Tunnel intelligent dimming method, system and computer storage medium

Abstract

This invention discloses a tunnel intelligent dimming method, system, and computer storage medium based on the isolated forest and 3Sigma principles, relating to the field of tunnel lighting control technology. The method includes the following steps: S1, collecting historical tunnel data and extracting historical feature data from it; S2, constructing a tunnel light intensity anomaly determination model Y using the isolated forest algorithm based on the historical feature data; S3, using the tunnel light intensity anomaly determination model Y to remove abnormal data from the historical feature data and calculating dimming baseline values ​​for different features using the 3Sigma principle; S4, collecting real-time feature data within the tunnel, using the light intensity anomaly determination model Y to judge the real-time dataset X, and outputting an intelligent dimming level based on the judgment result. It can determine whether historical and real-time light intensity data are abnormal and provide the correct dimming level. After using intelligent dimming, the average energy saving rate of the overall power consumption reaches 48.39%, indicating that the proposed model is of great significance for reducing tunnel operating costs.

Description

Technical Field

[0001] This invention relates to the field of tunnel lighting control technology, specifically to a tunnel intelligent dimming method, system, and computer storage medium based on the isolated forest and 3Sigma principles. Background Technology

[0002] In the field of tunnel lighting control, due to the "black hole effect" and "white hole effect" when vehicles enter and exit tunnels, traffic accidents caused by poor tunnel lighting account for more than half of all tunnel accidents each year. Therefore, the quality of tunnel lighting directly determines driving safety. Scientific and reasonable adjustment of tunnel lighting can not only create a good tunnel driving environment and reduce accidents, but also reduce tunnel energy consumption and operating costs. Currently, tunnel lighting faces the following challenges:

[0003] (1) Most tunnels only adjust the lighting according to the time of day. For example, the lighting is turned on to the maximum brightness during the day and only the basic lighting is turned on at night. This not only wastes energy, but may also cause the black hole effect and cause traffic accidents.

[0004] (2) The utilization rate of tunnel data is low. For example, the light intensity data outside the tunnel is basically idle and cannot generate corresponding value.

[0005] (3) The light intensity value measured by the equipment is easily affected by various factors, resulting in inaccurate light intensity data, and some data may even differ greatly from the actual value. Summary of the Invention

[0006] This invention provides a tunnel smart dimming method, system, and computer storage medium based on the isolated forest and 3Sigma principles. It can maximize the use of tunnel data resources, judge abnormal light intensity data, and output a smart dimming level based on the judgment result.

[0007] To achieve the above-mentioned technical objectives, the present invention mainly adopts the following technical solutions:

[0008] Firstly, this application provides a tunnel smart dimming method based on isolated forests and the 3Sigma principle, comprising the following steps:

[0009] S1 collects historical tunnel data and extracts historical feature data from the historical data;

[0010] S2 uses historical feature data and the isolated forest algorithm to construct a tunnel light intensity anomaly determination model Y;

[0011] S3 uses the tunnel light intensity anomaly judgment model Y to remove abnormal data from historical feature data and uses the 3Sigma principle to calculate the dimming reference value for different features.

[0012] S4 collects real-time feature data inside the tunnel, uses the light intensity anomaly judgment model Y to judge the real-time dataset X, and outputs the intelligent dimming level based on the judgment result.

[0013] In some embodiments, in step S1, the feature data includes light intensity, season, time of day, and climate feature data.

[0014] In some embodiments, step S2, the construction of the tunnel light intensity anomaly determination model Y includes the following steps:

[0015] Collect key data on indicators affecting light intensity;

[0016] The key indicator data are cleaned and preprocessed sequentially to extract feature data;

[0017] A light intensity anomaly detection model Y was constructed using the IFOrest algorithm.

[0018] In some embodiments, step S3, using the tunnel light intensity anomaly determination model Y to remove abnormal data from historical feature data, includes the following steps:

[0019] Using the light intensity anomaly detection model Y, the historical feature dataset Z0 is judged, the abnormal data that needs to be removed is filtered out, and the cleaned historical feature dataset Z1 is retained.

[0020] In some embodiments, the calculation of the dimming reference value in step S3 includes the following steps:

[0021] Using the historical feature dataset Z1, the dimming baseline values ​​for different features are calculated using the 3sigma principle and saved as dataset B. The specific format of dataset B is shown below:

[0022] B = {s} i ,h j ,w k ,b n}

[0023] in:

[0024] s i h represents the first feature. j Indicates the second characteristic, w k The third feature is represented by the range of values ​​for i, j, and k, which are determined based on the attributes or classifications of the first, second, and third features, respectively.

[0025] b n This represents the dimming reference value, where n ranges from 1 to i*j*k, representing dimming reference values ​​for different characteristics.

[0026] Preferably, the specific format of the dataset B is as follows:

[0027] B = {s} i ,h j ,w k ,b n}

[0028] in:

[0029] s i Indicates seasonal characteristics, with i ranging from 1 to 4, representing the four quarters of a year;

[0030] h j This represents a point-in-time characteristic, with j ranging from 0 to 23, representing 24 hours in a day;

[0031] w k It represents climate characteristics, with k ranging from 1 to k, representing k types of climate, such as sunny, rainy, cloudy, foggy, etc.

[0032] b n This represents the dimming reference value, where n ranges from 1 to i*j*k, representing the dimming reference value for different seasons, times, and climates.

[0033] In some embodiments, step S4, determining the real-time dataset includes the following steps:

[0034] Using the light intensity anomaly determination model Y, the real-time dataset X is determined. If the real-time dataset X is determined to be abnormal, the light intensity data in the dimming reference value dataset B is called; if the real-time dataset X is determined to be normal, the light intensity data of the real-time dataset X is retained, and the normal light intensity data returned in this step is recorded as Lo.

[0035] In some embodiments, the determination of the intelligent dimming level in step S4 includes the following steps:

[0036] The normal light intensity data (Lo) is compared with the execution level to determine the intelligent dimming level.

[0037] Secondly, this application discloses a tunnel smart dimming system based on isolated forests and the 3Sigma principle, the system comprising:

[0038] Data acquisition module: used to collect historical and real-time feature data within the tunnel;

[0039] Data analysis module: Analyzes historical feature data collected by the data acquisition module, and constructs a tunnel light intensity anomaly judgment model based on the analysis results;

[0040] Calculation module: Calculates the dimming reference values ​​for different features within the tunnel based on the tunnel light intensity anomaly determination model;

[0041] Light source control module: Based on the real-time data collected by the data acquisition module, the light source in the tunnel is controlled to ensure that the illuminance in the tunnel meets the execution level specified in the "Detailed Rules for Lighting Design of Highway Tunnels" (JTG / T D70 / 2-01-2014).

[0042] Thirdly, this application discloses a computer storage medium including a stored program, wherein the program, when running, controls the device where the storage medium is located to execute the method described in the first aspect.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The tunnel smart dimming method based on isolated forest and 3Sigma principles provided by this invention can determine whether historical and real-time light intensity data are abnormal by extracting features such as historical light intensity data, seasonal data, climate data, and time data from tunnel data resources, and give the correct dimming level. After using smart dimming, the average energy saving rate of the overall power consumption reaches 48.39%, indicating that the proposed model is of great significance for reducing tunnel operating costs. Attached Figure Description

[0045] Figure 1 This is a flowchart of the tunnel smart dimming method based on the isolated forest and 3Sigma principle proposed in this application.

[0046] Figure 2 The flowchart for constructing the light intensity anomaly determination model in this application is shown below;

[0047] Figure 3 This is a schematic diagram showing the reference values ​​of light intensity at different times under clear weather conditions and their implementation levels in this application.

[0048] Figure 4 This is a comparison chart of electricity consumption before and after the use of intelligent dimming in each tunnel in this application. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0050] Example 1

[0051] The tunnel smart dimming method based on isolated forests and the 3Sigma principle provided in this embodiment, such as Figure 1 As shown, the specific steps include the following:

[0052] (1) Collect historical data of the tunnel and extract historical feature data from the historical data.

[0053] Features such as historical light intensity data, seasonal data, climate data, and point-in-time data can be extracted from tunnel data resources. The extraction time should include at least one year of historical data, and if circumstances and funding permit, more than several years of historical data should be extracted to make the extracted historical feature data more accurate.

[0054] (2) Using historical feature data, the isolated forest algorithm is used to construct the tunnel light intensity anomaly determination model Y.

[0055] Isolation Forest (IForest) is a fast outlier detection method based on ensemble algorithms, characterized by low computational cost, linear time complexity, and high accuracy. To improve the accuracy of external light intensity data, key indicators affecting light intensity, such as light intensity, time of day, season, and climate, are first collected. Then, after data cleaning, preprocessing, and feature extraction, the IForest algorithm is used to construct a light intensity anomaly detection model Y. The specific process is as follows: Figure 2 As shown.

[0056] (3) The abnormal data in the historical feature data were removed by using the tunnel light intensity anomaly judgment model Y, and the dimming reference value of different features was calculated by using the 3Sigma principle.

[0057] First, the historical feature data collected in step (1) is collected, including historical data such as light intensity, time, season, and climate, to obtain historical feature dataset Z0. Then, the tunnel light intensity anomaly judgment model Y is used to judge the historical feature dataset Z0, filter out the abnormal data that needs to be removed, and retain the cleaned historical feature dataset Z1.

[0058] Using the historical feature dataset Z1, the lighting baseline values ​​for different seasons, times, and climates are calculated using the 3sigma principle and saved as dataset B. The specific format of dataset B is shown below:

[0059] B = {s} i ,h j ,w k ,b n}

[0060] in:

[0061] s i Indicates seasonal characteristics, with i ranging from 1 to 4, representing the four quarters of a year;

[0062] h jThis represents a point-in-time characteristic, with j ranging from 0 to 23, representing 24 hours in a day;

[0063] w k It represents climate characteristics, with k ranging from 1 to k, representing k types of climate, such as sunny, rainy, cloudy, foggy, etc.

[0064] b n This represents the dimming reference value, where n ranges from 1 to i*j*k, representing the dimming reference value for different seasons, times, and climates.

[0065] Among them, the 3sigma principle is also known as the Raida criterion. This method assumes that the dataset follows a normal or approximately normal distribution, and then calculates the mean μ and standard deviation σ of the dataset. The probability that the dataset is outside the interval (μ-3σ, μ+3σ) is less than 0.3%, so data values ​​outside the interval can be defined as critical outliers.

[0066] To ensure safe traffic flow in the tunnel while reducing energy consumption, the mean and variance of dataset B for different seasons, times, and climates were calculated, and μ+3σ was taken as b. n The baseline value.

[0067] (4) Collect real-time feature data in the tunnel, use the light intensity anomaly judgment model Y to judge the real-time feature dataset X, and output the intelligent dimming level according to the judgment result.

[0068] The light intensity anomaly detection model Y is used to determine the real-time feature dataset X. If the real-time feature dataset X is determined to be abnormal, the light intensity data in the dimming reference value dataset B is retrieved; if the real-time feature dataset X is determined to be normal, the light intensity data of the real-time dataset is retained. The normal light intensity data returned in this step is denoted as Lo.

[0069] According to the "Detailed Specifications for Lighting Design of Highway Tunnels" (JTG / T D70 / 2-01-2014), the standards for tunnel dimming within different light intensity ranges are shown in Table 1.

[0070] Table 1. Standards for Tunnel Lighting in Different Light Intensities

[0071]

[0072] Based on the normal light intensity data Lo returned in the previous step, calculate the required dimming level for the tunnel, denoted as Level.

[0073] The aforementioned feature data may also include other features, such as the air environment inside the tunnel and dust dispersion. Generally, if there is a lot of dust inside the tunnel, it will adhere to the lights, resulting in poor lighting conditions inside the tunnel. Feature parameters should be selected based on the actual situation. For areas with high pollution, dust dispersion parameters can be selected; for areas with better environments, this can be omitted to simplify the process.

[0074] Application examples

[0075] Data from seven tunnels on the Huiqing Expressway in Guangdong Province, collected between April 15th and April 22nd, 2021, was used for experimental testing. Data was collected at a 1-minute granularity, during a sunny weather period. Key data collected included road segment name, tunnel name, time, weather type, and light intensity. The data from the seven tunnels will be analyzed based on the tunnel light intensity assessment model from the previous section. First, abnormal light intensity data will be identified. Then, a baseline value for that time period will be calculated to determine the appropriate dimming level, and the energy savings after implementing intelligent dimming will be compared.

[0076] (1) Removal of abnormal light intensity values ​​and determination of benchmark values

[0077] By using Python toolkits to model the key indicator data of the seven tunnels and setting the two parameters nestimators and contamination, the abnormal light intensity results shown in Table 2 can be obtained.

[0078] Table 2 Results of Abnormal Light Intensity in Tunnels

[0079]

[0080] Abnormal light intensity data in Table 2 were removed, and the baseline light intensity values ​​for different time points in the tunnel were calculated using the 3 sigma principle (where 22:00-6:00 is nighttime, and only one baseline value is generated). The specific results are as follows: Figure 3 As shown.

[0081] from Figure 3 As can be seen, the light intensity level in the tunnel is 5 from 7 PM to 6 AM the next morning, meaning the enhanced lighting is completely off. Additionally, the light intensity level is highest from 11 PM to 2 PM, requiring the enhanced lighting to be turned on at its maximum. Overall, the variation in the baseline light intensity throughout the day follows a normal distribution, consistent with actual sunlight conditions on a clear day.

[0082] (2) Comparative analysis of results before and after intelligent dimming

[0083] The tunnel data from April 9, 2021 (without smart dimming) and April 19, 2021 (with smart dimming) were primarily used for comparative analysis. Specific results are as follows: Figure 4 As shown.

[0084] from Figure 4 As can be seen, the Chiling Tunnel achieved the largest energy saving rate after implementing intelligent dimming, reaching 67.51%. Furthermore, for the seven tunnels along the Huiqing Expressway, the average energy saving rate after implementing intelligent dimming reached 48.39%. This demonstrates that the proposed model is of significant importance in reducing tunnel operating costs.

[0085] Example 2

[0086] The tunnel smart dimming system based on isolated forests and the 3Sigma principle provided in this embodiment specifically includes:

[0087] Data acquisition module: used to collect historical and real-time feature data within the tunnel; including historical light intensity data, seasonal data, climate data, point-in-time data, etc. The collected historical feature data is aggregated to obtain a historical feature dataset; the collected real-time feature data is aggregated to obtain a real-time feature dataset.

[0088] Data Analysis Module: Analyzes historical feature data collected by the data acquisition module and constructs a tunnel light intensity anomaly judgment model based on the analysis results.

[0089] In this embodiment, the data analysis module is mainly used to perform data cleaning, data preprocessing, and feature extraction on historical feature data, and to construct a tunnel light intensity anomaly determination model based on the obtained data.

[0090] Calculation module: Calculates the dimming reference value for different features in the tunnel based on the tunnel light intensity anomaly judgment model.

[0091] In this embodiment, the calculation module mainly uses the tunnel light intensity anomaly judgment model Y to judge the historical feature dataset Z0, filter out the abnormal data that needs to be removed, and retain the cleaned historical feature dataset Z1. Then, using the historical feature dataset Z1, the dimming reference value for different seasons, different times and different climates is calculated using the 3sigma principle and retained as dataset B.

[0092] Finally, the light intensity anomaly detection model Y is used to determine the real-time feature dataset X collected by the data acquisition module. If the real-time feature dataset X is determined to be abnormal, the light intensity data in the dimming reference value dataset B is retrieved; if the real-time feature dataset X is determined to be normal, the light intensity data of the real-time dataset is retained. The normal light intensity data returned in this step is denoted as Lo.

[0093] Light source control module: Based on the real-time data collected by the data acquisition module, the light source in the tunnel is controlled to ensure that the illuminance in the tunnel meets the execution level specified in the "Detailed Rules for Lighting Design of Highway Tunnels" (JTG / T D70 / 2-01-2014).

[0094] In this embodiment, the light source control module mainly adjusts the light source based on the feedback results of the above steps, so that the illuminance in the tunnel reaches the execution level specified in the "Detailed Rules for Lighting Design of Highway Tunnels" (JTG / T D70 / 2-01-2014).

[0095] Example 3

[0096] The present invention also provides a computer-readable storage medium comprising a stored program, wherein, when the program is executed, it controls the device on which the storage medium is located to perform the method described in any of the preceding claims.

[0097] The computer storage medium disclosed in the above embodiments of the present invention can intelligently adjust the lighting in the tunnel, so as to maintain the lighting intensity in the tunnel at a stable and satisfactory level, thereby achieving the purpose of intelligent control of lighting and saving power resources.

[0098] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0099] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0100] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

Claims

1. A tunnel smart dimming method based on isolated forests and the 3Sigma principle, characterized in that, Includes the following steps: S1 collects historical tunnel data and extracts historical feature data from the historical data; The historical feature data includes light intensity, season, time of day, and climate feature data; S2 uses historical feature data and the isolated forest algorithm to construct a tunnel light intensity anomaly determination model Y; S3 uses the tunnel light intensity anomaly judgment model Y to remove abnormal data from historical feature data and uses the 3Sigma principle to calculate the dimming reference value for different features. The calculation of the dimming reference value includes the following steps: Using the historical feature dataset Z1 after removing the outlier data, the dimming baseline values ​​for different features are calculated using the 3sigma principle and retained as dataset B. The specific format of dataset B is as follows: B={s i ,h j, w k, b n } in: s i H represents the first feature. j Indicates the second characteristic, w k The third feature is represented by the range of values ​​for i, j, and k, which are determined based on the attributes or classifications of the first, second, and third features, respectively. b n This represents the dimming reference value, where n ranges from 1 to i*j*k, representing dimming reference values ​​for different characteristics. S4 collects real-time feature data inside the tunnel, uses the light intensity anomaly judgment model Y to judge the real-time dataset X, and outputs the intelligent dimming level based on the judgment result. The determination of real-time dataset X includes the following steps: Using the light intensity anomaly detection model Y, the real-time dataset X is evaluated. If the real-time dataset X is determined to be abnormal, the light intensity data from the dimming reference dataset B is retrieved; if the real-time dataset X is determined to be normal, the light intensity data of the real-time dataset X is retained, and the normal light intensity data returned in this step is recorded as L. o ; Determining the intelligent dimming level involves the following steps: (The text then abruptly shifts to a different topic) ...normal light intensity data L... o The smart dimming level is determined by comparing it with the execution level.

2. The tunnel smart dimming method based on isolated forest and 3Sigma principle according to claim 1, characterized in that, In step S2, the construction of the tunnel light intensity anomaly determination model Y includes the following steps: Collect key data on indicators affecting light intensity; The key indicator data are cleaned and preprocessed sequentially to extract feature data; A light intensity anomaly detection model Y was constructed using the IFOrest algorithm.

3. The tunnel smart dimming method based on isolated forest and 3Sigma principle according to claim 1, characterized in that, In step S3, the abnormal data in the historical feature data is removed using the tunnel light intensity anomaly determination model Y, including the following steps: Using the light intensity anomaly detection model Y, the historical feature dataset Z0 is judged, the abnormal data that needs to be removed is filtered out, and the cleaned historical feature dataset Z1 is retained.

4. The tunnel smart dimming method based on isolated forest and 3Sigma principle according to claim 1, characterized in that, The specific format of dataset B is as follows: B={s i ,h j, w k, b n } in: s i Indicates seasonal characteristics, with i ranging from 1 to 4, representing the four quarters of a year; h j This represents a point-in-time characteristic, with j ranging from 0 to 23, representing 24 hours in a day; w k It represents climate characteristics, with k ranging from 1 to k, representing k types of climate; b n This represents the dimming reference value, where n ranges from 1 to i*j*k, representing the dimming reference value for different seasons, times, and climates.

5. A tunnel smart dimming system based on isolated forests and the 3Sigma principle, employing the method described in any one of claims 1-4, characterized in that, The system includes: Data acquisition module: used to collect historical and real-time feature data within the tunnel; Data analysis module: Analyzes historical feature data collected by the data acquisition module, and constructs a tunnel light intensity anomaly judgment model based on the analysis results; Calculation module: Calculates the dimming reference values ​​for different features within the tunnel based on the tunnel light intensity anomaly determination model; Light source control module: Based on the real-time data collected by the data acquisition module, the light source in the tunnel is controlled to ensure that the illuminance in the tunnel meets the execution level specified in the "Detailed Rules for Lighting Design of Highway Tunnels" (JTG / T D70 / 2-01-2014).

6. A computer storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the storage medium is located to perform the method as described in any one of claims 1-4.

Images