Modularized intelligent control street lamp and single lamp control system thereof

Abstract

The invention relates to the technical field of street lamp control, in particular to a modular intelligent control street lamp and a single lamp control system thereof, and the system comprises an area division module which draws a circle by taking a vertical intersection point of the street lamp and a road center line as a circle center and taking half of the distance between two adjacent street lamps as a radius, and obtains a road path section in a circumscribed rectangle of the drawn circle; the data collection module is used for collecting historical illumination training parameters of a single lamp illumination area, and the model training module is used for training a machine learning model for predicting the brightness of the street lamp based on the historical illumination training parameters, collecting real-time illumination characteristic parameters of the single lamp illumination area and predicting the brightness of the street lamp based on the trained machine learning model. According to the invention, the lighting area of each street lamp is obtained through the area division module, the lighting characteristic parameters of the lighting area of the single lamp are collected, the brightness of the street lamp is dimmed, that is, each street lamp is dimmed independently based on the lighting characteristic parameters of the lighting area of the single lamp, and the lighting effect of the street lamp is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of street lamp control, and in particular to a modular intelligent controlled street lamp and a single lamp control system thereof. Background Art

[0002] Street lamps are public facilities used to provide lighting for roads, providing a safer driving environment for vehicles on the road. With the development of urban construction and the increase in the number of vehicles, more and more roads are being built. In road construction, the management and control of street lamps and how to save energy while ensuring the lighting effect are of great significance.

[0003] The Chinese patent application with publication number CN115915547A discloses a smart street light control system, including: a central processing unit, a collection system for collecting sensor data and image data, a processing system for collecting and analyzing the data collected by the collection system, a control system for judging and performing corresponding operations on the data analyzed by the processing system, and a detection system for detecting the internal working state and operation status. This smart street light control system can determine the presence of a preset target image in the road surface video data by setting the collection system and the processing system, and then adjust the brightness of the street light according to the current brightness of the street light and the preset threshold, and send a brightness adjustment instruction to the central processing unit at the same time; the central processing unit can adjust the brightness of the street light according to the current brightness of the street light and the preset threshold, and by obtaining the visibility around the street light, the surrounding pedestrian and vehicle flow conditions, the brightness of the street light is automatically adjusted, which can significantly save electricity.

[0004] As in the above application, the existing control of street lamp brightness is generally regional control, that is, the brightness of street lamps in each area or on a section of road in a certain area is controlled by overall adjustment, and the control accuracy is low. Secondly, the street lamp brightness is automatically adjusted by collecting visibility conditions and traffic flow conditions around the street lamps. The factors considered are relatively simple, resulting in poor control effect, which affects the actual use effect of street lamp lighting. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a modular intelligent controlled street lamp and a single lamp control system thereof.

[0006] The present invention adopts the following technical solution, a modular intelligent control system for controlling street lamps, comprising:

[0007] The area division module draws a circle with the vertical intersection of the street lamp and the center line of the road as the center and half the distance between two adjacent street lamps as the radius, obtains the road path segment within the circumscribed rectangle of the drawn circle, marks the path segment as a single-lamp lighting area, and obtains the single-lamp lighting area of ​​each street lamp in turn according to the travel direction of the road;

[0008] A data collection module collects historical lighting training parameters of a single-lamp lighting area, where the historical lighting training parameters are collected when the illumination of the single-lamp lighting area reaches the road lighting standard, and the historical lighting training parameters include lighting characteristic parameters and street lamp brightness, where the lighting characteristic parameters include road type, road brightness coefficient, and meteorological brightness coefficient;

[0009] Model training module: Based on historical lighting training parameters, train a machine learning model to predict street lamp brightness, collect real-time lighting feature parameters of a single-lamp lighting area, and predict street lamp brightness based on the trained machine learning model;

[0010] The analysis control module collects the real-time street lamp brightness in the single-lamp lighting area, compares the real-time street lamp brightness with the predicted street lamp brightness, obtains the brightness decay value, compares and analyzes the brightness decay value with the preset safety brightness decay threshold, generates a first dimming instruction, and controls the working brightness of the street lamp based on the first dimming instruction.

[0011] As a further description of the above technical solution: the parameters affecting the road brightness coefficient include road slope, road surface reflectivity and road curvature;

[0012] The method for obtaining the road brightness coefficient is: performing a product operation on the road slope, road surface reflectivity and road curvature based on a piecewise function relationship to obtain the road brightness coefficient.

[0013] As a further description of the above technical solution: the parameters affecting the meteorological brightness coefficient include external environment brightness, visibility and precipitation;

[0014] The method for obtaining the meteorological brightness coefficient is: performing multiplication, piecewise function and conditional function operations on the external environment brightness, visibility and precipitation to obtain the meteorological brightness coefficient.

[0015] As a further description of the above technical solution: the method for constructing a machine learning model for predicting street light brightness includes:

[0016] Initialize the machine learning model structure for predicting street lamp brightness. The machine learning model structure for predicting street lamp brightness adopts an MLP type multi-layer forward network structure, with one input layer, two hidden layers, and one output layer; the input layer is the first input layer, and the number of nodes in the first input layer is 3, corresponding to the road type, road brightness coefficient, and meteorological brightness coefficient respectively; the hidden layer includes the first hidden layer and the second hidden layer, the number of nodes in the first hidden layer is 128, and the Relu function is used as the activation function; the number of nodes in the second hidden layer is 64, and the Relu function is used as the activation function; the output layer is the first output layer, the number of nodes in the first output layer is [1, n], and the applicable street lamp brightness is predicted, where n is the number of street lamp brightness levels.

[0017] After initializing the machine learning model structure for predicting street lamp brightness, the lighting characteristic parameters and the street lamp brightness corresponding to the lighting characteristic parameters are used to train the machine learning model for predicting street lamp brightness. The lighting characteristic parameters and the street lamp brightness corresponding to the lighting characteristic parameters are stored in a database. The database records each lighting characteristic parameter and the street lamp brightness corresponding to the lighting characteristic parameters for optimizing the model. The optimizer is Adam, the loss function is the MSE loss function, the batch size is set to 32, the number of iterations is 200 rounds, and the training ends when the loss function converges, indicating that the training is completed.

[0018] It should be noted that the Adam optimizer is one of the five major optimizers commonly used in machine learning, and its full name is Adaptive Moment Estimation.

[0019] The method for training a machine learning model for predicting street light brightness includes:

[0020] Convert the collected historical lighting training parameters into a corresponding set of feature vectors;

[0021] Each group of feature vectors is used as the input of the machine learning model, the machine learning model takes the street lamp brightness corresponding to each group of lighting characteristic parameters as the output, the street lamp brightness actually corresponding to each group of lighting characteristic parameters is used as the prediction target, and minimizing the loss function value of the machine learning model is used as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0022] As a further description of the above technical solution: the method for obtaining the brightness decay value includes:

[0023] ;

[0024] In the formula, is the brightness decay value, is the predicted street light brightness, The real-time street light brightness.

[0025] As a further description of the above technical solution: the first dimming instruction includes a first-level dimming instruction, a second-level dimming instruction, a third-level dimming instruction and a fourth-level dimming instruction, wherein the first-level dimming instruction and the second-level dimming instruction are instructions for increasing the brightness of the street lamp, and the increased brightness decreases successively, and the third-level dimming instruction and the fourth-level dimming instruction are instructions for reducing the brightness of the street lamp, and the reduced brightness increases successively.

[0026] As a further description of the above technical solution: the method of comparing and analyzing the brightness decay value with a preset safe brightness decay threshold to generate a first dimming instruction includes:

[0027] The preset safety brightness decay thresholds are SBLD1, SBLD2, SBLD3 and SBLD4, where SBLD1<SBLD2<0<SBLD3<SBLD4;

[0028] when When ≤SBLD1, a four-level dimming command is generated;

[0029] When SBLD1< When ≤SBLD2, a three-level dimming command is generated;

[0030] When SBLD2< When ≤SBLD3, the first dimming instruction is not generated;

[0031] When SBLD3< When ≤SBLD4, a secondary dimming command is generated;

[0032] when >SBLD4, a first-level dimming instruction is generated.

[0033] As a further description of the above technical solution: it also includes:

[0034] A brightness compensation module is used to collect real-time road traffic flow characteristic data of a single-lamp lighting area and whether it is a road safety lighting compensation area, and generate a brightness compensation coefficient according to the collected road traffic flow characteristic data, and generate a second dimming instruction for the street lamp based on the brightness compensation coefficient and whether it is a road safety lighting compensation area;

[0035] The road traffic flow characteristic data includes vehicle speed, traffic flow and traffic density;

[0036] The method for obtaining the brightness compensation coefficient is: performing a weighted sum operation on the road traffic flow characteristic data to obtain the brightness compensation coefficient.

[0037] The method for obtaining the road safety lighting compensation area includes obtaining historical traffic accident data information of the road area, wherein the traffic accident data information includes the accident location and the number of accidents occurring at the accident location per unit time, and a preset accident number threshold. When the number of accidents occurring per unit time exceeds the preset accident number threshold, a single-lamp lighting area corresponding to the accident location is obtained, and the single-lamp lighting area is marked as a road safety lighting compensation area.

[0038] As a further description of the above technical solution: the second dimming instruction includes a first-level light increase compensation instruction, a second-level light increase compensation instruction and a third-level light increase compensation instruction, and the lighting brightness increased by the first-level light increase compensation instruction, the second-level light increase compensation instruction and the third-level light increase compensation instruction increases in sequence;

[0039] The method for generating the second dimming instruction includes:

[0040] The preset brightness compensation coefficient thresholds are BC1 and BC2, where BC1<BC2;

[0041] when When ≤BC1, and the corresponding single-lamp lighting area is not marked, the second dimming command is not generated;

[0042] when When ≤BC1, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a first-level light-increasing compensation instruction is generated;

[0043] When B.C. 1＜ When ≤BC2, and the corresponding single-lamp lighting area is not marked, a first-level light increase compensation instruction is generated;

[0044] When B.C. 1＜ When ≤BC2, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a secondary light enhancement compensation instruction is generated;

[0045] when >BC2, and the corresponding single-lamp lighting area is not marked, a secondary light-increasing compensation instruction is generated;

[0046] when >BC2, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a three-level light enhancement compensation instruction is generated.

[0047] A modular intelligent controlled street lamp comprises a light source and a single lamp control system of the modular intelligent controlled street lamp is applied.

[0048] Beneficial effects:

[0049] The present invention provides a modular intelligent controlled street lamp. Firstly, it obtains the lighting area of ​​each street lamp through the area division module, collects the lighting characteristic parameters of the single-lamp lighting area, and dims the brightness of the street lamp, that is, each street lamp is individually dimmed based on the lighting characteristic parameters of its single-lamp lighting area, so as to improve its lighting effect and ensure its lighting accuracy. Secondly, it overcomes the problem in the prior art that the brightness of the street lamps in each area or the street lamps on a section of road in a certain area are all adjusted and controlled as a whole, and the control accuracy is low.

[0050] Secondly, by collecting historical lighting training parameters of the single-lamp lighting area, training a machine learning model that predicts the brightness of the street lamp, collecting the real-time street lamp brightness in the single-lamp lighting area, comparing the real-time street lamp brightness with the predicted street lamp brightness, generating a first dimming instruction, and controlling the working brightness of the street lamp based on the first dimming instruction. It comprehensively collects the road type, road brightness coefficient and meteorological brightness coefficient, and comprehensively considers many aspects, comprehensively considering the lighting brightness needs of the single-lamp lighting area, ensuring the reasonable lighting brightness requirements of the single-lamp lighting area, and overcoming the problem that many street lamp brightness control systems in the prior art mainly rely on changes in the intensity of external environmental light to adjust the brightness, but often cannot accurately respond to complex actual application scenarios, resulting in unsatisfactory lighting effects or uneven lighting brightness. The system overcomes these limitations by introducing machine learning models and real-time data collection, and significantly improves the lighting effect and the accuracy of brightness adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments:

[0052] Figure 1 A module connection diagram of a single-lamp control system for modular intelligent street lamp control provided in Embodiment 1 of the present invention;

[0053] Figure 2 A module connection diagram of a single-lamp control system for modular intelligent street lamp control provided in Embodiment 2 of the present invention;

[0054] Figure 3 A schematic diagram of dividing the lighting area of ​​a single lamp provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0055] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention is further described below with reference to specific diagrams. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0056] Example 1

[0057] See also Figure 1 and Figure 3 The embodiment of the present invention provides a technical solution: a single lamp control system for modular intelligent control of street lamps, comprising:

[0058] The area division module draws a circle with the vertical intersection of the street lamp and the center line of the road as the center and half the distance between two adjacent street lamps as the radius, obtains the road path segment within the circumscribed rectangle of the drawn circle, marks the path segment as a single-lamp lighting area, and obtains the single-lamp lighting area of ​​each street lamp in turn according to the travel direction of the road;

[0059] Through the area division module, the lighting area of ​​each street lamp, that is, the single-lamp lighting area, is obtained. Subsequently, the lighting characteristic parameters of the single-lamp lighting area are collected, and the brightness of the street lamp is predicted based on the trained machine learning model. That is, each street lamp is individually dimmed based on the lighting characteristic parameters of its single-lamp lighting area, thereby improving its lighting effect and ensuring its lighting accuracy.

[0060] A data collection module collects historical lighting training parameters of a single-lamp lighting area, where the historical lighting training parameters are collected when the illumination of the single-lamp lighting area reaches the road lighting standard, and the historical lighting training parameters include lighting characteristic parameters and street lamp brightness, where the lighting characteristic parameters include road type, road brightness coefficient, and meteorological brightness coefficient;

[0061] It should be noted that illumination refers to the luminous flux received per unit area, and a lux meter is used to measure light intensity. The "Urban Road Lighting Design Standard" lists specific road lighting standards, which stipulate the minimum illumination requirements for different types of roads (such as main roads, secondary roads, block roads, etc.).

[0062] The road types include express roads, medium-speed roads and slow-speed roads;

[0063] The vehicle speed limit on the expressway is between 60 and 100 km / h, allowing long-distance traffic with fewer restrictions, and is a main urban road or a dedicated urban expressway;

[0064] The speed limit on medium-speed roads is between 40 and 60 km / h, usually secondary roads or suburban roads;

[0065] The speed limit on slow roads is less than 40 kilometers per hour, and they are roads in residential areas, commercial areas, and around schools.

[0066] The parameters affecting the road brightness coefficient include road slope, road surface reflectivity and road curvature;

[0067] The road slope and road curvature can be obtained by checking road design drawings or construction documents, or directly through ground measurement. The greater the road slope, the greater the street lamp brightness required, and vice versa. The greater the road curvature, the greater the street lamp brightness required, and vice versa.

[0068] The road surface reflectivity can be collected and obtained by a photometer or a reflectometer. The photometer or reflectometer directly obtains the road surface reflectivity by measuring the ratio of the reflected light intensity to the incident light intensity. Areas with higher road surface reflectivity can effectively reflect more light, thereby enhancing the lighting effect of the road, especially at night or in low-light environments. For road lighting, a higher reflectivity can make the road surface brighter and reduce the brightness of lighting fixtures, and vice versa.

[0069] The method for obtaining the road brightness coefficient is: performing a product operation on the road slope, road surface reflectivity and road curvature based on a piecewise function relationship to obtain the road brightness coefficient.

[0070] Preferably, the expression of the road brightness coefficient is:

[0071] ;

[0072] In the formula, is the road brightness coefficient, The larger the value, the greater the brightness required. is the road slope, is the maximum value of the road slope, is the road curvature, is the maximum value of the road curvature, is the road surface reflectivity, is the maximum value of the road surface reflectivity, is the slope threshold influence coefficient, is the slope threshold, which indicates the critical value at which the slope has a significant impact on driving difficulty. is a piecewise function, when The value is 1 when , otherwise it is 0.

[0073] Parameters affecting the meteorological brightness coefficient include external environment brightness, visibility and precipitation;

[0074] It should be noted that precipitation can be obtained through meteorological monitoring equipment. When the precipitation is heavy, the brightness of street lights needs to be increased to ensure road visibility and driving safety, and vice versa.

[0075] The brightness of the external environment can be acquired through a light sensor, and the visibility can be acquired through a visibility sensor or a near-infrared sensor;

[0076] Among them, ambient brightness refers to the intensity of natural light in the environment, which affects whether the street lamp needs to be turned on or the brightness is adjusted. The higher the ambient brightness, the lower the street lamp lighting brightness requirement, and vice versa;

[0077] Visibility reflects the impact of air quality on sight distance, which is mainly determined by factors such as haze, precipitation, smoke and dust. When visibility is low, more lighting is needed to ensure road safety, and vice versa.

[0078] The method for obtaining the meteorological brightness coefficient is: performing multiplication, piecewise function and conditional function operations on the external environment brightness, visibility and precipitation to obtain the meteorological brightness coefficient.

[0079] Preferably, the calculation formula of the meteorological brightness coefficient is:

[0080] ;

[0081] In the formula, is the meteorological brightness coefficient, is the external environment brightness, For visibility, is the precipitation, is the maximum value of visibility, is the maximum amount of precipitation, is the precipitation threshold, is the precipitation threshold impact coefficient (dimensionless), which is used to adjust the additional impact when the precipitation exceeds the threshold. is a piecewise function, when The value is 1 when , otherwise it is 0.

[0082] The model training module trains a machine learning model for predicting street lamp brightness based on historical lighting training parameters, collects real-time lighting feature parameters of the single-lamp lighting area, and predicts the street lamp brightness based on the trained machine learning model; it should be noted that the street lamp brightness is the brightness of the street lamp in the single-lamp lighting area.

[0083] The method for constructing the machine learning model for predicting street lamp brightness includes:

[0084] Initialize the machine learning model structure for predicting street lamp brightness. The machine learning model structure for predicting street lamp brightness adopts an MLP type multi-layer forward network structure, with one input layer, two hidden layers, and one output layer; the input layer is the first input layer, and the number of nodes in the first input layer is 3, corresponding to the road type, road brightness coefficient, and meteorological brightness coefficient respectively; the hidden layer includes the first hidden layer and the second hidden layer, the number of nodes in the first hidden layer is 128, and the Relu function is used as the activation function; the number of nodes in the second hidden layer is 64, and the Relu function is used as the activation function; the output layer is the first output layer, the number of nodes in the first output layer is [1, n], and the applicable street lamp brightness is predicted, where n is the number of street lamp brightness levels.

[0085] After initializing the machine learning model structure for predicting street lamp brightness, the lighting characteristic parameters and the street lamp brightness corresponding to the lighting characteristic parameters are used to train the machine learning model for predicting street lamp brightness. The lighting characteristic parameters and the street lamp brightness corresponding to the lighting characteristic parameters are stored in a database. The database records each lighting characteristic parameter and the street lamp brightness corresponding to the lighting characteristic parameters for optimizing the model. The optimizer is Adam, the loss function is the MSE loss function, the batch size is set to 32, the number of iterations is 200 rounds, and the training ends when the loss function converges, indicating that the training is completed.

[0086] It should be noted that the Adam optimizer is one of the five major optimizers commonly used in machine learning, and its full name is Adaptive Moment Estimation.

[0087] The method for training a machine learning model for predicting street light brightness includes:

[0088] Convert the collected historical lighting training parameters into a corresponding set of feature vectors;

[0089] Each group of feature vectors is used as the input of the machine learning model, the machine learning model takes the street lamp brightness corresponding to each group of lighting characteristic parameters as the output, the street lamp brightness actually corresponding to each group of lighting characteristic parameters is used as the prediction target, and minimizing the loss function value of the machine learning model is used as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0090] The machine learning model may be one of support vector machine regression, random forest regression or neural network regression models.

[0091] The loss function value of the machine learning model is the mean square error.

[0092] By transforming the loss function The model is trained with minimization as the goal, so that the machine learning model can better fit the data, thereby improving the performance and accuracy of the model.

[0093] In the loss function is the loss function value of the machine learning model, is the feature vector group number; is the number of eigenvector groups; For the The street light brightness corresponding to the group eigenvector, For the The street light brightness that the group of feature vectors actually corresponds to.

[0094] Other model parameters of the machine learning model, target loss value, optimization algorithm, training set test set validation set ratio, and loss function optimization are all achieved through actual engineering implementation and continuous experimental tuning.

[0095] The analysis control module collects the real-time street lamp brightness in the single-lamp lighting area, compares the real-time street lamp brightness with the predicted street lamp brightness, obtains the brightness decay value, compares and analyzes the brightness decay value with the preset safety brightness decay threshold, generates a first dimming instruction, and controls the working brightness of the street lamp based on the first dimming instruction.

[0096] It should be noted that the method for collecting the real-time street lamp brightness in a single-lamp lighting area includes:

[0097] Direct measurement: a sensor is set at the connection between the side of the light panel and the light housing to collect scattered light, which does not affect the external lighting and can realize real-time collection of street light brightness.

[0098] Indirect measurement, by measuring the current, voltage and power of the street lamp, combined with the light efficiency parameters of the lamp, the brightness is indirectly calculated, which can also avoid shadow problems.

[0099] The method for obtaining the brightness decay value includes:

[0100] ;

[0101] In the formula, is the brightness decay value, is the predicted street light brightness, The real-time street light brightness.

[0102] The first dimming instruction includes a first-level dimming instruction, a second-level dimming instruction, a third-level dimming instruction and a fourth-level dimming instruction, wherein the first-level dimming instruction and the second-level dimming instruction are instructions for increasing the brightness of the street lamp, and the increased brightness decreases successively, and the third-level dimming instruction and the fourth-level dimming instruction are instructions for reducing the brightness of the street lamp, and the reduced brightness increases successively.

[0103] The method of comparing and analyzing the brightness decay value with a preset safe brightness decay threshold to generate a first dimming instruction includes:

[0104] The preset safety brightness decay thresholds are SBLD1, SBLD2, SBLD3 and SBLD4, wherein SBLD1<SBLD2<0<SBLD3<SBLD4, and the safety brightness decay thresholds are determined by those skilled in the art according to data fitting;

[0105] when When ≤SBLD1, a four-level dimming command is generated;

[0106] When SBLD1< When ≤SBLD2, a three-level dimming command is generated;

[0107] When SBLD2< When ≤SBLD3, the first dimming instruction is not generated;

[0108] When SBLD3< When ≤SBLD4, a secondary dimming command is generated;

[0109] when When >SBLD4, a first-level dimming instruction is generated.

[0110] In this implementation, firstly, the lighting area of ​​each street lamp is obtained through the area division module, the lighting characteristic parameters of the single-lamp lighting area are collected, and the brightness of the street lamp is dimmed, that is, each street lamp is individually dimmed based on the lighting characteristic parameters of its single-lamp lighting area, so as to improve its lighting effect and ensure its lighting accuracy. Secondly, the problem of low control accuracy in the prior art that the brightness of street lamps in each area or on a section of road in a certain area is controlled by overall adjustment is overcome.

[0111] Secondly, by collecting historical lighting training parameters of the single-lamp lighting area, training a machine learning model that predicts the brightness of street lights, collecting real-time street light brightness in the single-lamp lighting area, comparing the real-time street light brightness with the predicted street light brightness, generating the first dimming instruction, and controlling the working brightness of the street light based on the first dimming instruction, it comprehensively collects road types, road brightness coefficients, and meteorological brightness coefficients, comprehensively considers many aspects, and comprehensively considers the lighting brightness needs of the single-lamp lighting area, ensuring the reasonable lighting brightness requirements of the single-lamp lighting area, and overcoming the existing technology that many street light brightness control systems mainly rely on changes in the intensity of external ambient light to adjust the brightness, but often cannot accurately respond to complex actual application scenarios, resulting in unsatisfactory lighting effects or uneven lighting brightness. The system overcomes these limitations by introducing machine learning models and real-time data collection, and significantly improves the accuracy of lighting effects and brightness adjustment.

[0112] Example 2

[0113] like Figure 2 As shown, this embodiment discloses a brightness compensation module based on the above embodiment;

[0114] A brightness compensation module is used to collect real-time road traffic flow characteristic data of a single-lamp lighting area and whether it is a road safety lighting compensation area, and generate a brightness compensation coefficient according to the collected road traffic flow characteristic data, and generate a second dimming instruction for the street lamp based on the brightness compensation coefficient and whether it is a road safety lighting compensation area;

[0115] The road traffic flow characteristic data includes vehicle speed, traffic flow and traffic density;

[0116] It should be noted that vehicle speed data is usually obtained through radar sensors, video surveillance systems or traffic monitoring systems (such as automatic vehicle speed detectors). The relationship between vehicle speed and lighting brightness is that high vehicle speeds require higher lighting brightness, especially on highways or main roads, to ensure that drivers can see potential obstacles and pedestrians in time; at low speeds, the lighting brightness requirement is relatively low, especially in traffic congestion or low-speed driving on urban roads;

[0117] Traffic flow data can be obtained through geomagnetic sensors, video surveillance, radar sensors or traffic flow counters. The relationship between traffic flow and lighting brightness is that high traffic (such as peak hours) usually requires higher lighting brightness to ensure sufficient sight distance and safety, while low traffic (such as late night hours) can appropriately reduce the brightness to save energy;

[0118] By analyzing traffic flow and vehicle speed, traffic density data can be calculated. The relationship between traffic density and lighting brightness is that high traffic density means more vehicles are concentrated on a smaller road section, which usually increases the risk of accidents. Therefore, higher lighting brightness is required to enhance the driver's sight distance. When traffic density is low, the lighting brightness can be reduced to reduce energy consumption. Traffic density indicates the number of vehicles per unit length of road.

[0119] The method for calculating traffic density based on traffic flow and vehicle speed is:

[0120] ;

[0121] In the formula, is the traffic density, For traffic flow, For vehicle speed.

[0122] The method for obtaining the brightness compensation coefficient is: performing a weighted sum operation on the road traffic flow characteristic data to obtain the brightness compensation coefficient.

[0123] Preferably, the calculation formula of the brightness compensation coefficient is:

[0124] ;

[0125] In the formula, is the brightness compensation coefficient, is the vehicle speed, For traffic flow, is the traffic density, , and is the weight coefficient, and , and Greater than 0;

[0126] It should be noted that the size of the weight coefficient is a specific value obtained by quantifying each data to facilitate subsequent comparison. The size of the weight coefficient depends on the number of comprehensive parameters and the preliminary setting of the corresponding weight coefficient for each set of comprehensive parameters by technical personnel in this field.

[0127] The method for obtaining the road safety lighting compensation area includes obtaining historical traffic accident data information of the road area, wherein the traffic accident data information includes the accident location and the number of accidents occurring at the accident location within a unit time, and a preset accident number threshold. When the number of accidents occurring within a unit time exceeds the preset accident number threshold, obtaining a single-lamp lighting area corresponding to the accident location, marking the single-lamp lighting area as a road safety lighting compensation area;

[0128] It should be noted that if a road area has a history of multiple traffic accidents, traffic crimes or traffic conflicts, a higher lighting brightness is required to enhance the sense of security and visibility and reduce the probability of accidents.

[0129] The second dimming instruction includes a first-level light increase compensation instruction, a second-level light increase compensation instruction and a third-level light increase compensation instruction, and the illumination brightness increased by the first-level light increase compensation instruction, the second-level light increase compensation instruction and the third-level light increase compensation instruction increases in sequence;

[0130] The method for generating the second dimming instruction includes:

[0131] The preset brightness compensation coefficient thresholds are BC1 and BC2, where BC1<BC2; the brightness compensation coefficient thresholds are determined by those skilled in the art according to data fitting.

[0132] when When ≤BC1, and the corresponding single-lamp lighting area is not marked, the second dimming command is not generated;

[0133] when When ≤BC1, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a first-level light-increasing compensation instruction is generated;

[0134] When B.C. 1＜ When ≤BC2, and the corresponding single-lamp lighting area is not marked, a first-level light increase compensation instruction is generated;

[0135] When B.C. 1＜ When ≤BC2, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a secondary light enhancement compensation instruction is generated;

[0136] when >BC2, and the corresponding single-lamp lighting area is not marked, a secondary light-increasing compensation instruction is generated;

[0137] when >BC2, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a three-level light enhancement compensation instruction is generated.

[0138] Based on the above embodiments, this embodiment further adds a brightness compensation instruction, that is, real-time road traffic flow characteristic data (vehicle speed, traffic flow and traffic density) is collected, and a brightness compensation coefficient is generated according to the collected road traffic flow characteristic data, and a second dimming instruction is generated based on the brightness compensation coefficient and whether the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, so that the single-lamp control system of the modular intelligent control street lamp can further automatically adjust the lighting scheme of the street lamp brightness according to the road traffic flow characteristic data and traffic safety, further improve energy efficiency, reduce energy waste, and at the same time ensure the safety of road traffic and lighting quality.

[0139] Example 3

[0140] A modular intelligent controlled street lamp comprises a light source and a single lamp control system of the modular intelligent controlled street lamp is applied.

[0141] The basic principles, main features and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and the description in the specification are only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A modular intelligent street lamp control system, characterized in that: include: The area division module draws a circle with the vertical intersection of the street lamp and the center line of the road as the center and half the distance between two adjacent street lamps as the radius, obtains the road path segment within the circumscribed rectangle of the drawn circle, marks the path segment as a single-lamp lighting area, and obtains the single-lamp lighting area of ​​each street lamp in turn according to the travel direction of the road; A data collection module collects historical lighting training parameters of a single-lamp lighting area, where the historical lighting training parameters are collected when the illumination of the single-lamp lighting area reaches the road lighting standard, and the historical lighting training parameters include lighting characteristic parameters and street lamp brightness, where the lighting characteristic parameters include road type, road brightness coefficient, and meteorological brightness coefficient; Model training module: Based on historical lighting training parameters, train a machine learning model to predict street lamp brightness, collect real-time lighting feature parameters of a single-lamp lighting area, and predict street lamp brightness based on the trained machine learning model; The analysis control module collects the real-time street lamp brightness in the single-lamp lighting area, compares the real-time street lamp brightness with the predicted street lamp brightness, obtains the brightness decay value, compares and analyzes the brightness decay value with the preset safety brightness decay threshold, generates a first dimming instruction, and controls the working brightness of the street lamp based on the first dimming instruction.

2. According to claim 1, a modular intelligent street lamp control system for single lamps, characterized in that: The parameters affecting the road brightness coefficient include road slope, road surface reflectivity and road curvature; The method for obtaining the road brightness coefficient is: performing a product operation on the road slope, road surface reflectivity and road curvature based on a piecewise function relationship to obtain the road brightness coefficient.

3. According to claim 1, a modular intelligent street lamp control system for single lamps is characterized in that: Parameters affecting the meteorological brightness coefficient include external environment brightness, visibility and precipitation; The method for obtaining the meteorological brightness coefficient is: performing multiplication, piecewise function and conditional function operations on the external environment brightness, visibility and precipitation to obtain the meteorological brightness coefficient.

4. According to claim 1, a modular intelligent street lamp control system for single lamps is characterized in that: The method for constructing the machine learning model for predicting street lamp brightness includes: Initialize the machine learning model structure for predicting street lamp brightness. The machine learning model structure for predicting street lamp brightness adopts an MLP type multi-layer forward network structure, with one input layer, two hidden layers, and one output layer; the input layer is the first input layer, and the number of nodes in the first input layer is 3, corresponding to the road type, road brightness coefficient, and meteorological brightness coefficient respectively; the hidden layer includes the first hidden layer and the second hidden layer, the number of nodes in the first hidden layer is 128, and the Relu function is used as the activation function; the number of nodes in the second hidden layer is 64, and the Relu function is used as the activation function; the output layer is the first output layer, the number of nodes in the first output layer is [1, n], and the applicable street lamp brightness is predicted, and n is the number of street lamp brightness levels; After initializing the machine learning model structure for predicting street lamp brightness, the lighting characteristic parameters and the street lamp brightness corresponding to the lighting characteristic parameters are used to train the machine learning model for predicting street lamp brightness. The lighting characteristic parameters and the street lamp brightness corresponding to the lighting characteristic parameters are stored in a database. The database records each lighting characteristic parameter and the street lamp brightness corresponding to the lighting characteristic parameters for optimizing the model. The optimizer is Adam, the loss function is the MSE loss function, the batch size is set to 32, the number of iterations is 200 rounds, and the training ends when the loss function converges. The method for training a machine learning model for predicting street light brightness includes: Convert the collected historical lighting training parameters into a corresponding set of feature vectors; Each group of feature vectors is used as the input of the machine learning model, the machine learning model takes the street lamp brightness corresponding to each group of lighting characteristic parameters as the output, the street lamp brightness actually corresponding to each group of lighting characteristic parameters is used as the prediction target, and minimizing the loss function value of the machine learning model is used as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

5. According to claim 1, a modular intelligent street lamp control system for single lamps is characterized in that: The method for obtaining the brightness decay value comprises: ； In the formula, is the brightness decay value, is the predicted street light brightness, The real-time street light brightness.

6. According to claim 5, a modular intelligent street lamp control system for single lamps is characterized in that: The first dimming instruction includes a first-level dimming instruction, a second-level dimming instruction, a third-level dimming instruction and a fourth-level dimming instruction, wherein the first-level dimming instruction and the second-level dimming instruction are instructions for increasing the brightness of the street lamp, and the increased brightness decreases successively, and the third-level dimming instruction and the fourth-level dimming instruction are instructions for reducing the brightness of the street lamp, and the reduced brightness increases successively.

7. A modular intelligent street lamp control system according to claim 6, characterized in that: The method of comparing and analyzing the brightness decay value with a preset safe brightness decay threshold to generate a first dimming instruction includes: The preset safety brightness decay thresholds are SBLD1, SBLD2, SBLD3 and SBLD4, where SBLD1<SBLD2<0<SBLD3<SBLD4; when When ≤SBLD1, a four-level dimming command is generated; When SBLD1< When ≤SBLD2, a three-level dimming command is generated; When SBLD2< When ≤SBLD3, the first dimming instruction is not generated; When SBLD3< When ≤SBLD4, a secondary dimming command is generated; when >SBLD4, a first-level dimming instruction is generated.

8. According to claim 1, a modular intelligent street lamp control system for single lamps, characterized in that: Also includes: A brightness compensation module is used to collect real-time road traffic flow characteristic data of a single-lamp lighting area and whether it is a road safety lighting compensation area, and generate a brightness compensation coefficient according to the collected road traffic flow characteristic data, and generate a second dimming instruction for the street lamp based on the brightness compensation coefficient and whether it is a road safety lighting compensation area; The road traffic flow characteristic data includes vehicle speed, traffic flow and traffic density; The method for obtaining the brightness compensation coefficient is: performing a weighted sum operation on the road traffic flow characteristic data to obtain the brightness compensation coefficient; The method for obtaining the road safety lighting compensation area includes obtaining historical traffic accident data information of the road area, wherein the traffic accident data information includes the accident location and the number of accidents occurring at the accident location per unit time, and a preset accident number threshold. When the number of accidents occurring per unit time exceeds the preset accident number threshold, a single-lamp lighting area corresponding to the accident location is obtained, and the single-lamp lighting area is marked as a road safety lighting compensation area.

9. The modular intelligent street lamp control system according to claim 8, characterized in that: The second dimming instruction includes a first-level light increase compensation instruction, a second-level light increase compensation instruction and a third-level light increase compensation instruction, and the illumination brightness increased by the first-level light increase compensation instruction, the second-level light increase compensation instruction and the third-level light increase compensation instruction increases in sequence; The method for generating the second dimming instruction includes: The preset brightness compensation coefficient thresholds are BC1 and BC2, where BC1<BC2; when When ≤BC1, and the corresponding single-lamp lighting area is not marked, the second dimming command is not generated; when When ≤BC1, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a first-level light-increasing compensation instruction is generated; When B.C. 1＜ When ≤BC2, and the corresponding single-lamp lighting area is not marked, a first-level light increase compensation instruction is generated; When B.C. 1＜ When ≤BC2, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a secondary light-increasing compensation instruction is generated; when >BC2, and the corresponding single-lamp lighting area is not marked, a secondary light-increasing compensation instruction is generated; when >BC2, and the corresponding single-lamp lighting area is marked as a road safety lighting compensation area, a three-level light enhancement compensation instruction is generated.

10. A modular intelligent controlled street lamp, comprising a light source, characterized in that: A single lamp control system for modular intelligently controlling street lamps according to any one of claims 8 to 9 is applied.

Images