analytic hierarchy process to assign weights. The model also integrates real-time wind speed data to perform diffusion compensation for SO2 concentration and generate a color temperature judgment value.

[0011] S3. Based on the brightness judgment value and the color temperature judgment value, the brightness adjustment level and the color temperature adjustment level are divided respectively, and the brightness and color temperature of the street light are dynamically adjusted.

[0012] As a preferred embodiment, it also includes:

[0013] S4. Achieve linkage between upstream and downstream streetlights through collaborative control strategies:

[0014] The upstream centralized controller calculates the adjustment time, hold time, and target brightness / color temperature of downstream streetlights based on the detected traffic volume, pedestrian volume, average lane speed, and lane occupancy, combined with the LSTM-based dynamic traffic flow prediction model.

[0015] By sending adjustment instructions to the downstream centralized controller based on the LoRaWAN and 5G dual-mode redundant communication protocol, the downstream street lights can complete pre-adjustment of brightness and color temperature before the arrival of traffic.

[0016] As a preferred method, the dynamic weight correction factor in the combined weighting algorithm is calculated as follows:

[0017] ;

[0018] in:

[0019] The objective weight calculated by the entropy weight method;

[0020] The absolute deviation between the current parameter value and the historical mean, where the historical mean is the mean of the same time period over the past seven days;

[0021] The maximum fluctuation range allowed for the parameter is set according to national standards;

[0022] The comprehensive weight after dynamic correction is: in, It is The comprehensive weight of each parameter indicates the final influence of the parameter in the brightness or color temperature adjustment model; is the weight obtained by the hierarchical analysis method, Indicates the parameters, The total number of parameters entered into the model, the specific value depends on the adjustment target. Brightness adjustment model: (traffic flow, pedestrian flow, PM2.5 concentration, humidity, ambient illumination); color temperature adjustment model: (PM2.5 concentration, SO2 concentration, humidity).

[0023] As a preferred method, the brightness adjustment level is divided into the following ways: the brightness judgment value interval [0,1] is divided into adaptive Gaussian intervals based on the standard deviation of historical traffic flow. The specific formula is: interval ;

[0024] in, For the The center value of each brightness level is 25%, 50%, 75%, and 100% basic brightness; is the interval standard deviation, which is dynamically adjusted based on the standard deviation of traffic volume in the past hour.

[0025] Brightness adjustment levels are dynamically adjusted based on historical traffic fluctuations. The system divides the brightness range [0, 1] into four levels (25%, 50%, 75%, and 100%), with the center value of each level corresponding to the base brightness. By dynamically calculating the standard deviation of traffic flow over the past hour, the range for each brightness level is calculated. If traffic fluctuates significantly (large standard deviation), the range is automatically widened to avoid frequent adjustments. If traffic is stable (small standard deviation), the range is narrowed for a more responsive brightness response. This adaptive mechanism intelligently balances energy consumption and lighting needs based on actual traffic flow, ensuring driving safety while reducing energy waste.

[0026] As a preferred embodiment, in the brightness adjustment model, the inverse correlation between ambient illumination and brightness is realized by a piecewise function:

[0027] ;

[0028] in, is the base brightness (unit: %); is the ambient illuminance (unit: lux), the coefficients 0.03 and 0.01 imply the dimensionless proportional coefficient / lux, or directly take Substitute the numerical value into the calculation, This is the minimum brightness threshold (default 20%) to avoid negative values or over-dimming.

[0029] As a preferred embodiment, the color temperature adjustment levels are divided as follows: the color temperature judgment value interval [0, 1] is divided into three sub-intervals, which respectively correspond to dynamic color temperature values based on the haze index:

[0030] ;

[0031] in, is the color temperature judgment value, and the calculation formula is:

[0032] ;

[0033] :PM2.5 normalized value, when PM2.5 concentration ≤300μg / m 3 hour, ; When PM2.5 concentration>300μg / m 3 hour, ;

[0034] : SO2 normalized value, (Based on GB 3095-2012 24-hour average limit);

[0035] : Humidity normalized value, ;

[0036] is the combined weight, satisfying , the calculation formula is:

[0037] ;

[0038] The entropy weight method is used to calculate the objective weights of parameters (based on parameter volatility); :The first The subjective weight of each parameter (expert experience).

[0039] The entropy weight method calculates the objective weight:

[0040] (brightness model);

[0041] (color temperature model);

[0042] is the objective weight of PM2.5 parameters in the brightness model and color temperature model, corresponding to (brightness model) and (Color temperature model).

[0043] ;

[0044] : The original PM2.5 objective weight calculated by the entropy weight method;

[0045] : the absolute deviation of the current PM2.5 concentration from the historical mean;

[0046] , the maximum fluctuation range allowed for PM2.5, such as .

[0047] When PM2.5 concentration deviates hour, The dynamic correction factor is amplified to reflect the real-time fluctuation of pollution concentration. The corrected weights need to be renormalized to ensure that the sum of all parameter weights is 1. , among which, only the weight of PM2.5 is modified, and the weights of other parameters remain the original values.

[0048] As a preferred method, the remote communication adopts LRaWAN and 5G dual-mode redundant transmission protocol, with a transmission delay of less than 2 seconds and a packet loss rate of less than 0.1%.

[0049] As a preferred embodiment, in the collaborative control strategy, the adjustment time of the downstream street lamp is determined by an improved fuzzy PID control algorithm, specifically:

[0050] Average lane speed (Unit: km / h) and lane occupancy (Unit: %) is input, and a new congestion index is added The membership function of : in, The maximum design speed for the road (unit: km / h);

[0051] Output regulation time The membership function of the time series (unit: seconds) adopts a dynamic asymmetric Gaussian distribution:

[0052] ;

[0053] in, is the expected value of the adjustment time calculated based on historical data, Standard deviation, adjusts the fluctuation range of time.

[0054] As a preferred method, in the improved fuzzy PID control algorithm, the proportional coefficient is dynamically adjusted The formula is:

[0055] ;

[0056] in, is the initial proportional coefficient; The vehicle speed predicted by the LSTM model (unit: km / h); is the actual detected vehicle speed (unit: km / h).

[0057] As a preferred embodiment, after receiving the adjustment instruction, if the actual traffic flow deviates from the predicted value by more than 20%, the downstream centralized controller re-predicts the target brightness / color temperature value through the Kalman filter algorithm: State equation:

[0058] ;

[0059] Observation equation:

[0060] ;

[0061] in:

[0062] is the state vector, including the traffic volume (unit: vehicles / minute) and the vehicle speed (unit: km / h);

[0063] is the system matrix;

[0064] are process noise and observation noise.

[0065] As a preferred method, it also includes an abnormal data processing step: when the PM2.5 concentration exceeds 300μg / m 3 When the fog penetration mode is activated, the setting is forced , directly trigger , color temperature is locked at 3000K;

[0066] The brightness is increased to 120% of the rated value until PM2.5 drops to 200 μg / m 3 The following is executed after verifying the adjustment effect through the digital twin platform.

[0067] An urban smart street light control system, comprising:

[0068] Data acquisition module, used to obtain road environment parameters and traffic status parameters in real time;

[0069] The dynamic weight calculation module is deployed in the centralized controller and includes:

[0070] The brightness weight unit integrates the exponential smoothing filter algorithm of PM2.5 concentration. The filtering formula is:

[0071] ;

[0072] in, is the historical average (unit: μg / m 3 ); is the current value, It is the PM2.5 concentration value after being processed by the exponential smoothing filter algorithm;

[0073] Color temperature weight unit, integrated The wind speed compensation model of concentration, the compensation formula is:

[0074] in, is the current wind speed;

[0075] An adjustment decision module is used to generate a brightness level instruction and a color temperature level instruction according to the brightness judgment value and the color temperature judgment value;

[0076] Collaborative control module, including:

[0077] The upstream controller is configured to run the LSTM-based traffic flow prediction model and output the downstream adjustment time, hold time, and target brightness / color temperature values;

[0078] The downstream controller is configured to calibrate the target brightness / color temperature value in real time through a Kalman filter algorithm and perform adjustment when the predicted time arrives.

[0079] The present invention has at least the following beneficial effects: By comprehensively collecting road environment and traffic status parameters, it achieves refined adjustment of streetlight brightness and color temperature. By constructing brightness and color temperature adjustment models and using a combined weighting algorithm to calculate the weights of each parameter, the scientific and accurate basis for adjustment is ensured. Furthermore, the introduction of a dynamic weight correction factor and real-time wind speed data to compensate for the diffusion of SO2 concentration further improves the real-time and adaptability of the adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] To reveal the technical details of the embodiments of the present invention, the following is a brief introduction to the drawings involved in the embodiments. It should be emphasized that these drawings only illustrate several embodiments of the present invention and should not be considered as defining the scope of the invention. Those skilled in the art can deduce other relevant drawings based on these drawings without engaging in creative work.

[0081] Figure 1 This is a flow chart of Example 1;

[0082] Figure 2 This is a structural diagram of an urban smart street light control system. DETAILED DESCRIPTION

[0083] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0084] In the following, embodiments of the present disclosure are described in detail with the aid of accompanying drawings. However, please be aware that the present disclosure is not limited to the specific forms shown herein. Rather, it should be understood to encompass various variations, equivalents, and / or alternatives to the embodiments of the present disclosure. In describing the drawings, the same reference numerals will be used to indicate similar components.

[0085] It should be understood that while the following description provides extensive specific details intended to facilitate a comprehensive understanding of the example embodiments, those skilled in the art will appreciate that the example embodiments can be implemented without these specific details. For example, systems may be presented in block diagram form to avoid excessive detail that would obscure the clarity of the examples. In other cases, unnecessary details regarding well-known processes, structures, and techniques may be omitted to maintain clarity of the examples.

[0086] like Figure 1 As shown, a method for controlling urban smart street lights includes the following contents:

[0087] S1. Collect road environmental parameters and traffic status parameters, including PM2.5 concentration, humidity, ambient illumination, SO2 concentration, and traffic status parameters including traffic flow, pedestrian flow, average lane speed, and lane occupancy;

[0088] S2. Construct a brightness adjustment model and a color temperature adjustment model, calculating the weights of each parameter using a combined weighting algorithm. The brightness adjustment model uses vehicle flow, pedestrian flow, PM2.5 concentration, humidity, and ambient illumination as input parameters, and uses a combined entropy weighting method and the analytic hierarchy process to assign weights. A dynamic weight correction factor is introduced to generate a brightness judgment value. The color temperature adjustment model uses PM2.5 concentration, SO2 concentration, and humidity as input parameters, and uses a combined entropy weighting method and the analytic hierarchy process to assign weights. The model also integrates real-time wind speed data to perform diffusion compensation for SO2 concentration and generate a color temperature judgment value.

[0089] S3. Based on the brightness judgment value and the color temperature judgment value, the brightness adjustment level and the color temperature adjustment level are divided respectively, and the brightness and color temperature of the street light are dynamically adjusted.

[0090] This embodiment dynamically adjusts streetlight brightness and color temperature by monitoring environmental and traffic data in real time, balancing road safety and energy conservation. For data collection and parameter integration, the system uses sensors to collect two types of data in real time: environmental parameters, including PM2.5 concentration (indicating haze), humidity (affecting visibility), ambient illumination (natural light intensity), SO2 concentration (a pollution indicator), and wind speed (affecting pollutant dispersion); and traffic parameters, including vehicle and pedestrian flow (determining lighting needs), average lane speed, and lane occupancy (determining congestion levels).

[0091] The brightness adjustment model integrates five parameters: traffic flow, pedestrian flow, PM2.5, humidity, and ambient illumination. It assigns weights using a combination of the "entropy weight method" and the analytic hierarchy process (AHP) method (taking into account both data objectivity and expert experience), and introduces a dynamic correction mechanism. For example, when PM2.5 concentration suddenly increases, the system automatically increases its weight, prioritizing fog permeability. Brightness adjustment: The brightness judgment value is divided into four levels (25%, 50%, 75%, and 100%). For example, during the evening rush hour, the brightness is adjusted to 100%, and when traffic is light in the early morning, the brightness is reduced to 25%, achieving "bright lights when there are many cars, dim lights when there are no cars." This part is described in detail in the subsequent examples.

[0092] The color temperature adjustment model is based on three parameters: PM2.5, SO2, and humidity, and incorporates wind speed data to compensate for SO2 concentration. (Higher wind speeds increase pollutant diffusion, reducing the weight of SO2.) For example, in windy weather, even with high SO2 concentrations, the system maintains a higher color temperature to optimize lighting efficiency. Color temperature adjustment: The system divides color temperature into three levels: 3000K warm yellow light, 4000K neutral light, and 5000K cool white light. On foggy days, the system automatically switches to 3000K for enhanced fog penetration, while on clear nights, the system uses 5000K high-efficiency cool light to save energy.

[0093] When the environment suddenly changes (such as a sudden increase in PM2.5) or traffic fluctuates, the model rapidly adjusts weights to avoid regulation lag. Tiered regulation balances energy consumption and demand, ensuring safe fog penetration during haze days and reducing ineffective lighting during low-traffic periods. Wind speed data corrects for the effects of SO2, preventing excessive color temperature reduction on windy days and improving regulation efficiency. This invention uses multi-parameter collaborative decision-making to enable streetlights to perceive the environment and anticipate demand, enabling intelligent and refined control in complex scenarios. This will be described in detail in the following examples.

[0094] In a preferred embodiment, the invention further comprises: S4. Linking upstream and downstream streetlights through a collaborative control strategy:

[0095] The upstream centralized controller calculates the adjustment time, hold time, and target brightness / color temperature of downstream streetlights based on detected traffic flow, pedestrian flow, average lane speed, and lane occupancy, combined with a dynamic traffic flow prediction model based on an LSTM. LSTM is a recurrent neural network specifically designed to handle long-term dependencies in time series data. Compared to conventional RNNs, LSTM effectively addresses the vanishing or exploding gradient issues inherent in traditional RNNs, enabling it to memorize long-term historical information and dynamically adjust its memory content. The upstream centralized controller implements dynamic traffic flow prediction using the LSTM model. Input data includes historical traffic flow data, temporal features, and environmental parameters. Historical traffic flow data includes time series data such as traffic flow, average lane speed, and lane occupancy. Temporal features include time period (peak / off-peak) and date (weekday / holiday). Environmental parameters include factors affecting visibility, such as weather (smog, rain), PM2.5 concentration, and other factors. The LSTM network is trained using historical data to learn the spatiotemporal patterns of traffic flow, such as sudden increases in traffic flow during the evening rush hour and light traffic during the early morning hours. Input data is segmented into sequences based on time windows (e.g., the past hour), and the output is a future time period (e.g., the arrival time and duration of downstream traffic). Prediction targets include adjustment time and hold time. Adjustment time: The time required for traffic to travel from upstream to downstream, used to trigger dimming of downstream streetlights in advance. Hold time: The time it takes for traffic to pass through the downstream area, which determines how long the streetlights maintain high brightness. The upstream controller collects current traffic data (e.g., speed, lane occupancy) in real time, inputs it into the LSTM model, and predicts the arrival time of downstream traffic. The prediction results are transmitted to the downstream controller via LoRaWAN+5G dual-mode communication, allowing the downstream controller to adjust the brightness and color temperature of the streetlights in advance.

[0096] Adjustment instructions are sent to the downstream centralized controller based on the LoRaWAN and 5G dual-mode redundant communication protocol (transmission delay <2 seconds, packet loss rate <0.1%), allowing the downstream street lights to pre-adjust their brightness and color temperature before traffic arrives.

[0097] The upstream centralized controller is typically located at the entrance of a road section or the starting point of traffic flow. It monitors the area's traffic volume, pedestrian flow, average lane speed, and lane occupancy in real time. Sensors also collect environmental parameters such as PM2.5, humidity, and SO2 concentration. It predicts the arrival time of downstream traffic and generates adjustment instructions (brightness, color temperature, and adjustment time). The downstream centralized controller is typically located on adjacent roads or in areas where traffic is about to arrive. It receives upstream instructions, calibrates target parameters (such as brightness and color temperature) in real time, and executes the adjustment at the designated time. It adjusts the streetlight status in advance to ensure optimal or near-optimal lighting conditions when traffic arrives.

[0098] The upstream controller uses an LSTM model to analyze historical traffic data and predict the arrival time of downstream traffic. Based on current environmental parameters (such as haze concentration), it calculates the target brightness (e.g., 100%) and color temperature (e.g., 3000K warm yellow light) required by the downstream controller. It then generates a command containing the adjustment time (when to adjust), the hold time (how long to hold), and the brightness / color temperature value. This command is sent to the downstream controller via LoRaWAN+5G dual-mode communication, ensuring fast (latency < 2 seconds) and reliable (packet loss rate < 0.1%) delivery. After receiving the command, the downstream controller uses a Kalman filter algorithm to eliminate sensor errors and calibrate the target brightness / color temperature.

[0099] In a preferred embodiment, the dynamic weight correction factor in the combined weighting algorithm is calculated as follows:

[0100] ;

[0101] in:

[0102] The objective weight calculated by the entropy weight method;

[0103] The absolute deviation between the current parameter value and the historical mean, where the historical mean is the mean of the same time period over the past seven days;

[0104] The maximum fluctuation range allowed for the parameter is set according to national standards;

[0105] The comprehensive weight after dynamic correction is: in, It is The comprehensive weight of each parameter indicates the final influence of the parameter in the brightness or color temperature adjustment model; is the weight obtained by the hierarchical analysis method, Indicates the parameters, The total number of parameters entered into the model, the specific value depends on the adjustment target. Brightness adjustment model: (traffic flow, pedestrian flow, PM2.5 concentration, humidity, ambient illumination); color temperature adjustment model: (PM2.5 concentration, SO2 concentration, humidity). The dynamically adjusted weights ( ) and the subjective weight of the AHP method ( ) combined and normalized to generate the final comprehensive weight ( The dynamic weight correction factor is used to adjust the influence of various environmental parameters in the brightness / color temperature adjustment model in real time, enabling the system to quickly respond to sudden environmental or traffic changes and optimize the sensitivity and stability of the adjustment.

[0106] In a preferred embodiment, the brightness adjustment level is divided into the following ways: the brightness judgment value interval [0,1] is divided into adaptive Gaussian intervals based on the standard deviation of historical traffic flow. The specific formula is: ,in, For the The center value of each brightness level is 25%, 50%, 75%, and 100% basic brightness; is the interval standard deviation, which is dynamically adjusted based on the standard deviation of traffic volume in the past hour.

[0107] Brightness adjustment levels are dynamically adjusted based on historical traffic fluctuations. The system divides the brightness range [0, 1] into four levels (25%, 50%, 75%, and 100%), with the center value of each level corresponding to the base brightness. By dynamically calculating the standard deviation of traffic flow over the past hour, the range for each brightness level is calculated. If traffic fluctuates significantly (large standard deviation), the range is automatically widened to avoid frequent adjustments. If traffic is stable (small standard deviation), the range is narrowed for a more responsive brightness response. This adaptive mechanism intelligently balances energy consumption and lighting needs based on actual traffic flow, ensuring driving safety while reducing energy waste.

[0108] In a preferred embodiment, in the brightness adjustment model, the inverse correlation between ambient illumination and brightness is realized by a piecewise function:

[0109] ;

[0110] in, is the base brightness (unit: %); is the ambient illuminance (unit: lux), the coefficients 0.03 and 0.01 imply the dimensionless proportional coefficient / lux, or directly take Substitute the numerical value into the calculation, This is the minimum brightness threshold (default 20%) to avoid negative values or over-dimming.

[0111] This embodiment dynamically adjusts the brightness of street lamps based on ambient illumination. The stronger the ambient light, the lower the demand for artificial lighting. The ambient illumination is divided into two intervals, and different adjustment coefficients are used to achieve brightness control. When the ambient illumination is lower than 20 lux, the brightness of the street lamp decreases rapidly as the ambient illumination increases. At this time, the basic brightness is reduced by 3% for every 1 lux increase in illumination. When the ambient illumination reaches or exceeds 20 lux, the brightness reduction slows down, and the brightness is only reduced by 1% per lux. It ensures a sensitive response to brightness changes in low-illuminance environments and avoids frequent fluctuations caused by over-adjustment in high-illuminance environments. The difference between coefficients 0.03 and 0.01 reflects the energy-saving strategy under different lighting conditions. At night, the ambient illumination is usually lower than 20 lux. At this time, the street lamp needs to assume the main lighting function, and the brightness is finely adjusted according to the changes in weak light sources such as moonlight and car lights. During the day or in strong light environments, natural light can meet basic needs. The narrowing of the brightness reduction can avoid the impact of frequent starting and stopping on the life of the lamp.

[0112] In a preferred embodiment, the color temperature adjustment levels are divided as follows: the color temperature judgment value interval [0, 1] is divided into three sub-intervals, each corresponding to a dynamic color temperature value based on the haze index:

[0113] ;

[0114] in, is the color temperature judgment value, and the calculation formula is:

[0115] ;

[0116] :PM2.5 normalized value, when PM2.5 concentration ≤300μg / m 3 hour, ; When PM2.5 concentration>300μg / m 3 hour, ;

[0117] : SO2 normalized value, ;

[0118] : Humidity normalized value, ;

[0119] is the combined weight, satisfying , the calculation formula is:

[0120] ;

[0121] The entropy weight method is used to calculate the objective weights of parameters (based on parameter volatility); :The first The subjective weight of each parameter (expert experience).

[0122] The entropy weight method calculates the objective weight:

[0123] (brightness model), Represents the first The entropy value of environmental parameters ( ), corresponding to the five parameters of traffic flow, pedestrian flow, humidity, PM2.5, and ambient illumination.

[0124] (color temperature model), Represents the first The entropy value of environmental parameters ( ), corresponding to the three parameters of PM2.5, SO2, and humidity.

[0125] The brightness model and the color temperature model are basically the same, except for the number of parameters and the meaning of the parameters. The present invention also introduces a dynamic weight correction factor, using PM2.5 as an example, and the same applies to other parameters.

[0126] is the objective weight of PM2.5 parameters in the brightness model and color temperature model, corresponding to (brightness model) and (Color temperature model).

[0127] ;

[0128] : The original PM2.5 objective weight calculated by the entropy weight method;

[0129] : the absolute deviation of the current PM2.5 concentration from the historical mean;

[0130] , the maximum fluctuation range allowed for PM2.5, such as .

[0131] When PM2.5 concentration deviates hour, Amplify and update by dynamic correction factor , to reflect the real-time fluctuation of pollution concentration. The corrected weights need to be renormalized to ensure that the sum of all parameter weights is 1. , among which, only the weight of PM2.5 is modified, and the weights of other parameters remain the original values.

[0132] In a preferred embodiment, the remote communication adopts LoRaWAN and 5G dual-mode redundant transmission protocols, with a transmission delay of less than 2 seconds and a packet loss rate of less than 0.1%.

[0133] In a preferred embodiment, in the collaborative control strategy, the adjustment time of the downstream street lamp is determined by an improved fuzzy PID control algorithm, specifically:

[0134] Average lane speed (Unit: km / h) and lane occupancy (Unit: %) is input, and a new congestion index is added The membership function of : in, The maximum design speed for the road (unit: km / h);

[0135] Output regulation time The membership function adopts dynamic asymmetric Gaussian distribution:

[0136] ;

[0137] in, is the expected value of the adjustment time calculated based on historical data, Standard deviation, adjusts the fluctuation range of time.

[0138] The downstream streetlight adjustment time is determined based on an improved fuzzy PID algorithm, which achieves precise control through dynamic perception of traffic conditions. Input parameter collection: Real-time acquisition of lane average speed (v) and lane occupancy (p). Speed reflects the efficiency of traffic flow, while occupancy reflects the degree of road congestion. For example, a speed of 30 km / h and an occupancy of 80% indicate severe congestion on the road section. Congestion index calculation: A congestion index is generated by combining speed and occupancy. , which comprehensively quantifies the degree of congestion.

[0139] Fuzzy PID dynamic adjustment includes fuzzy control and PID correction. Fuzzy control: maps vehicle speed and occupancy rate into fuzzy rules (e.g., slow vehicle speed + high occupancy rate corresponds to extended adjustment time) to generate preliminary adjustment time. PID correction: introduces the deviation between the predicted vehicle speed (LSTM model output) and the actual vehicle speed to dynamically adjust the proportional coefficient. The greater the deviation, The higher the weight, the faster the adjustment time response. The preliminary value is generated by fuzzy rules, and the dynamic Gaussian distribution adjusts the fluctuation range according to the congestion level. The correction further reduces the error, forming a closed-loop control logic of fuzzy decision-PID correction-dynamic distribution. Fuzzy control body: processing nonlinear input (vehicle speed v, lane occupancy p, congestion index ), generate preliminary adjustment time through fuzzy rules . If there is high congestion (such as >0.6), the fuzzy rule is directly extended To cope with traffic stagnation. If the vehicle speed is low (e.g. 30 km / h) and the lane occupancy is high (e.g. 80%), then the output = 100 seconds (high congestion scenario). Fuzzy rule base: specifies the downstream street light adjustment time The fuzzy rule base is based on three input variables: lane average speed v: divided into 5 fuzzy sets (slow, slower, medium, faster, fast). Lane occupancy p: divided into 5 fuzzy sets (low, lower, medium, higher, high). Congestion index , and divided into three fuzzy sets (low, medium, and high). By combining the fuzzy sets of the three input variables, the rule base is expanded to 36 IF-THEN rules. The specific rules are generated through expert experience and historical data optimization. For example, input: v = slow, p = high, = High, output: = long (corresponding to the expected value of the adjustment time =180 s , σ =0.1 ). The fuzzy rule base is pre-determined, just like a database, which can be called by any program. The present invention only retains the proportional term Dynamic adjustment, the integration and differentiation are replaced by fuzzy rules. When the prediction error is large (such as sudden congestion or sensor noise), Increase, enhance proportional action, and quickly correct the adjustment time of fuzzy output , reducing delay. You can use Indicates that It is based on fuzzy rules or fuzzy controller output preliminary , is the error or prediction error, which is the initial adjustment time of the fuzzy controller output and (historical) actual demand time in, Indicates the Adjustment operations (historical operations). The system uses the sliding window method to count the recent The mean error of the adjustment operation: , set according to actual needs (such as It can be an integer between 5-20, such as 10), balancing real-time performance and stability. The final value of is output through a dynamic asymmetric Gaussian distribution, whose mean Calculated from historical data. Through the above analysis, the output adjustment time The membership function of using dynamic asymmetric Gaussian distribution can be changed to:

[0140] ;

[0141] Preliminary values generated for fuzzy control; adjusted based on prediction error ,generate ; Gaussian distribution constraint: is the mean, standard deviation Dynamic adjustment, output final , .

[0142] In general, the present invention includes a fuzzy control stage and a PID correction stage. In the fuzzy control stage: the membership function is used to fuzzify the vehicle speed v and lane occupancy p into language variables (such as "slow + high occupancy"). Combined with fuzzy rules (such as "slow speed and high occupancy translate into long adjustment time"), a preliminary Fuzzy output. PID correction stage: (Proportional coefficient) is dynamically adjusted by the prediction error. When the prediction error is large (such as sudden congestion), Increase, force shorten Membership function: determines the range of fuzzy rule output (such as fluctuation range). Adjustment: Based on the error correction, the deviation of the fuzzy output is ensured. Quickly converge to the actual demand value. The membership function defines the basis of fuzzy rules. Provide dynamic correction, and the two together realize coarse adjustment (fuzzy) + fine adjustment (PID) closed-loop control.

[0143] In a preferred embodiment, in the improved fuzzy PID control algorithm, the proportional coefficient is dynamically adjusted The formula is:

[0144] ;

[0145] in, is the initial proportional coefficient; The vehicle speed predicted by the LSTM model (unit: km / h); is the actual detected vehicle speed (unit: km / h). , PID output ,

[0146] ;

[0147] ,in for The initial value of is calculated based on historical traffic data. Equivalent to the above embodiment , such as the distance d1 from the upstream centralized controller to the downstream centralized controller, the average speed of traffic through the road section over the past period of time (such as 1 hour) , .

[0148] In a preferred embodiment, after receiving the adjustment instruction, if the actual traffic flow deviates from the predicted value by more than 20%, the downstream centralized controller re-predicts the target brightness / color temperature value through the Kalman filter algorithm: State equation:

[0149] ;

[0150] Observation equation:

[0151] ;

[0152] in, The system is The predicted state vector at time t, is the state vector, including traffic flow and speed; , It's traffic flow. is the average vehicle speed; is the system matrix, such as , is the sampling period, is the coefficient of traffic flow changing with vehicle speed (empirical value); ,in, The impact of brightness adjustment on traffic flow, The effect of color temperature adjustment on vehicle speed; it means that brightness adjustment only suppresses vehicle speed. , is the observation matrix. is the control input, , and Respectively represent the brightness adjustment amount and color temperature adjustment amount. is the process noise and observation noise. Process noise Describes the dynamic disturbances not modeled in the state equation (such as sudden congestion, sudden weather changes, etc.), which obey the zero-mean Gaussian distribution, and the covariance matrix is , , : The process noise of traffic flow, the variance is determined by the covariance matrix in representation. The process noise of vehicle speed, the variance is given by the covariance matrix in Characterization. Observation noise Describes sensor measurement errors (such as traffic flow counting deviation and speed estimation noise), obeys zero-mean Gaussian distribution, and the covariance matrix is , , The observation noise of traffic flow, the variance is given by the covariance matrix in representation. The observation noise of vehicle speed, the variance is given by the covariance matrix in representation. yes The actual observation vector at time .

[0153] When actual traffic flow deviates from the predicted value by more than 20%, the downstream centralized controller activates the Kalman filter algorithm to recalibrate the target brightness and color temperature. The core of this algorithm is to continuously optimize adjustment parameters by dynamically balancing predicted and measured data. The system first predicts future conditions based on historical traffic flow models, such as estimating road load for the next time period based on current traffic flow and speed. This prediction process takes into account natural fluctuations in traffic flow (such as the regularity of morning and evening rush hours) as well as random interference factors (such as sudden accidents or weather effects).

[0154] In a preferred embodiment, the abnormal data processing step is also included: when the PM2.5 concentration exceeds 300 μg / m 3 When the fog penetration mode is activated, the setting is forced , directly trigger , lock the color temperature to 3000K, or force the color temperature to:

[0155] ;

[0156] The brightness is increased to 120% of the rated value until PM2.5 drops to 200 μg / m 3 The following is executed after verifying the adjustment effect through the digital twin platform.

[0157] A smart urban street light control system (see Figure 2 ),include:

[0158] Data acquisition module, used to obtain road environment parameters and traffic status parameters in real time;

[0159] The dynamic weight calculation module is deployed in the upstream centralized controller and includes:

[0160] The brightness weight unit integrates the exponential smoothing filter algorithm of PM2.5 concentration. The filtering formula is:

[0161] ;

[0162] in, is the historical average (unit: μg / m 3 ); is the current value, It is the PM2.5 concentration value after being processed by the exponential smoothing filter algorithm;

[0163] The objective weights of vehicle flow, pedestrian flow, humidity, PM2.5, and ambient illumination are calculated using the entropy weight method. The subjective weights of these parameters are calculated using the analytic hierarchy process. These weights are then combined to generate a comprehensive weight. Furthermore, a dynamic correction factor is introduced to adjust the weights and perform normalization.

[0164] Color temperature weight unit, integrated The wind speed compensation model of concentration, the compensation formula is:

[0165] in, is the current wind speed;

[0166] The objective weights of PM2.5, SO2, and humidity were calculated using the entropy weight method; the subjective weights of the above parameters were calculated using the analytic hierarchy process; and the combined weights were used to generate a comprehensive weight.

[0167] It should be noted that the combined weighting algorithm, dynamic correction factor and parameter normalization are all in the dynamic weight calculation module. Please refer to the description of the method embodiment for details. This part of the method and system corresponds to each other and is only briefly described here.

[0168] The adjustment decision module is used to generate brightness and color temperature level instructions based on the brightness and color temperature judgment values. It normalizes parameters such as vehicle flow and humidity, weights them, and generates brightness and color temperature judgment values, dividing them into brightness levels (25%, 50%, 75%, 100%) and color temperature levels (3000K, 4000K, 5000K). For more information on flexible division and level adjustment, please refer to the method examples and will not be repeated here.

[0169] Collaborative control module, including:

[0170] The upstream controller is configured to run the LSTM-based traffic flow prediction model and output the downstream adjustment time, hold time, and target brightness / color temperature values;

[0171] The downstream controller is configured to calibrate the target brightness / color temperature value in real time through a Kalman filter algorithm and perform adjustment when the predicted time arrives.

[0172] The city's smart streetlight control system achieves intelligent regulation through the collaborative operation of multiple modules. The system first uses a data acquisition module to acquire real-time road environment parameters and traffic status data, including PM2.5 concentration, wind speed, and vehicle volume. The dynamic weight calculation module optimizes this raw data, using, for example, an exponential smoothing algorithm to eliminate transient fluctuations in PM2.5 concentration monitoring values, ensuring they more accurately reflect actual pollution levels. For sulfur dioxide concentration monitoring values, the system incorporates a wind speed compensation mechanism to dynamically adjust for the impact of gas diffusion on detection accuracy.

[0173] The adjustment decision module compares processed environmental parameters with preset standard values and generates corresponding brightness and color temperature adjustment instructions. For example, it automatically increases brightness to enhance penetration during haze, or increases color temperature to enhance lighting effects during peak traffic hours. The system features a specially designed collaborative control mechanism. The upstream controller uses a deep learning model to predict traffic trends within the next 15 minutes and pre-calculate the required lighting parameter adjustment time points and duration for downstream sections.

[0174] After receiving the prediction instructions, the downstream controller uses a dynamic calibration algorithm based on real-time traffic speed changes to eliminate prediction errors. When a vehicle is about to enter the controlled area, the streetlights adjust their brightness and color temperature 3-5 seconds in advance, ensuring optimal lighting parameters as the vehicle passes through, completing a closed-loop control process of prediction, calibration, and execution.

[0175] In a preferred embodiment, the collaborative control module integrates a fuzzy PID hybrid controller, and its fuzzy rule base is expanded to 36 IF-THEN rules, and the congestion index is increased. as the third input variable.

[0176] The congestion index, calculated using real-time sensor data on vehicle and pedestrian traffic, reflects the level of road congestion. The fuzzy rule base contains 36 if-then logic rules that define control strategies for different input combinations. For example, when a large brightness error is detected and the congestion index is high, the system triggers the "rapid brightness increase" rule; if the error is small and the congestion index is low, "smooth fine-tuning" is implemented to reduce energy consumption. During actual control, the controller continuously receives environmental data, calculates PID parameters through fuzzy reasoning, and then outputs the final brightness or color temperature adjustment command based on the PID algorithm. This hybrid control mode maintains the stability of PID while using fuzzy logic to cope with complex and changing road scenarios.

[0177] In a preferred embodiment, the communication unit utilizes a blockchain-based multi-hop verification mechanism. Specifically, a hash value is generated before command transmission and recorded in a distributed ledger. Upon receiving the command, the downstream controller verifies the hash consistency and triggers a retransmission if any discrepancy occurs. The communication unit of the smart streetlight system utilizes a blockchain-based multi-hop verification mechanism to ensure the reliability and security of command transmission. When the upstream controller generates an adjustment command (such as brightness or color temperature parameters), the system generates a unique digital fingerprint (hash value) for the command and simultaneously records this fingerprint in the blockchain network's distributed ledger. The distributed ledger is maintained by multiple nodes, each of which maintains a complete hash record, ensuring data immutability and network-wide transparency. Upon receiving the command, the downstream controller immediately recalculates the hash value for the command content and compares it with the original hash value stored in the blockchain ledger. If the two match, the command has not been tampered with during transmission, and the downstream controller executes the adjustment operation. If they differ, indicating possible data corruption or malicious attack during transmission, the system automatically triggers a retransmission mechanism, requiring the upstream controller to resend the command until verification passes. The entire process does not require human intervention. Through the distributed verification characteristics of the blockchain, it avoids the risk of single point failure and achieves efficient error self-repair capabilities.

[0178] In a preferred embodiment, it also includes: an adaptive compensation module, which automatically switches to the reinforcement learning model to generate adjustment instructions when the environmental parameters exceed the safety threshold, and executes them after virtual verification through the digital twin platform. The adaptive compensation module is a safety mechanism of the smart street light system, which can ensure the reliability and safety of lighting control under extreme environmental conditions. When environmental parameters such as PM2.5 concentration and wind speed exceed the preset safety threshold, the system automatically triggers the exception handling process. The module switches to the reinforcement learning model, which dynamically generates brightness and color temperature adjustment instructions based on historical environmental data and adjustment effects. If the PM2.5 concentration suddenly soars to 300μg / m 3 The model prioritizes increasing brightness to 90% and lowering color temperature to 3000K to enhance fog penetration. These commands are then sent to the digital twin platform for virtual verification. By constructing a virtual image of the road lighting system, the platform simulates the adjusted light distribution, energy consumption, and traffic impact, avoiding the risks associated with directly manipulating physical equipment. Once verified, the commands are sent to the physical streetlights for execution, and the results of the adjustments are recorded to optimize subsequent decision-making.

[0179] The urban smart streetlight control system described in this invention achieves intelligent control through the collaboration of multiple modules. The system's data acquisition module acquires real-time environmental parameters (PM2.5, humidity, SO2 concentration, wind speed) and traffic parameters (traffic volume, speed, and lane occupancy). The dynamic weight calculation module filters and compensates this data (e.g., PM2.5 exponential smoothing filtering, SO2 wind speed compensation), generating denoised parameter values. The adjustment decision module dynamically assigns weights to each parameter using a combined weighting algorithm (entropy weighting plus analytic hierarchy process), outputs brightness and color temperature judgment values, and categorizes adaptive adjustment levels based on historical standard deviations (e.g., four levels for brightness and three levels for color temperature). The collaborative control module uses the upstream centralized controller's LSTM traffic flow prediction model to predict downstream adjustment needs, generating target brightness / color temperature values and time parameters. This is combined with the downstream Kalman filter to calibrate errors in real time, achieving a closed prediction-command-execution loop. The communication unit utilizes blockchain multi-hop verification to ensure secure command transmission. The adaptive compensation module switches to a reinforcement learning model when environmental limits are exceeded, executing emergency strategies after virtual verification using a digital twin.

[0180] Although the preferred embodiments of the present invention have been described in detail, it is still possible for those skilled in the art to make further adjustments and changes to these embodiments after mastering their basic innovative concepts. Therefore, the appended claims are intended to cover the preferred embodiments and all changes and modifications that are consistent with the scope of the invention. The above content is only a description of the preferred embodiments of the present invention and is not intended to limit the scope of the invention. It should be clear that any modification, equivalent substitution or improvement made within the spirit and principle framework of the present invention should be deemed to be included in the scope of protection of the present invention.

Claims

1. A method for controlling urban smart streetlights, characterized in that: Includes the following: S1. Collect road environmental parameters and traffic status parameters, including PM2.5 concentration, humidity, ambient illumination, SO2 concentration, and traffic status parameters including traffic flow, pedestrian flow, average lane speed, and lane occupancy; S2. Construct a brightness adjustment model and a color temperature adjustment model, calculating the weights of each parameter using a combined weighting algorithm. The brightness adjustment model uses vehicle flow, pedestrian flow, PM2.5 concentration, humidity, and ambient illumination as input parameters, and uses a combined entropy weighting method and the analytic hierarchy process to assign weights. A dynamic weight correction factor is introduced to generate a brightness judgment value. The color temperature adjustment model uses PM2.5 concentration, SO2 concentration, and humidity as input parameters, and uses a combined entropy weighting method and the analytic hierarchy process to assign weights. The model also integrates real-time wind speed data to perform diffusion compensation for SO2 concentration and generate a color temperature judgment value. S3 based on the brightness judgment value and color temperature judgment value, respectively, the brightness adjustment level and color temperature adjustment level, dynamically adjust the brightness and color temperature of the streetlights; S4. Achieve linkage between upstream and downstream streetlights through collaborative control strategies: The upstream centralized controller calculates the adjustment time, hold time, and target brightness / color temperature of downstream streetlights based on the detected traffic volume, pedestrian volume, average lane speed, and lane occupancy, combined with the LSTM-based dynamic traffic flow prediction model. By sending adjustment instructions to the downstream centralized controller based on the LoRaWAN and 5G dual-mode redundant communication protocol, the downstream street lights can complete the pre-adjustment of brightness and color temperature before the arrival of traffic; The color temperature adjustment level is divided into three sub-intervals: the color temperature judgment value interval [0, 1] is divided into three sub-intervals, which correspond to dynamic color temperature values based on the haze index: ； in, is the color temperature judgment value, and the calculation formula is: ； :PM2.5 normalized value, when PM2.5 concentration ≤300μg / m 3 hour, ; When PM2.5 concentration>300μg / m 3 hour, ; : SO2 normalized value, ; : Humidity normalized value, ; is the combined weight, satisfying , the calculation formula is: ; is calculated by the entropy weight method. objective weights of parameters; The first The subjective weight of each parameter.

2. The urban smart streetlight control method according to claim 1, characterized in that: The calculation method of the dynamic weight correction factor in the combined weighting algorithm is: ； in: The objective weight calculated by the entropy weight method; The absolute deviation between the current parameter value and the historical mean, where the historical mean is the mean of the same time period over the past seven days; The maximum fluctuation range allowed for the parameter; The comprehensive weight after dynamic correction is: in, It is The comprehensive weight of each parameter indicates the final influence of the parameter in the brightness or color temperature adjustment model; is the weight obtained by the hierarchical analysis method, Indicates the parameters, is the total number of parameters entered into the model.

3. The urban smart streetlight control method according to claim 1, characterized in that: The brightness adjustment level is divided into the following ways: the brightness judgment value interval [0,1] is divided into adaptive Gaussian intervals based on the standard deviation of historical traffic flow. The specific formula is: ,in, For the The center value of each brightness level is 25%, 50%, 75%, and 100% basic brightness; is the interval standard deviation, which is dynamically adjusted based on the standard deviation of traffic volume in the past hour.

4. The urban smart streetlight control method according to claim 1, characterized in that: In the brightness adjustment model, the inverse correlation between ambient illumination and brightness is realized by a piecewise function: ； in, is the basic brightness; is the ambient illumination, is the minimum brightness threshold.

5. The urban smart streetlight control method according to claim 1, characterized in that: In the collaborative control strategy, the adjustment time of the downstream streetlights is determined by an improved fuzzy PID control algorithm, specifically: Average lane speed and lane occupancy Added congestion index for input The membership function of : in, Designing a maximum speed for the road; Output regulation time The membership function adopts dynamic asymmetric Gaussian distribution: ； in, is the expected value of the adjustment time calculated based on historical data, is the standard deviation.

6. The urban smart streetlight control method according to claim 5, characterized in that: In the improved fuzzy PID control algorithm, the proportional coefficient is dynamically adjusted The formula is: ； in, is the initial proportional coefficient; The vehicle speed predicted by the LSTM model; is the actual detected vehicle speed.

7. The urban smart streetlight control method according to claim 1, characterized in that: It also includes abnormal data processing steps: when the PM2.5 concentration exceeds 300μg / m 3 When the fog penetration mode is activated, the setting is forced , directly trigger , lock the color temperature at 3000K; and increase the brightness to 120% of the rated value until PM2.5 drops to 200μg / m 3 the following.

8. The urban smart streetlight control system according to the urban smart streetlight control method of claim 1, characterized in that: include: Data acquisition module, used to obtain road environment parameters and traffic status parameters in real time; The dynamic weight calculation module is deployed in the centralized controller and includes: The brightness weight unit integrates the exponential smoothing filter algorithm of PM2.5 concentration. The filtering formula is: ； in, is the historical mean; is the current value, It is the PM2.5 concentration value after being processed by the exponential smoothing filter algorithm; Color temperature weight unit, integrated The wind speed compensation model of concentration, the compensation formula is: in, is the current wind speed; An adjustment decision module is used to generate a brightness level instruction and a color temperature level instruction according to the brightness judgment value and the color temperature judgment value; Collaborative control module, including: The upstream controller is configured to run the LSTM-based traffic flow prediction model and output the downstream adjustment time, hold time, and target brightness / color temperature values; The downstream controller is configured to calibrate the target brightness / color temperature value in real time through a Kalman filter algorithm and perform adjustment when the predicted time arrives.

Images