Self-cleaning control method for road lighting lamp

Abstract

The invention discloses a road lighting lamp self-cleaning control method, and belongs to the technical field of road lighting lamp cleaning. The method comprises the following steps: S1, feature selection: collecting feature parameters such as PM2.5 and PM10 concentrations; s2, feature fusion: carrying out dust, rain and fog pollution identification and the like; s3, intelligent judgment: calculating various pollution risk values; and S4, in a response mode, the cleaning frequency and intensity are dynamically adjusted according to a three-stage mode. Through multi-source data sensing and an intelligent algorithm, automatic cleaning of the lamp is achieved, the cleaning efficiency is improved, safety and traffic hidden dangers caused by manual cleaning are avoided, it is ensured that the lighting effect of the lamp is stable, and the traffic accident risk is reduced.

Description

Technical Field

[0001] This invention belongs to the field of highway lighting fixture cleaning technology, and particularly relates to a self-cleaning control method for highway lighting fixtures. Background Technology

[0002] Highway lighting fixtures are exposed to complex environments for extended periods. Dust and exhaust fumes kicked up by passing vehicles, particulate matter in smoggy weather, and water films in high-humidity environments all adhere to the surface of the fixtures, leading to reduced lighting performance and affecting traffic safety. Current methods of cleaning these fixtures mainly rely on regular manual cleaning, which suffers from problems such as untimely cleaning, low efficiency, poor results, traffic disruption, and safety risks for cleaning personnel.

[0003] To address these issues, this invention proposes a system that combines a self-cleaning device on the lamp with an intelligent judgment algorithm to dynamically adjust the cleaning frequency and control the lamp to clean itself automatically, ensuring that the lamp is always in optimal lighting condition. Summary of the Invention

[0004] The purpose of this invention is to provide a self-cleaning control method for highway lighting fixtures to solve the problems mentioned in the background art.

[0005] In view of this, the present invention provides a self-cleaning control method for highway lighting fixtures, comprising the following steps:

[0006] S1. Feature selection: Collect characteristic parameters that affect the surface pollution of highway lighting fixtures, including PM2.5 concentration, PM10 concentration, special weather parameters, traffic flow parameters, and lighting fixture brightness parameters;

[0007] Among them, special weather parameters include wind speed and humidity corresponding to sandstorm and strong wind weather, and rainfall, visibility and humidity corresponding to rain and fog weather;

[0008] Traffic flow parameters include traffic volume and the proportion of large vehicles;

[0009] The luminaire brightness parameters include the actual luminaire brightness and the theoretical luminaire brightness;

[0010] S2. Feature Fusion: The feature parameters collected in step S1 are fused, including dust pollution identification, rain and fog pollution identification, composite pollution index calculation and lighting pollution degree quantification.

[0011] S3. Intelligent Judgment: Based on the feature fusion results of step S2, calculate the risk value of sand and dust pollution, the risk value of rain and fog pollution, the risk value of composite pollution, and the direct quantitative value of lamp pollution to determine the degree of lamp pollution.

[0012] S4. Response Mode: Based on the judgment result of step S3, a three-level cleaning response mode is adopted to dynamically adjust the cleaning frequency and cleaning intensity, and control the lamp self-cleaning device to perform cleaning operations.

[0013] In a further embodiment of the present invention, in step S1, PM2.5 is the mass concentration of particulate matter with a diameter of less than or equal to 2.5 micrometers in the air, and PM10 is the mass concentration of particulate matter with a diameter of less than or equal to 10 micrometers in the air.

[0014] Large vehicles include heavy-duty trucks, large buses, and engineering vehicles.

[0015] In a further embodiment of the present invention, the dust pollution identification in step S2 is based on PM10 concentration, wind speed, and humidity, and is determined to be dust pollution when the following conditions are met:

[0016]

[0017] Among them, 200 μg / m is the lower limit of PM10 concentration for light dust, wind speed > 5 m / s is a necessary condition for dust to rise, and humidity < 40% is a dry environment characteristic for dust to rise easily.

[0018] In a further embodiment of the present invention, the rain and fog pollution identification in step S2 is based on rainfall, visibility, and humidity, and is determined to be rain and fog pollution when the following conditions are met:

[0019]

[0020] Among them, rainwater has the ability to wash away pollutants when the rainfall is greater than 0.5 mm / h, visibility is less than 500m is a typical characteristic of foggy weather, and water film is easily formed when the humidity is greater than 80%, which aggravates the adhesion of pollutants.

[0021] In a further embodiment of this invention, the composite pollution index calculation in step S2 integrates PM2.5 concentration, PM10 concentration, traffic flow, and the proportion of large vehicles, using the following formula:

[0022]

[0023] in, K Q K R This is based on historical experience.

[0024] In a further embodiment of the present invention, the quantification of the lamp pollution level in step S2 includes calculating the theoretical brightness of the lamp:

[0025] L theory (t)=L initial ·e -λt λ = 0.015 / year

[0026] Among them, L theory Indicates the theoretical luminance value of the lamp; L initial λ represents the factory brightness value of the lamp; λ represents the aging rate of the lamp, with a statistical period of years; t represents the years the lamp has been used.

[0027] Calculate the luminance deviation rate (BAR):

[0028] Among them, L actual This indicates the actual brightness value of the lamp.

[0029] In a further embodiment of this invention, the dust pollution risk value in step S3 is calculated using the Sigmoid function based on PM10, wind speed, and humidity, with the following formula:

[0030]

[0031] The risk value is mapped to [0, 1].

[0032] In a further embodiment of this invention, the rain and fog pollution risk value in step S3 is calculated based on rainfall, visibility, and humidity, using the following formula:

[0033] In a further embodiment of this invention, the calculation of the composite pollution risk value in step S3 incorporates a dynamic adjustment factor, the formula of which is: w i =Softmax(f LSTM (PM 2.5 PM 10 Traffic flow, time)

[0034] threshold i (t) = threshold i (0)·(1+DAF(t)),

[0035]

[0036] Among them, w i The value represents the feature weight; α represents the adjustment coefficient; x i (t) represents the measurement value of the i-th sensor at time t; σ represents the average measurement value of the i-th sensor over a historical period; i This represents the standard deviation of the measurements taken by the i-th sensor over a historical period.

[0037] In a further embodiment of the present invention, the three-level cleaning response mode in step S4 includes a daily cleaning mode: when there is no special weather, the compound pollution index is ≤0.4 and BAR is ≤10%, cleaning is performed once every 24 hours, with cleaning liquid sprayed each time and the scraper reciprocating 5 times.

[0038] Enhanced cleaning mode: When the compound pollution index is 0.4-0.7, BAR is 10%-15%, or the dust / rain / fog risk value is 0.4-0.6, clean once every hour, spray cleaning liquid each time, and scraper blades move back and forth 5 times;

[0039] Where BAR is the luminance deviation rate, and its prediction formula is:

[0040]

[0041] L theory (t)=L0·e -λt +β·Cumulative cleaning times.

[0042] The beneficial effects of this invention are:

[0043] This invention achieves a breakthrough improvement by constructing a fully intelligent system. The system utilizes an edge computing terminal to access multi-source data from particulate matter concentration sensors, meteorological sensors, millimeter-wave radar sensors, and brightness sensors in real time. This data covers key indicators such as PM2.5 and PM10 concentrations, special weather parameters such as sandstorms, strong winds, and rain / fog, traffic flow (including the proportion of large vehicles), and the deviation between the actual and theoretical brightness of lighting fixtures. Based on this data, the system intelligently determines, through feature fusion and quantitative calculation, parameters that can be directly used for decision-making, such as sandstorm pollution risk values, rain / fog pollution risk values, composite pollution index, and brightness deviation rate (BAR). This dynamically matches a three-level cleaning response mode. For example, when the PM10 concentration exceeds 200 μg / m³, the wind speed is greater than 5 m / s, and the humidity is less than 40%, the system determines it as sandstorm pollution and triggers an emergency cleaning mode, starting cleaning every 15 minutes. When there is no special weather, the pollution index is ≤0.4, and the BAR is ≤10%, the system automatically switches to a daily cleaning mode, cleaning every 24 hours. This dynamic adjustment mechanism completely eliminates the blindness of manual cleaning, enabling precise matching of cleaning frequency with the speed of contamination. Cleaning efficiency is several times higher than traditional methods, while significantly reducing resource consumption caused by ineffective cleaning.

[0044] In terms of ensuring safety and reducing risks, this invention solves the inherent drawbacks of manual cleaning. Traditional manual cleaning requires high-altitude operations in dangerous sections of roads with heavy traffic and complex environments, such as highway tunnel entrances and exits, and interchanges. This not only exposes cleaning personnel to safety threats such as falls and vehicle collisions, but also easily causes traffic congestion and even accidents due to lane closures during operations. This invention, however, achieves fully unmanned operation of lamp cleaning through the collaboration of a self-cleaning device and a remote control system. The entire process requires no on-site personnel and does not interrupt traffic. This design completely eliminates the safety risks for cleaning personnel and avoids traffic disruptions caused by the operation, significantly improving the safety and smoothness of road traffic. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0047] In the description of this application, it should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. For ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items, and therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0048] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, the first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0049] It should be noted that in the description of this application, the directional terms such as "front, back, up, down, left, right", "horizontal, vertical, horizontal" and "top, bottom" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description. Unless otherwise stated, these directional terms do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the scope of protection of this application. The directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself.

[0050] It should be noted that, in this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0051] This embodiment provides a self-cleaning control method for highway lighting fixtures, including the following steps:

[0052] S1. Feature selection: Collect characteristic parameters that affect the surface pollution of highway lighting fixtures, including PM2.5 concentration, PM10 concentration, special weather parameters, traffic flow parameters, and lighting fixture brightness parameters;

[0053] Among them, special weather parameters include wind speed and humidity corresponding to sandstorm and strong wind weather, and rainfall, visibility and humidity corresponding to rain and fog weather;

[0054] Traffic flow parameters include traffic volume and the proportion of large vehicles;

[0055] The luminaire brightness parameters include the actual luminaire brightness and the theoretical luminaire brightness;

[0056] S2. Feature Fusion: The feature parameters collected in step S1 are fused, including dust pollution identification, rain and fog pollution identification, composite pollution index calculation and lighting pollution degree quantification.

[0057] S3. Intelligent Judgment: Based on the feature fusion results of step S2, calculate the risk value of sand and dust pollution, the risk value of rain and fog pollution, the risk value of composite pollution, and the direct quantitative value of lamp pollution to determine the degree of lamp pollution.

[0058] S4. Response Mode: Based on the judgment result of step S3, a three-level cleaning response mode is adopted to dynamically adjust the cleaning frequency and cleaning intensity, and control the lamp self-cleaning device to perform cleaning operations.

[0059] In a further embodiment of the present invention, in step S1, PM2.5 is the mass concentration of particulate matter with a diameter of less than or equal to 2.5 micrometers in the air, and PM10 is the mass concentration of particulate matter with a diameter of less than or equal to 10 micrometers in the air.

[0060] Large vehicles include heavy-duty trucks, large buses, and engineering vehicles.

[0061] In a further embodiment of the present invention, the dust pollution identification in step S2 is based on PM10 concentration, wind speed, and humidity, and is determined to be dust pollution when the following conditions are met:

[0062]

[0063] Among them, 200 μg / m is the lower limit of PM10 concentration for light dust, wind speed > 5 m / s is a necessary condition for dust to rise, and humidity < 40% is a dry environment characteristic for dust to rise easily.

[0064] In a further embodiment of the present invention, the rain and fog pollution identification in step S2 is based on rainfall, visibility, and humidity, and is determined to be rain and fog pollution when the following conditions are met:

[0065]

[0066] Among them, rainwater has the ability to wash away pollutants when the rainfall is greater than 0.5 mm / h, visibility is less than 500m is a typical characteristic of foggy weather, and water film is easily formed when the humidity is greater than 80%, which aggravates the adhesion of pollutants.

[0067] In step S2, the dust pollution identification process involves the system acquiring PM10 concentration in real time using a particulate matter concentration sensor. Combined with wind speed and humidity data collected by a meteorological sensor, dust pollution is identified when the PM10 concentration exceeds 200 μg / m3 (reaching the lower limit of light dust), the wind speed is greater than 5 m / s (meeting the dynamic conditions for dust generation), and the humidity is less than 40% (in a dry environment where dust is easily stirred up). Rain and fog pollution identification, on the other hand, involves collecting data using a rain sensor, a visibility detector, and a humidity sensor. Rain and fog pollution is identified when the rainfall exceeds 0.5 mm / h (rainwater has the ability to wash away dirt), or the visibility is less than 500 m (a typical characteristic of foggy weather) and the humidity is greater than 80% (easily forming a water film). All of the above identification processes are automatically completed on the control terminal through a preset logic algorithm, directly outputting the pollution identification results to support the formulation of subsequent cleaning strategies.

[0068] In a further embodiment of this invention, the composite pollution index calculation in step S2 integrates PM2.5 concentration, PM10 concentration, traffic flow, and the proportion of large vehicles, using the following formula:

[0069]

[0070] in, K Q K R This is based on historical experience.

[0071] In the calculation of the composite pollution index in step S2, the system first obtains the PM2.5 and PM10 concentrations through a particulate matter concentration sensor, obtains the traffic flow and the proportion of large vehicles through traffic monitoring equipment, and then introduces weighting coefficients k1, k2, and k3 set by historical experience values ​​(where k3 is adjusted based on the basic traffic flow weighting coefficient k3′ combined with the large vehicle proportion correction coefficient K_R, i.e., k3=k3′+0.1×K_R), and combines it with the traffic flow coefficient K_Q, and calculates according to the formula "Composite Pollution Index Calculation Formula", so as to comprehensively quantify the impact of PM2.5, PM10, traffic flow, and the proportion of large vehicles on lamp pollution. The calculation process is automatically executed by the algorithm module of the control terminal, and the output composite pollution index is used for subsequent cleaning mode determination.

[0072] In a further embodiment of the present invention, the quantification of the lamp pollution level in step S2 includes calculating the theoretical brightness of the lamp:

[0073] L theory (t)=L initial ·e -λt λ = 0.015 / year

[0074] Among them, L theory Indicates the theoretical luminance value of the lamp; L initial λ represents the factory brightness value of the lamp; λ represents the aging rate of the lamp, with a statistical period of years; t represents the years the lamp has been used.

[0075] Calculate the luminance deviation rate (BAR):

[0076] Among them, L actual This indicates the actual brightness value of the lamp.

[0077] In step S2, the system first retrieves the initial brightness value recorded when the lamp leaves the factory (i.e., the factory brightness value), combines it with the preset aging rate of the lamp (fixed at 0.015 / year) and the years of use of the lamp recorded by the device, and calculates the theoretical brightness value of the lamp according to the formula "theoretical brightness value of lamp = factory brightness value of lamp × e^(-aging rate × years of use)". At the same time, the system collects the current actual luminous intensity of the lamp in real time through the brightness sensor (i.e., the actual brightness value of the lamp), and then calculates the BAR value according to the formula "brightness deviation rate BAR = (theoretical brightness value of lamp - actual brightness value of lamp) / theoretical brightness value of lamp × 100%". This process is automatically completed by the control terminal. The BAR value is used to intuitively quantify the degree of brightness decay of the lamp surface caused by pollution, providing a basis for subsequent cleaning judgment.

[0078] In a further embodiment of this invention, the dust pollution risk value in step S3 is calculated using the Sigmoid function based on PM10, wind speed, and humidity, with the following formula:

[0079]

[0080] The risk value is mapped to [0, 1].

[0081] In step S3, the dust pollution risk value calculation first obtains real-time data on PM10 concentration, wind speed, and humidity using particulate matter concentration sensors, wind speed sensors, and humidity sensors, respectively. Based on this data, the system substitutes the data into a preset calculation formula (based on the Sigmoid function) for calculation. The Sigmoid function maps the calculation result to the interval [0, 1], thereby obtaining the dust pollution risk value. This calculation process is automatically executed by the control terminal, and the resulting risk value can intuitively reflect the degree of risk of light fixtures being polluted during dusty weather, providing data support for the selection of subsequent clean response modes.

[0082] In a further embodiment of this invention, the rain and fog pollution risk value in step S3 is calculated based on rainfall, visibility, and humidity, using the following formula: In step S3, the system calculates the rain and fog pollution risk value by collecting real-time data on rainfall, visibility, and humidity using a rain sensor, a visibility detector, and a humidity sensor, respectively. This data is then used to calculate the risk using a preset formula. This formula quantifies the pollution risk in rainy and foggy weather by processing visibility-related parameters. The resulting rain and fog pollution risk value directly reflects the degree of risk of light fixtures being polluted in rainy and foggy weather. The entire calculation process is automatically completed by the control terminal, providing data support for the subsequent determination of the clean response mode.

[0083] In a further embodiment of this invention, the calculation of the composite pollution risk value in step S3 incorporates a dynamic adjustment factor, the formula of which is: w i =Softmax(f LSTM (PM 2.5 PM 10 Traffic flow, time)

[0084] threshold i (t) = threshold i (0)·(1+DAF(t)),

[0085]

[0086] Among them, w i The value represents the feature weight; α represents the adjustment coefficient; x i(t) represents the measurement value of the i-th sensor at time t; σ represents the average measurement value of the i-th sensor over a historical period; i This represents the standard deviation of the measurements taken by the i-th sensor over a historical period.

[0087] In the calculation of the composite pollution risk value in step S3, the system first obtains real-time measured values ​​of relevant parameters such as PM2.5 concentration, PM10 concentration, and traffic flow through corresponding sensors (such as particulate matter concentration sensors, traffic flow monitoring equipment, etc.). (That is, the measured value x of the i-th sensor at time t) i (t)), and simultaneously retrieves measurement data from each sensor over a historical period to calculate the historical measurement average and historical measurement standard deviation; then, based on the preset feature weights (w i The basic risk value (DAF) is calculated by substituting the base risk value (w) and adjustment coefficient (α) into the dynamic adjustment factor formula. Then, combining the feature weight (w) and the eigenvector (X) composed of multi-source parameters, the composite pollution risk value is calculated using the core formula (Composite pollution risk value = w × X × (1 + DAF(t))). The entire process is automatically executed by the control terminal. The basic risk value is corrected using a dynamic adjustment factor to better reflect real-time environmental changes. The resulting composite pollution risk value accurately reflects the impact of current comprehensive pollution on lighting fixtures, providing a basis for the formulation of subsequent cleaning strategies.

[0088] In a further embodiment of the present invention, the three-level cleaning response mode in step S4 includes a daily cleaning mode: when there is no special weather, the compound pollution index is ≤0.4 and BAR is ≤10%, cleaning is performed once every 24 hours, with cleaning liquid sprayed each time and the scraper reciprocating 5 times.

[0089] Enhanced cleaning mode: When the compound pollution index is 0.4-0.7, BAR is 10%-15%, or the dust / rain / fog risk value is 0.4-0.6, clean once every hour, spray cleaning liquid each time, and scraper blades move back and forth 5 times;

[0090] Where BAR is the luminance deviation rate, and its prediction formula is:

[0091]

[0092] L theory (t)=L0·e -λt +β·Cumulative cleaning times.

[0093] In the Level 3 Clean Response mode of step S4, the system first obtains judgment parameters such as the composite pollution index, BAR value (brightness deviation rate), and dust / rain / fog risk value through preliminary calculations. The prediction of the BAR value is achieved through the ARIMA model, which calculates the predicted BAR value for future times based on historical BAR values, PM2.5 concentration, and time interval Δt, and combines this with preset conditions (BAR...). t+1 >15% or a combined pollution risk value >0.7 and BAR t+1 A cleaning decision is triggered when the pollution level exceeds 10%. When the following conditions are met: no special weather, compound pollution index ≤ 0.4 and BAR ≤ 10%, the system activates the daily cleaning mode, controlling the self-cleaning device to perform cleaning once every 24 hours. Each time, cleaning fluid is sprayed through the spray device, and the scraper blades reciprocate 5 times. When the following conditions are met: compound pollution index 0.4-0.7, BAR 10%-15%, or dust / rain / fog risk value 0.4-0.6, the system automatically switches to enhanced cleaning mode, adjusting the cleaning frequency to once per hour, similarly performing the spraying of cleaning fluid and the scraper blades reciprocating 5 times. The entire mode switching and cleaning execution process is automatically controlled by the control terminal, requiring no manual intervention, ensuring precise matching of cleaning operations to the pollution level.

[0094] This embodiment uses the operating logic of a highway tunnel entrance / exit system as an example (12:00-14:00 on a certain day);

[0095] The lighting system at the entrance and exit of a highway tunnel has been equipped with self-cleaning lamps, PM2.5 / PM10 sensors, weather sensors, traffic flow detectors, brightness sensors, and edge computing terminals. The system achieves automated cleaning control through a closed-loop logic of "data acquisition - analysis and calculation - strategy determination - execution feedback." The following details the operation process using actual environmental data from that period:

[0096] I. Data Acquisition and Input: Real-time Access to Multi-Source Parameters

[0097] The edge computing terminal synchronously collects environmental and equipment status data for the specified time period through a pre-set sensor network:

[0098] Particulate matter pollution data: Real-time monitoring by particulate matter concentration sensors yielded PM2.5 concentrations of 85 g / m3 and PM10 concentrations of 180 g / m3, reflecting the pollution level of suspended particulate matter in the air;

[0099] Meteorological data: Meteorological sensors recorded a wind speed of 6.5 m / s and humidity of 35%. Combined with on-site observations, the weather conditions were determined to be sandstorm with strong winds and no rainfall, providing a basis for the assessment of sandstorm pollution.

[0100] Traffic pollution data: The traffic flow detector counted 1200 vehicles per hour, of which large vehicles (heavy trucks, construction vehicles, etc.) accounted for 25%, which was used to quantify the pollution contribution of exhaust emissions and road dust to the lamps.

[0101] Lighting brightness data: The brightness sensor collects the actual luminous brightness of the lighting fixture as 800 lumens. At the same time, the edge computing terminal retrieves the factory brightness parameters and service life of the lighting fixture and calculates that the theoretical brightness should be 850 lumens (after deducting normal aging losses).

[0102] After all data is preprocessed (outliers removed, timestamps standardized), it is transmitted in real time to the algorithm module of the edge computing terminal to provide a basis for subsequent analysis.

[0103] II. Feature Fusion and Pollution Risk Calculation: Quantifying Pollution Levels

[0104] Edge computing terminals use preset algorithms to fuse and process collected data, transforming multi-dimensional parameters into pollution risk values ​​that can be directly used for decision-making.

[0105] Dust pollution risk value calculation: Combining PM10 concentration (180g / m³), wind speed (6.5m / s), and humidity (35%), the values ​​are substituted into the Sigmoid function formula. Although PM10 did not reach the light dust threshold of 200g / m³, the wind speed was >5m / s (meeting the dynamic conditions for dust generation) and the humidity was <40% (dry environments are prone to dust generation). Considering the dusty and windy weather conditions, the calculated dust pollution risk value is approximately 1.0 (mapped to the [0,1] interval, with values ​​close to 1 indicating extremely high risk).

[0106] The composite pollution index and risk value are calculated by combining PM2.5, PM10 concentrations and traffic flow parameters according to the corresponding formulas mentioned above. Large vehicles account for 25%, and the traffic flow weight is increased by the correction coefficient K_R to obtain the composite pollution index. Then, the dynamic adjustment factor DAF (based on the historical data mean and standard deviation correction) is introduced to finally obtain a composite pollution risk value of approximately 0.8.

[0107] Lighting pollution can be directly quantified: Based on the brightness deviation rate formula, the calculated BAR = (850-800) / 850×100%≈5.88%, which directly reflects the degree of brightness decay caused by surface pollution in current lighting fixtures.

[0108] III. Cleaning Strategy Determination: Matching the Three-Level Response Mode

[0109] The edge computing terminal compares the above risk values ​​with preset thresholds to determine the cleaning mode:

[0110] According to the three-level response mode rules, the dust pollution risk value is 1.0 > 0.8, and the compound pollution risk value is 0.8 > 0.7. Although the BAR value is 5.88% < 10%, it meets the emergency cleaning mode triggering conditions of "dust pollution risk value > 0.8" or "compound pollution risk value > 0.7". Therefore, it was finally determined to execute the emergency cleaning mode.

[0111] IV. Cleaning Execution: Automated High-Frequency Cleaning Operations

[0112] The edge computing terminal sends an execution command to the microcontroller of the lamp self-cleaning device to start the cleaning process:

[0113] The microcontroller controls the spray pump through a PWM voltage regulation circuit to spray cleaning liquid (a special formula for sand and dust, which can soften the adhesion between sand and dust particles) according to a preset dosage; at the same time, it drives a high-precision stepper motor, which drives the rubber scraper to reciprocate five times along a set trajectory through a four-bar linkage mechanism to thoroughly scrape away the sand and dust and cleaning liquid residue on the surface of the lamp. Each cleaning takes about 30 seconds.

[0114] The system is set to emergency cleaning mode, automatically repeating the above cleaning operation every 15 minutes to ensure that sand and dust do not accumulate on the surface of the lamps. During this period, the edge computing terminal continuously monitors environmental data. When the sand and dust pollution risk value drops below 0.8 and the composite pollution risk value is ≤0.7, it will automatically switch to enhanced cleaning mode (once every 1 hour) until the pollution risk is further reduced, at which point it will be adjusted to daily cleaning mode, achieving the goal of "cleaning on demand and dynamic adaptation".

[0115] This example demonstrates that the invention can accurately identify pollution risks during sandstorms and strong winds, and effectively maintain the lighting effect of lamps through high-frequency cleaning operations, avoiding the lag and safety hazards of manual cleaning. This fully reflects the core advantages of the invention: "intelligent, automated, and precise".

[0116] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A self-cleaning control method for highway lighting fixtures, characterized in that, Includes the following steps: S1. Feature Selection: Collect characteristic parameters affecting surface pollution of highway lighting fixtures, including PM2.5 concentration, PM10 concentration, special weather parameters, traffic flow parameters, and fixture brightness parameters. Among them, special weather parameters include wind speed and humidity corresponding to sandstorms and strong winds, and rainfall, visibility, and humidity corresponding to rain and fog: Traffic flow parameters include traffic volume and the proportion of large vehicles: The luminaire brightness parameters include the actual luminaire brightness and the theoretical luminaire brightness; S2. Feature Fusion: The feature parameters collected in step S1 are fused, including dust pollution identification, rain and fog pollution identification, composite pollution index calculation and lighting pollution degree quantification. S3. Intelligent Judgment: Based on the feature fusion results of step S2, calculate the risk value of sand and dust pollution, the risk value of rain and fog pollution, the risk value of composite pollution, and the direct quantitative value of lamp pollution to determine the degree of lamp pollution. S4. Response Mode: Based on the judgment result of step S3, a three-level cleaning response mode is adopted to dynamically adjust the cleaning frequency and cleaning intensity, and control the lamp self-cleaning device to perform cleaning operations.

2. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, In step S1, PM2.5 is the mass concentration of particulate matter with a diameter of 2.5 micrometers or less in the air, and PM10 is the mass concentration of particulate matter with a diameter of 10 micrometers or less in the air. Large vehicles include heavy-duty trucks, large buses, and engineering vehicles.

3. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, The dust pollution identification in step S2 is based on PM10 concentration, wind speed, and humidity. Dust pollution is determined when the following conditions are met: Among them, 200 μg / m is the lower limit of PM10 concentration for light dust, wind speed > 5 m / s is a necessary condition for dust to rise, and humidity < 40% is a dry environment characteristic for dust to rise easily.

4. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, The rain and fog pollution identification in step S2 is based on rainfall, visibility, and humidity. Rain and fog pollution is determined when the following conditions are met: Among them, rainwater has the ability to wash away pollutants when the rainfall is greater than 0.5 mm / h, visibility is less than 500 m is a typical characteristic of foggy weather, and water film is easily formed when the humidity is greater than 80%, which aggravates the adhesion of pollutants.

5. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, The composite pollution index calculation in step S2 combines PM2.5 concentration, PM10 concentration, traffic flow, and the proportion of large vehicles. The formula is as follows: in, K Q K R This is based on historical experience.

6. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, The quantification of lamp pollution levels in step S2 includes calculating the theoretical luminance of the lamps: L theory (t)=L initial ·e -λt =0.015 / year, Among them, L theory Indicates the theoretical luminance value of the lamp; L initial λ represents the factory brightness value of the lamp; λ represents the aging rate of the lamp, with a statistical period of years; t represents the years the lamp has been used. Calculate the luminance deviation rate (BAR): Among them, L actual This indicates the actual brightness value of the lamp.

7. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, The dust pollution risk value in step S3 is calculated using the Sigmoid function based on PM10, wind speed, and humidity. The formula is as follows: The risk value is mapped to [0, 1].

8. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, The rain and fog pollution risk value in step S3 is calculated based on rainfall, visibility, and humidity, using the following formula:

9. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, Step S3 introduces a dynamic adjustment factor in the calculation of the composite pollution risk value. The formula is as follows: w i =Softmax(f LSTM (PM 2.5 PM 10 Traffic flow, time) threshold i (t) = threshold i (0)·(1+DAF(t)), Among them, w i The value represents the feature weight; α represents the adjustment coefficient; x i (t) represents the measurement value of the i-th sensor at time t; σ represents the average measurement value of the i-th sensor over a historical period; i This represents the standard deviation of the measurements taken by the i-th sensor over a historical period.

10. The self-cleaning control method for highway lighting fixtures according to claim 1, characterized in that, The three-level cleaning response mode in step S4 includes the daily cleaning mode: when there is no special weather, the compound pollution index is ≤0.4 and the BAR is ≤10%, cleaning is carried out once every 24 hours, with cleaning liquid sprayed each time and the scraper blade moving back and forth 5 times. Enhanced cleaning mode: When the compound pollution index is 0.4-0.7, BAR is 10%-15%, or the dust / rain / fog risk value is 0.4-0.6, clean once every hour, spray cleaning liquid each time, and scraper blades move back and forth 5 times; Where BAR is the luminance deviation rate, and its prediction formula is: L theory (t)=L0·e -λt +β·Cumulative cleaning times.

Images