An intelligent control system and method for LED light source based on data analysis

Abstract

The invention discloses an intelligent control system and method for LED light sources based on data analysis, and belongs to the field of artificial intelligence technology. A sensor group is responsible for monitoring the ambient light intensity, vehicle driving conditions, and the existence of surrounding objects to form a data group to be analyzed; according to the ambient light intensity near different street lamps in the data group to be analyzed and the geographical locations of different street lamps, the light intensity of the street lamp interval area under the mutual influence of different street lamps is calculated to form a light intensity data set; for the abnormal light intensity data set, whether the abnormal light intensity causes visual obstruction to the car driver is analyzed, and the brightness of the LED light source of the corresponding street lamp is linearly adjusted; according to the data group to be analyzed, the alarm-type LED light sources under all street lamp poles in the road section from the current position of the vehicle to the position of the surrounding objects are controlled to flash; when surrounding objects exist, the objects are identified, and when no vehicles and personnel are detected, the brightness of the street lamp is automatically dimmed.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent control system and method for an LED light source based on data analysis. Background Art

[0002] When driving at night, the driver's vision is limited by the lighting conditions, and driving safety is easily affected by too strong or too weak light. Especially when encountering emergencies or abnormal road conditions, the driver needs a clear field of vision to react in advance. Therefore, intelligent street light LED light source control can improve the driver's driving safety.

[0003] Traditional street light control systems usually control the brightness of street lights based on a fixed schedule or by using photoresistors to sense light intensity. This method cannot be adjusted according to real-time traffic conditions and environmental changes, which can easily lead to energy waste or visual fatigue. Traditional street light systems cannot provide instant warnings based on the real-time driving conditions of vehicles and the real-time existence of surrounding objects, which can cause drivers to be unable to take timely measures in potentially dangerous situations, increasing the risk of traffic accidents. Traditional street light control systems cannot adjust the brightness of street lights according to real-time conditions, which leads to energy waste and increases urban energy consumption. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent control system and method for LED light sources based on data analysis to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A method for intelligently controlling an LED light source based on data analysis, the method comprising the following steps:

[0007] S100, the sensor group installed near the street lamp on the highway is responsible for monitoring the ambient light intensity, vehicle driving conditions and the existence of surrounding objects, collecting data transmitted by the sensor group in real time to form a data group to be analyzed, and adding the geographical location and number of the street lamp to the data group to be analyzed;

[0008] S200, calculating the light intensity of the area between street lamps under the mutual influence of different street lamps according to the ambient light intensity near different street lamps in the data group to be analyzed and the geographical locations of different street lamps, so as to form a light intensity data set of each area on the road section;

[0009] S300, according to the light intensity data sets of each area, obtaining an abnormal light intensity data set, using a machine learning algorithm to analyze whether the abnormal light intensity causes visual obstruction to the car driver, and linearly adjusting the brightness of the LED light source of the corresponding street lamp to maintain the light intensity of each area within a normal range;

[0010] S400, according to the vehicle driving conditions and the existence of surrounding objects in the data group to be analyzed, when a sensor near a distant street lamp detects that a vehicle at a distance of A meters is driving in the direction of a location where there are surrounding objects, control the alarm-type LED light sources under all street lamp poles in the road section from the current position of the vehicle to the location where there are surrounding objects to flash;

[0011] S500 uses the SSD algorithm to identify objects when the sensor detects the presence of surrounding objects. When no cars, pedestrians or cyclists are detected around, the current to control the LED light source is sent to the street lamp to automatically dim the brightness of the LED light source of the street lamp in the road section.

[0012] Further, step S200 includes:

[0013] S201, obtaining geographic location information and number information of each street lamp, and determining the boundary of the lighting range of each street lamp, wherein the boundary defines the interval area between adjacent street lamps;

[0014] S202, taking into account the lighting range of the street lamps, the mutual influence between the street lamps and the ambient light intensity, and using the light attenuation model to calculate the light intensity of the area between the street lamps;

[0015] S203, integrating the light intensity calculated for each street lamp interval area with the ambient light intensity near different street lamps to form a light intensity data set for each area of ​​the road section;

[0016] S204, verify the consistency of each data point in the light intensity data set to ensure that the format and unit of the data are unified and there are no errors or abnormal values; perform field measurements to verify whether the light intensity in the light intensity data set is consistent with the actual situation, and correct the data if it is inconsistent.

[0017] Further, according to step S202, the illumination attenuation model uses the inverse fourth law to describe the attenuation law of light in space, and the mathematical expression is as follows:

[0018]

[0019] Where I is the illumination at the target point, is the initial illuminance of the LED light source, and d is the distance from the target point to the LED light source; for each street lamp, the initial illuminance of the light source is determined according to the shape and size of its lighting range. ; For each target point within the street lamp interval area, calculate the distance d from the target point to the nearest street lamp.

[0020] Further, according to step S203, light intensity integration is performed in the following manner:

[0021] For a target point in the overlapping area of ​​several street lamp lighting ranges, the influence of several street lamps on the illumination of the target point is considered, and the illumination I of the target point of several street lamps is superimposed to obtain the total illumination of the target point; the influence of the ambient light intensity on the illumination of the target point is considered, and the ambient light intensity is added to the total illumination of the target point to obtain the final illumination of the target point;

[0022] For each target point in the street lamp interval area, the final illuminance is obtained according to the illuminance attenuation model and light intensity integration method. By traversing all target points, the illuminance distribution of each point in the area is calculated to form a light intensity data set for each area under the road section.

[0023] Further, according to step S204, each data point in the light intensity data set is checked to ensure the uniformity of format and unit, wherein the checking includes checking whether the data is recorded in the same format and ensuring that all data use the same unit to represent the light intensity; checking whether there are abnormal values, wherein the abnormal values ​​include values ​​outside a reasonable range, null values ​​or inconsistent data points;

[0024] In the field, select the area between street lamps for field sampling to perform field measurement of light intensity. The area can cover different lighting conditions, including positions close to street lamps, intervals between street lamps, and positions far away from street lamps. The light intensity obtained by field measurement is compared with the light intensity at the corresponding position in the light intensity dataset. If the comparison results are inconsistent, data correction and adjustment are required, and steps S201-S203 are re-executed.

[0025] Furthermore, step S300 includes:

[0026] S301, extracting an abnormal light intensity data set from the light intensity data sets of each region, wherein the abnormal light intensity refers to a data point that deviates significantly from the surrounding light intensity, indicating that the light is too bright or too dark;

[0027] S302, selecting a random forest algorithm to analyze whether the abnormal light intensity causes visual obstruction to the car driver, using the abnormal light intensity data set to train a machine learning model, using a test set to evaluate the performance of the model, and using the trained machine learning model to analyze the abnormal light intensity data;

[0028] S303. According to the analysis results of the machine learning model, linearly adjust the brightness of the street lamp LED light source in the area where the abnormal light intensity is located to maintain the light intensity in each area within the normal range.

[0029] Further, according to step S302, a training data set is prepared from the abnormal light intensity data set, wherein the training data set includes input features and corresponding labels, wherein the input features are abnormal light intensity data, and the labels describe whether visual field obstruction is caused, and are used for the machine learning model to learn the relationship between the features and the labels; the prepared training data set is divided into a training set and a test set, wherein the training set is used to train the model, and the test set is used to evaluate the performance of the model;

[0030] The machine learning model uses a random forest algorithm and constructs several decision trees at the same time, each of which performs random sampling and feature selection on the data set; the trained random forest model is evaluated using test set data, and the evaluation indicators include accuracy, precision, recall rate and F1 value, and the model parameters are adjusted according to the evaluation results of the test set; the trained random forest model is used to analyze new abnormal light intensity data, and the model predicts whether the input abnormal light intensity data will cause visual obstruction to the car driver based on the input abnormal light intensity data.

[0031] Furthermore, step S400 includes:

[0032] S401: When a sensor near a distant street lamp detects that a vehicle at a distance of A meters is traveling in a direction where there are objects around, it is determined that there is a potential danger;

[0033] S402, when a potential danger is detected, immediately controlling the warning LED light sources under all street lamp poles in the road section from the current position of the vehicle to the position of surrounding objects to flash;

[0034] S403: The warning LED light source continues to flash until the vehicle leaves the potential danger area or the potential danger situation is resolved.

[0035] Further, according to step S401, the method for calculating the distance A meters is as follows:

[0036] S401-1. When a vehicle is detected at the farthest safe distance from the position where there are surrounding objects, the current vehicle speed is obtained through the sensor group at the vehicle position. , the current vehicle model and the road surface slip coefficient c are obtained through the convolutional neural network; the moving speed of the object is obtained through the sensor at the location where there is an object around ;According to the distance between the two sensor groups, obtain the distance D between the vehicle and the object;

[0037] S401-2. According to the type of the current vehicle, obtain the conventional braking distance S of the current vehicle from the reference braking distance table of the current vehicle model. The calculation formula for the predicted braking distance P is as follows: ;

[0038] S401-3. Considering the driver's reaction distance G, use the following mathematical formula to calculate the distance A of the potential danger situation:

[0039]

[0040] in, is the relative time between the vehicle and the object.

[0041] Further, according to step S500, when the sensor detects the presence of an object in the surrounding area, the system uses the SSD algorithm to identify the object. SSD is a classic target detection algorithm that can simultaneously detect and classify targets in a single neural network, with high detection speed and accuracy. The SSD algorithm classifies the detected objects and identifies whether they include traffic participants such as cars, pedestrians, or cyclists. The identification of these categories is to determine whether there is a traffic situation, so as to decide whether to adjust the brightness of the LED light source of the street lamp.

[0042] If the sensor does not detect any traffic participants such as cars, pedestrians or cyclists around, the system will send instructions to the street lights based on this situation to control the current of the LED light source and automatically dim the brightness of the LED light source of the street lights in the road section. This can be achieved through an intelligent control system that adjusts the brightness level of the LED light source to save energy and reduce light pollution.

[0043] An intelligent control system for LED light sources based on data analysis, comprising:

[0044] Sensor module: responsible for real-time monitoring of ambient light intensity, vehicle driving conditions and the presence of surrounding objects to form a data group to be analyzed; including: a light sensor unit, a vehicle detection sensor unit, an object detection sensor unit, a data transmission unit and a data processing unit, wherein the light sensor unit is used to monitor the ambient light intensity, the vehicle detection sensor unit is used to monitor the vehicle driving conditions, the object detection sensor unit is used to monitor the presence of surrounding objects, the data transmission unit is used to transmit the data collected by the sensor to the system, and the data processing unit is used to pre-process the sensor data to form a data group to be analyzed and add the geographical location and number of the street lamp;

[0045] Light intensity calculation module: according to the ambient light intensity near different street lamps in the data group to be analyzed and the geographical locations of different street lamps, the light intensity of the area between street lamps under the mutual influence of different street lamps is calculated to form a light intensity data set of each area under the road section; including: a light intensity calculation unit and a data integration unit, wherein the light intensity calculation unit is used to calculate the light intensity of each area of ​​the road section according to the sensor data and the street lamp location information, and the data integration unit is used to integrate the calculated light intensity data into a data set;

[0046] Abnormal light intensity analysis module: obtains abnormal light intensity data sets from the light intensity data sets of each area, uses machine learning algorithms to analyze whether the abnormal light intensity causes visual obstruction to the car driver, and makes corresponding light source brightness adjustments; includes: abnormal light intensity detection unit, machine learning algorithm unit and light source brightness adjustment unit, wherein the abnormal light intensity detection unit is used to detect whether the light intensity is abnormal, the machine learning algorithm unit is used to analyze the abnormal light intensity data and determine whether it causes visual obstruction to the driver, and the light source brightness adjustment unit is used to adjust the LED light source brightness of the street lamp according to the analysis results;

[0047] Alarm LED light source control module: when a remote sensor detects that the vehicle is traveling in the direction of a location where there are objects around, the alarm LED light sources under all street lamp poles in the road section from the current location of the vehicle to the location where there are objects around are controlled to flash; including: an alarm LED light source control unit, used to control the alarm LED light source to flash according to the location of the vehicle and the location of the surrounding objects;

[0048] Object recognition module: When the sensor detects the existence of surrounding objects, it recognizes the objects and adjusts the brightness of the LED light source of the street lamp according to the recognition results; it includes: an object recognition unit and a light source brightness control unit, wherein the object recognition unit uses the SSD algorithm to recognize the surrounding objects detected by the sensor, and the light source brightness control unit adjusts the brightness of the LED light source of the street lamp according to the recognition results.

[0049] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0050] The present invention can timely warn drivers of potential dangers by real-time monitoring of vehicle driving conditions and the presence of surrounding objects, as well as intelligently controlling the brightness of LED light sources of street lamps, thereby effectively reducing the incidence of traffic accidents and improving traffic safety.

[0051] Traditional street lamp systems often use a fixed brightness control method, which cannot adjust the brightness according to real-time conditions, which easily leads to energy waste. However, the present invention uses intelligent control of the LED light source brightness and adjusts the street lamp brightness according to real-time data, achieving more efficient energy utilization and energy saving.

[0052] The present invention can optimize the brightness distribution of street lamps and improve the visual comfort of drivers and the road lighting effect by calculating the light intensity of the street lamp interval area according to the ambient light intensity and geographical location near different street lamps, and adjusting the LED light source brightness of the street lamp according to abnormal light intensity data. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0054] Figure 1 It is a schematic diagram of the steps of an intelligent control method of an LED light source based on data analysis of the present invention;

[0055] Figure 2 It is a system structure diagram of an intelligent control system of an LED light source based on data analysis of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] See also Figure 1 and Figure 2 , the present invention provides a technical solution:

[0058] like Figure 1 A step diagram of an LED light source intelligent control method based on data analysis is shown in the figure, an LED light source intelligent control method based on data analysis, the method comprises the following steps:

[0059] S100, the sensor group installed near the street lamp on the highway is responsible for monitoring the ambient light intensity, vehicle driving conditions and the existence of surrounding objects, collecting data transmitted by the sensor group in real time to form a data group to be analyzed, and adding the geographical location and number of the street lamp to the data group to be analyzed;

[0060] S200, calculating the light intensity of the area between street lamps under the mutual influence of different street lamps according to the ambient light intensity near different street lamps in the data group to be analyzed and the geographical locations of different street lamps, so as to form a light intensity data set of each area on the road section;

[0061] S300, according to the light intensity data sets of each area, obtaining an abnormal light intensity data set, using a machine learning algorithm to analyze whether the abnormal light intensity causes visual obstruction to the car driver, and linearly adjusting the brightness of the LED light source of the corresponding street lamp to maintain the light intensity of each area within a normal range;

[0062] S400, according to the vehicle driving conditions and the existence of surrounding objects in the data group to be analyzed, when a sensor near a distant street lamp detects that a vehicle at a distance of A meters is driving in the direction of a location where there are surrounding objects, control the alarm-type LED light sources under all street lamp poles in the road section from the current position of the vehicle to the location where there are surrounding objects to flash;

[0063] S500 uses the SSD algorithm to identify objects when the sensor detects the presence of surrounding objects. When no cars, pedestrians or cyclists are detected around, the current to control the LED light source is sent to the street lamp to automatically dim the brightness of the LED light source of the street lamp in the road section.

[0064] Step S200 includes:

[0065] S201, obtain the geographical location information and number information of each street lamp, determine the boundary of the lighting range of each street lamp, and the boundary defines the interval area between adjacent street lamps; street lamp number: L1, L2, L3, ...; geographical location (longitude, latitude): (100.123, 30.456), (100.456, 30.789), (100.789, 30.123), ...; according to the geographical location information of each street lamp, the boundary of its lighting range can be determined. The boundary of the lighting range is determined by the distance calculation method of longitude and latitude, and the great circle distance between two points is calculated using the Vincenty formula. The radius of the lighting range of each street lamp is R, which is 50 meters. According to the determined lighting range boundary, the area between adjacent street lamps is divided into interval areas, and the size of the interval area is divided into grids.

[0066] S202, taking into account the lighting range of the street lamps, the mutual influence between the street lamps and the ambient light intensity, and using the light attenuation model to calculate the light intensity of the area between the street lamps;

[0067] S203, integrating the light intensity calculated for each street lamp interval area with the ambient light intensity near different street lamps to form a light intensity data set for each area of ​​the road section;

[0068] S204, verify the consistency of each data point in the light intensity data set to ensure that the format and unit of the data are unified and there are no errors or abnormal values; perform field measurements to verify whether the light intensity in the light intensity data set is consistent with the actual situation, and correct the data if it is inconsistent.

[0069] In one embodiment, according to step S202, the illumination attenuation model uses the inverse fourth law to describe the attenuation law of light in space, and the mathematical expression of the light intensity in a street lamp interval area is as follows:

[0070]

[0071] Where I is the illumination at the target point, is the initial illuminance of the LED light source, is the distance from the target point to the LED light source;

[0072] According to step S203, light intensity integration is performed as follows:

[0073] For a target point in the overlapping area of ​​several street lamp lighting ranges, the influence of several street lamps on the illumination of the target point is considered, and the illumination I of the target point of several street lamps is superimposed to obtain the total illumination of the target point; the influence of the ambient light intensity on the illumination of the target point is considered, and the ambient light intensity is added to the total illumination of the target point to obtain the final illumination of the target point;

[0074] For each target point in the street lamp interval area, the final illuminance is obtained according to the illuminance attenuation model and light intensity integration method. By traversing all target points, the illuminance distribution of each point in the area is calculated to form a light intensity data set for each area under the road section.

[0075] Street lamp lighting range radius R = 50 meters, street lamp spacing d = 100 meters, ambient light intensity lux, we select a street lamp interval area, which contains 3 target points, and their distances from the street lamps are rice, rice, Meters, we use the light attenuation model for calculation, the model formula is: , the initial illuminance of the street lamp lux, we first calculate the illuminance of each target point, then consider the influence of the ambient light intensity on it, and finally get the final illuminance of the target point.

[0076] For the first target point: lux, for the second target point: lux, for the third target point: lux;

[0077] Consider the ambient light intensity: ; then for the first target point: lux, for the second target point: lux, for the third target point: lux.

[0078] According to step S204, each data point in the light intensity data set is checked to ensure the uniformity of format and unit, wherein the checking includes checking whether the data is recorded in the same format and ensuring that all data use the same unit to represent the light intensity; checking whether there are abnormal values, wherein the abnormal values ​​include values ​​outside a reasonable range, null values, or inconsistent data points;

[0079] In the field, select the area between street lamps for field sampling to perform field measurement of light intensity. The area can cover different lighting conditions, including positions close to street lamps, intervals between street lamps, and positions far away from street lamps. The light intensity obtained by field measurement is compared with the light intensity at the corresponding position in the light intensity dataset. If the comparison results are inconsistent, data correction and adjustment are required, and steps S201-S203 are re-executed.

[0080] Step S300 includes:

[0081] S301, extracting an abnormal light intensity data set from the light intensity data sets of each region, wherein the abnormal light intensity refers to a data point that deviates significantly from the surrounding light intensity, indicating that the light is too bright or too dark;

[0082] S302, selecting a random forest algorithm to analyze whether the abnormal light intensity causes visual obstruction to the car driver, using the abnormal light intensity data set to train a machine learning model, using a test set to evaluate the performance of the model, and using the trained machine learning model to analyze the abnormal light intensity data;

[0083] S303. According to the analysis results of the machine learning model, linearly adjust the brightness of the street lamp LED light source in the area where the abnormal light intensity is located to maintain the light intensity in each area within the normal range.

[0084] According to step S302, a training data set is prepared from the abnormal light intensity data set, wherein the training data set includes input features and corresponding labels, wherein the input features are abnormal light intensity data, and the labels describe whether visual field obstruction is caused, and are used for the machine learning model to learn the relationship between the features and the labels; the prepared training data set is divided into a training set and a test set, wherein the training set is used to train the model, and the test set is used to evaluate the performance of the model;

[0085] The machine learning model uses a random forest algorithm and constructs several decision trees at the same time, each of which performs random sampling and feature selection on the data set; the trained random forest model is evaluated using test set data, and the evaluation indicators include accuracy, precision, recall rate and F1 value, and the model parameters are adjusted according to the evaluation results of the test set; the trained random forest model is used to analyze new abnormal light intensity data, and the model predicts whether the input abnormal light intensity data will cause visual obstruction to the car driver based on the input abnormal light intensity data.

[0086] Step S400 includes:

[0087] S401: When a sensor near a distant street lamp detects that a vehicle at a distance of A meters is traveling in a direction where there are objects around, it is determined that there is a potential danger;

[0088] S402, when a potential danger is detected, immediately controlling the warning LED light sources under all street lamp poles in the road section from the current position of the vehicle to the position of surrounding objects to flash;

[0089] S403: The warning LED light source continues to flash until the vehicle leaves the potential danger area or the potential danger situation is resolved.

[0090] In another embodiment, according to step S401, the method for calculating the distance A meters is as follows:

[0091] S401-1. When a vehicle is detected at the farthest safe distance from the position where there are surrounding objects, the current vehicle speed is obtained through the sensor group at the vehicle position. , the current vehicle model and the road surface slip coefficient c=0.2 are obtained through the convolutional neural network; the moving speed of the object is obtained through the sensor at the location where there is an object around ;According to the distance between the two sensor groups, the distance between the vehicle and the object is obtained as D=100m;

[0092] S401-2. According to the type of the current vehicle, obtain the conventional braking distance S=20m of the current vehicle from the reference braking distance table of the current vehicle model. The calculation formula of the predicted braking distance P is as follows: m;

[0093] S401-3. Considering the driver's reaction distance G = 10m, use the following mathematical formula to calculate the distance A of the potential danger situation:

[0094]

[0095] According to step S500, when the sensor detects the presence of an object in the surrounding area, the system uses the SSD algorithm to identify the object. SSD is a classic target detection algorithm that can simultaneously detect and classify targets in a single neural network, with high detection speed and accuracy. The SSD algorithm classifies the detected objects and identifies whether they include traffic participants such as cars, pedestrians, or cyclists. The identification of these categories is to determine whether there is a traffic situation, so as to decide whether to adjust the brightness of the LED light source of the street lamp.

[0096] If the sensor does not detect any traffic participants such as cars, pedestrians or cyclists around, the system will send instructions to the street lights based on this situation to control the current of the LED light source and automatically dim the brightness of the LED light source of the street lights in the road section. This can be achieved through an intelligent control system that adjusts the brightness level of the LED light source to save energy and reduce light pollution.

[0097] like Figure 2 As shown in the system structure diagram of an LED light source intelligent control system based on data analysis, an LED light source intelligent control system based on data analysis includes:

[0098] Sensor module: responsible for real-time monitoring of ambient light intensity, vehicle driving conditions and the presence of surrounding objects to form a data group to be analyzed; including: a light sensor unit, a vehicle detection sensor unit, an object detection sensor unit, a data transmission unit and a data processing unit, wherein the light sensor unit is used to monitor the ambient light intensity, the vehicle detection sensor unit is used to monitor the vehicle driving conditions, the object detection sensor unit is used to monitor the presence of surrounding objects, the data transmission unit is used to transmit the data collected by the sensor to the system, and the data processing unit is used to pre-process the sensor data to form a data group to be analyzed and add the geographical location and number of the street lamp;

[0099] Light intensity calculation module: according to the ambient light intensity near different street lamps in the data group to be analyzed and the geographical locations of different street lamps, the light intensity of the area between street lamps under the mutual influence of different street lamps is calculated to form a light intensity data set of each area under the road section; including: a light intensity calculation unit and a data integration unit, wherein the light intensity calculation unit is used to calculate the light intensity of each area of ​​the road section according to the sensor data and the street lamp location information, and the data integration unit is used to integrate the calculated light intensity data into a data set;

[0100] Abnormal light intensity analysis module: obtains abnormal light intensity data sets from the light intensity data sets of each area, uses machine learning algorithms to analyze whether the abnormal light intensity causes visual obstruction to the car driver, and makes corresponding light source brightness adjustments; includes: abnormal light intensity detection unit, machine learning algorithm unit and light source brightness adjustment unit, wherein the abnormal light intensity detection unit is used to detect whether the light intensity is abnormal, the machine learning algorithm unit is used to analyze the abnormal light intensity data and determine whether it causes visual obstruction to the driver, and the light source brightness adjustment unit is used to adjust the LED light source brightness of the street lamp according to the analysis results;

[0101] Alarm LED light source control module: when a remote sensor detects that the vehicle is traveling in the direction of a location where there are objects around, the alarm LED light sources under all street lamp poles in the road section from the current location of the vehicle to the location where there are objects around are controlled to flash; including: an alarm LED light source control unit, used to control the alarm LED light source to flash according to the location of the vehicle and the location of the surrounding objects;

[0102] Object recognition module: When the sensor detects the existence of surrounding objects, it recognizes the objects and adjusts the brightness of the LED light source of the street lamp according to the recognition results; it includes: an object recognition unit and a light source brightness control unit, wherein the object recognition unit uses the SSD algorithm to recognize the surrounding objects detected by the sensor, and the light source brightness control unit adjusts the brightness of the LED light source of the street lamp according to the recognition results.

[0103] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0104] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for intelligent control of LED light sources based on data analysis, characterized in that: The method comprises the following steps: S100, the street lamp sensor group installed on the highway is responsible for monitoring the ambient light intensity, vehicle driving conditions and the existence of surrounding objects, collecting data transmitted by the sensor group in real time to form a data group to be analyzed, and adding the geographical location and number of the street lamp to the data group to be analyzed; S200, calculating the light intensity of the area between street lamps under the mutual influence of different street lamps according to the light intensity of different street lamp environments and the geographical locations of different street lamps in the data group to be analyzed, so as to form a light intensity data set of each area under the road section; S300, according to the light intensity data sets of each area, obtaining an abnormal light intensity data set, using a machine learning algorithm to analyze whether the abnormal light intensity causes visual obstruction to the car driver, and linearly adjusting the brightness of the LED light source of the corresponding street lamp to maintain the light intensity of each area within a normal range; S400, according to the vehicle driving conditions and the existence of surrounding objects in the data group to be analyzed, when a sensor near a distant street lamp detects that a vehicle at a distance of A meters from the street lamp is driving in the direction of a location where an object exists around the street lamp, control the alarm-type LED light sources under all street lamp poles in the road section from the current position of the vehicle to the location where the object exists to flash, where A is a positive integer; S500. Using the SSD algorithm, when the sensor detects the presence of surrounding objects, the object is identified, where the object represents a car, a pedestrian or a cyclist. When no cars, pedestrians or cyclists are detected around, a current is sent to the street lamp to control the LED light source, and the brightness of the LED light source of the street lamp in the road section is automatically dimmed.

2. The method for intelligently controlling an LED light source based on data analysis according to claim 1, characterized in that: Step S200 includes: S201, obtaining geographic location information and number information of each street lamp, and determining the boundary of the lighting range of each street lamp, wherein the boundary defines the interval area between adjacent street lamps; S202, taking into account the lighting range of the street lamps, the mutual influence between the street lamps and the ambient light intensity, and using the light attenuation model to calculate the light intensity of the area between the street lamps; S203, integrating the light intensity calculated for each street lamp interval area with the light intensity of different street lamp environments to form a light intensity data set for each area of ​​the road section; S204, verify the consistency of each data point in the light intensity data set to ensure that the format and unit of the data are unified and there are no errors or abnormal values; perform field measurements to verify whether the light intensity in the light intensity data set is consistent with the actual situation, and correct the data if it is inconsistent.

3. The method for intelligently controlling an LED light source based on data analysis according to claim 2, characterized in that: According to step S202, the illumination attenuation model uses the inverse fourth law to describe the attenuation law of light in space, and the mathematical expression is as follows: Where I is the illumination at the target point, is the initial illuminance of the LED light source, and d is the distance from the target point to the LED light source; for each street lamp, the initial illuminance of the light source is determined according to the shape and size of its lighting range. ; For each target point within the street lamp interval area, calculate the distance d from the target point to the nearest street lamp.

4. The method for intelligently controlling an LED light source based on data analysis according to claim 2, characterized in that: According to step S203, light intensity integration is performed as follows: For a target point in the overlapping area of ​​several street lamp lighting ranges, the influence of several street lamps on the illumination of the target point is considered, and the illumination I of the target point of several street lamps is superimposed to obtain the total illumination of the target point; the influence of the ambient light intensity on the illumination of the target point is considered, and the ambient light intensity is added to the total illumination of the target point to obtain the final illumination of the target point; For each target point in the street lamp interval area, the final illuminance is obtained according to the illuminance attenuation model and light intensity integration method. By traversing all target points, the illuminance distribution of each point in the area is calculated to form a light intensity data set for each area under the road section.

5. The method for intelligently controlling an LED light source based on data analysis according to claim 2, characterized in that: According to step S204, each data point in the light intensity data set is checked to ensure uniformity of format and unit, wherein the checking includes checking whether the data is recorded in the same format and ensuring that all data use the same unit to represent light intensity; Checking for outliers, including values ​​outside a reasonable range, null values, or inconsistent data points; In the field, select the area between street lamps for field sampling to perform field measurement of light intensity. The area can cover different lighting conditions, including positions close to street lamps, intervals between street lamps, and positions far away from street lamps. The light intensity obtained by field measurement is compared with the light intensity at the corresponding position in the light intensity dataset. If the comparison results are inconsistent, data correction and adjustment are required, and steps S201-S203 are re-executed.

6. The method for intelligently controlling an LED light source based on data analysis according to claim 1, characterized in that: Step S300 includes: S301, extracting an abnormal light intensity data set from the light intensity data sets of each region, wherein the abnormal light intensity refers to a data point that deviates significantly from the surrounding light intensity, indicating that the light is too bright or too dark; S302, selecting a random forest algorithm to analyze whether the abnormal light intensity causes visual obstruction to the car driver, using the abnormal light intensity data set to train a machine learning model, using a test set to evaluate the performance of the model, and using the trained machine learning model to analyze the abnormal light intensity data; S303. According to the analysis results of the machine learning model, linearly adjust the brightness of the street lamp LED light source in the area where the abnormal light intensity is located to maintain the light intensity in each area within the normal range.

7. The method for intelligently controlling an LED light source based on data analysis according to claim 6, characterized in that: According to step S302, a training data set is prepared from the abnormal light intensity data set, wherein the training data set includes input features and corresponding labels, wherein the input features are abnormal light intensity data, and the labels describe whether visual field obstruction is caused, and are used for the machine learning model to learn the relationship between the features and the labels; the prepared training data set is divided into a training set and a test set, wherein the training set is used to train the model, and the test set is used to evaluate the performance of the model; The machine learning model uses a random forest algorithm to construct several decision trees at the same time, each of which performs random sampling and feature selection on the data set; Use the test set data to evaluate the trained random forest model. The evaluation indicators include accuracy, precision, recall and F1 value. Adjust the model parameters according to the evaluation results of the test set. The trained random forest model is used to analyze new abnormal light intensity data. The model predicts whether the input abnormal light intensity data will cause visual obstruction to the car driver based on the input abnormal light intensity data.

8. The method for intelligently controlling an LED light source based on data analysis according to claim 1, characterized in that: Step S400 includes: S401: When a sensor near a distant street lamp detects that a vehicle at a distance of A meters is traveling in a direction where there are objects around, it is determined that there is a potential danger; S402, when a potential danger is detected, immediately controlling the warning LED light sources under all street lamp poles in the road section from the current position of the vehicle to the position of surrounding objects to flash; S403: The warning LED light source continues to flash until the vehicle leaves the potential danger area or the potential danger situation is resolved.

9. The method for intelligently controlling an LED light source based on data analysis according to claim 8, characterized in that: According to step S401, the method for calculating the distance A meters is as follows: S401-1. When a vehicle is detected at the farthest safe distance from the position where there are surrounding objects, the current vehicle speed is obtained through the sensor group at the vehicle position. , the current vehicle model and the road surface slip coefficient c are obtained through the convolutional neural network; the moving speed of the object is obtained through the sensor at the location where there is an object around ;According to the distance between the two sensor groups, obtain the distance D between the vehicle and the object; S401-2. According to the type of the current vehicle, obtain the conventional braking distance S of the current vehicle from the reference braking distance table of the current vehicle model. The calculation formula for the predicted braking distance P is as follows: ; S401-3. Considering the driver's reaction distance G, use the following mathematical formula to calculate the distance A of the potential danger situation: in, is the relative time between the vehicle and the object.

10. An intelligent control system for LED light sources based on data analysis, using an intelligent control method for LED light sources based on data analysis according to any one of claims 1 to 9, characterized in that: include: Sensor module: responsible for real-time monitoring of ambient light intensity, vehicle driving conditions and the presence of surrounding objects to form a data group to be analyzed; including: a light sensor unit, a vehicle detection sensor unit, an object detection sensor unit, a data transmission unit and a data processing unit, wherein the light sensor unit is used to monitor the ambient light intensity, the vehicle detection sensor unit is used to monitor the vehicle driving conditions, the object detection sensor unit is used to monitor the presence of surrounding objects, the data transmission unit is used to transmit the data collected by the sensor to the system, and the data processing unit is used to pre-process the sensor data to form a data group to be analyzed and add the geographical location and number of the street lamp; Light intensity calculation module: according to the light intensity of different street lamp environments and the geographical locations of different street lamps in the data group to be analyzed, the light intensity of the street lamp interval area under the mutual influence of different street lamps is calculated to form a light intensity data set of each area under the road section; including: a light intensity calculation unit and a data integration unit, wherein the light intensity calculation unit is used to calculate the light intensity of each area of ​​the road section according to the sensor data and the street lamp location information, and the data integration unit is used to integrate the calculated light intensity data into a data set; Abnormal light intensity analysis module: obtains abnormal light intensity data sets from the light intensity data sets of each area, uses machine learning algorithms to analyze whether the abnormal light intensity causes visual obstruction to the car driver, and makes corresponding light source brightness adjustments; includes: abnormal light intensity detection unit, machine learning algorithm unit and light source brightness adjustment unit, wherein the abnormal light intensity detection unit is used to detect whether the light intensity is abnormal, the machine learning algorithm unit is used to analyze the abnormal light intensity data and determine whether it causes visual obstruction to the driver, and the light source brightness adjustment unit is used to adjust the LED light source brightness of the street lamp according to the analysis results; Alarm LED light source control module: when a remote sensor detects that the vehicle is traveling in the direction of a location where there are objects around, the alarm LED light sources under all street lamp poles in the road section from the current location of the vehicle to the location where there are objects around are controlled to flash; including: an alarm LED light source control unit, used to control the alarm LED light source to flash according to the location of the vehicle and the location of the surrounding objects; Object recognition module: When the sensor detects the existence of surrounding objects, it recognizes the objects and adjusts the brightness of the LED light source of the street lamp according to the recognition results; it includes: an object recognition unit and a light source brightness control unit, wherein the object recognition unit uses the SSD algorithm to recognize the surrounding objects detected by the sensor, and the light source brightness control unit adjusts the brightness of the LED light source of the street lamp according to the recognition results.

Images