A method and system for generating illuminance adjustment strategies for railway passenger station hubs

Abstract

This invention provides a method and system for generating illuminance adjustment strategies for railway passenger station hubs. The method includes: determining correction coefficients for various information types in environmental and passenger flow indicators for each functional area based on environmental and passenger flow parameters; determining standard illuminance values ​​for each functional area based on correction coefficients related to the overall illuminance of the station and lighting design standards for each functional area; inputting environmental, passenger flow, and monitoring image data into an illuminance prediction model to obtain predicted illuminance values ​​for each functional area; fusing the standard illuminance values ​​and predicted illuminance values ​​to obtain a target illuminance fusion value; and for each functional area, obtaining an adjustment strategy for lighting equipment based on the target illuminance fusion value, the natural illuminance of the area, and the correction coefficients corresponding to various information types. This invention enables precise control of illuminance in passenger stations.

Description

Technical Field

[0001] This invention relates to the field of railway passenger station optimization technology, and in particular to a method and system for generating illumination adjustment strategies for railway passenger station hubs. Background Technology

[0002] As a core node in the integrated transportation network, the lighting system of railway passenger station hubs not only directly affects passengers' travel experience, visual comfort, and traffic safety, but is also closely related to energy consumption and operating costs. Currently, railway passenger station hub lighting often adopts manual adjustment or illuminance settings based on time of day, resulting in problems such as uneven illuminance (e.g., alternating light and dark passageways, excessively bright or dim areas in waiting areas), energy consumption imbalance (e.g., insufficient illuminance during peak hours and excessive lighting during off-peak hours), and low energy utilization (e.g., lamps operating at full capacity when natural light is abundant).

[0003] Therefore, there is an urgent need for an illuminance adjustment scheme that can adapt to the combined effects of multiple factors such as regional characteristics (e.g., differences in functional zones such as entrance halls, waiting areas, and transfer passages), passenger flow tidal changes, and regional light environment differences, so that the railway passenger station hub lighting system can achieve precise control, on-demand lighting, and energy efficiency. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and system for generating illuminance adjustment strategies for railway passenger station hubs, which can adjust illuminance by combining multiple influencing factors to achieve precise control, on-demand lighting, and energy efficiency.

[0005] One aspect of the present invention provides a method for generating illuminance adjustment strategies for railway passenger station hubs, the method comprising the following steps: Acquire environmental index parameters, passenger flow index parameters, and monitoring image data of various functional areas of the railway passenger station; among them, environmental index and passenger flow index both include information types related to the overall illuminance of the passenger station and information types related to the illuminance of functional areas; Based on environmental and passenger flow parameters, correction coefficients are determined for various types of information in environmental and passenger flow parameters for each functional area. Based on correction coefficients for information related to the overall illuminance of the station and lighting design standards for each functional area, illuminance standard values ​​for each functional area are determined for future adjustment periods. By inputting environmental index parameters, passenger flow index parameters, and monitoring image data into a pre-trained illuminance prediction model, the illuminance prediction values ​​for each functional area during future adjustment periods are obtained. By combining the standard illuminance value and the predicted illuminance value, the illuminance fusion target value for each functional area during the future adjustment period can be obtained; For each functional area, the lighting equipment adjustment strategy for the area in the future adjustment period is obtained based on the illuminance fusion target value, the natural illuminance of the area, and the correction coefficients corresponding to various information.

[0006] In some embodiments of the present invention, the types of information related to the overall illuminance of the passenger station in the environmental indicators include the passenger station's geographical location, natural light environment, and weather conditions; the types of information related to the overall illuminance of the passenger station in the passenger flow indicators include the overall passenger flow of the passenger station. The types of information related to functional area illuminance in environmental indicators include area illuminance and its changes; the types of information related to functional area illuminance in passenger flow indicators include area passenger flow and its changes. Information related to natural light environment includes natural light intensity.

[0007] In some embodiments of the present invention, the illuminance standard value for each functional area during future adjustment periods is determined based on correction coefficients corresponding to information related to the overall illuminance of the passenger station and the lighting design standards for each functional area, including: The lighting design standards for each functional area are adjusted based on the correction coefficients corresponding to the natural light environment, weather conditions, and overall passenger flow of the station. For each functional area, the hardware feasibility of the lighting equipment in the area is verified based on the station's geographical location and the adjusted lighting design standards. If the verification passes, the adjusted lighting design standards will be used as the illuminance standard value for the area during the future adjustment period. If the verification fails, the correction coefficients corresponding to the natural light environment, weather conditions, and overall passenger flow of the station will be updated according to the correction coefficients corresponding to the station's geographical location.

[0008] In some embodiments of the present invention, environmental index parameters, passenger flow index parameters, and monitoring image data are input into a pre-trained illuminance prediction model to obtain illuminance prediction values ​​for each functional area during future adjustment periods, including: Image brightness values ​​are extracted from the monitoring image data of each functional area; Environmental index parameters, passenger flow index parameters, and image brightness values ​​are converted into feature representations, and the converted feature representations are input into a pre-trained illuminance prediction model to output the illuminance prediction values ​​for each functional area during future adjustment periods.

[0009] In some embodiments of the present invention, the image brightness value is extracted from the monitoring image data of each functional area, including: A Gaussian filtering algorithm is used to denoise the monitoring image data, and a grayscale world algorithm is used to correct the color cast problem of the denoised monitoring image data. For each functional area, an image semantic segmentation algorithm is used to divide the corrected monitoring image data of the area into multiple sub-images, and all sub-images are converted to the HSV color space. In this way, the luminance channel features are extracted from the HSV color space to obtain the image luminance value of the area.

[0010] In some embodiments of the present invention, a lighting equipment adjustment strategy for the area during a future adjustment period is obtained based on the illuminance fusion target value, the natural illuminance of the area, and correction coefficients corresponding to various types of information, including: The difference between the target illuminance value and the natural illuminance of the area is used as the theoretical adjustment value. The theoretical adjustment value is then corrected based on the comprehensive correction coefficient to obtain the lighting equipment adjustment strategy for the area in the future adjustment period. The comprehensive correction coefficient is determined based on the correction coefficients corresponding to various information in environmental indicators and passenger flow indicators.

[0011] In some embodiments of the present invention, before correcting the theoretical adjustment value, the method further includes: subtracting the regional hardware illuminance gain from the theoretical adjustment value to update the theoretical adjustment value.

[0012] Another aspect of the present invention provides an illumination adjustment strategy generation system for railway passenger station hubs, including a processor, a memory, and a computer program / instructions stored in the memory. The processor is used to execute the computer program / instructions, and when the computer program / instructions are executed, the system implements the steps of the method described in any of the above embodiments.

[0013] Another aspect of the present invention provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the method described in any of the above embodiments.

[0014] Another aspect of the present invention provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method described in any of the above embodiments.

[0015] This invention proposes a method and system for generating illuminance adjustment strategies for railway passenger station hubs. It determines standard illuminance values ​​by combining multiple influencing factors (such as environmental and passenger flow dimensions), obtains predicted illuminance values ​​based on an illuminance prediction model, and integrates illuminance values ​​for various functional areas determined through different methods during future adjustment periods to obtain a target illuminance value. The illuminance of lighting equipment is then adjusted adaptively based on this target value. The method proposed in this application not only ensures passenger safety and experience but also reduces operating energy consumption and costs, thereby achieving "on-demand lighting, precise control, and energy efficiency" for railway passenger station hub lighting.

[0016] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the description and drawings.

[0017] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. In the drawings: Figure 1 This is a flowchart illustrating a method for generating illumination adjustment strategies for railway passenger station hubs according to an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of the process of obtaining the standard value of illuminance using an illuminance prediction model in one embodiment of the present invention.

[0020] Figure 3 This is a schematic diagram of the system architecture applicable to the illumination adjustment strategy generation method of this application in one embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0022] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0023] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0024] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0025] Existing illuminance adjustment schemes for railway passenger stations are prone to problems such as uneven illuminance, energy consumption imbalance, and unnecessary energy increases. Based on the shortcomings of existing technologies, this application proposes a complete technical system combining standard guidance, predictive support, intelligent adjustment, and experimental verification. This system can ensure passenger travel safety and experience while reducing operational energy consumption and costs, providing core technical support for the construction of smart railway passenger stations. Specifically, the illuminance adjustment strategy generation method proposed in this application aims to achieve "on-demand lighting, precise control, and energy efficiency" in railway passenger station hub lighting. Its implementation logic is as follows: Illuminance standard values ​​are determined by combining multiple influencing factors (such as environmental and passenger flow dimensions) and differentiated lighting design standards. Taking into account the illuminance standard values ​​determined according to current standards. Illuminance prediction values ​​obtained based on illuminance prediction models The target value for illuminance fusion can be obtained. According to the illuminance fusion target value Adaptive adjustment of lighting equipment illuminance to achieve the desired illuminance value for the current area. Approaching the target value of illuminance fusion .

[0026] Figure 1 This is a flowchart illustrating a method for generating illumination adjustment strategies for railway passenger station hubs according to an embodiment of the present invention. Figure 1 As shown, the method includes steps S110 to S150, as detailed below: Step S110: Obtain environmental indicator parameters, passenger flow indicator parameters, and monitoring image data of various functional areas of the railway passenger station. The monitoring image data can be images captured by cameras within the station during a specific historical period. The environmental and passenger flow indicator parameters can be historical values ​​for those parameters during a specific historical period, or estimated values ​​for future adjustments. The design can be tailored to the desired adjustment precision.

[0027] This application aims to achieve illuminance adjustment based on multi-dimensional influencing factors. The factors influencing the illuminance of passenger stations can be designed to include at least environmental dimension factors (i.e., environmental indicators) and passenger flow dimension factors (i.e., passenger flow indicators). Both environmental and passenger flow indicators include information types related to the overall illuminance of the passenger station and information types related to the illuminance of functional areas. For example, the information types related to the overall illuminance of the passenger station in the environmental indicators may include the station's geographical location, natural light environment (such as natural light illuminance, natural light flux, and natural light colorimetry), and weather conditions. The information types related to the illuminance of functional areas in the environmental indicators may include regional illuminance and its changes. Similarly, the information types related to the overall illuminance of the passenger station in the passenger flow indicators may include the overall passenger flow of the passenger station, and the information types related to the illuminance of functional areas in the passenger flow indicators may include regional passenger flow and its changes. It should be noted that the regional illuminance mentioned in this application refers to the illuminance of functional areas within the passenger station, which can be determined jointly by the illuminance of the illuminance equipment in that area and the natural light illuminance (when natural light can illuminate the area). The natural light illuminance only represents the illuminance of solar radiation and is unrelated to the illuminance of the passenger station's lighting equipment.

[0028] As an example, by quantifying the various information contained in environmental and passenger flow indicators, the parameters of these indicators can be determined. The geographical location of a passenger station can be quantified using its latitude and climate zone, and specific values ​​for latitude and climate zone can be obtained from a geographic information system (GIS). The natural light environment can be quantified using natural light intensity (also known as natural light strength), sunshine duration, and solar altitude angle. The values ​​of natural light intensity, sunshine duration, and solar altitude angle can be determined from historical meteorological data in the meteorological system or by monitoring specific values ​​using sensors installed in the natural environment (such as illuminance sensors and solar angle sensors). Weather conditions can be quantified using weather types (sunny, cloudy, rainy, snowy, etc.), and weather types can be obtained from the meteorological system or determined by meteorological sensors. The overall passenger flow of a passenger station can be quantified using overall passenger flow density, tidal characteristics, and peak fluctuations, and can be determined based on passenger flow sensors. The specific values ​​are determined by combining the monitoring image data captured by the ticketing system and / or station cameras; the regional illuminance and its changes can be quantified using the true value of regional illuminance and instantaneous light change. The true value of regional illuminance can be determined by combining the monitoring image data of the corresponding functional area and the illuminance values ​​collected by the illuminance sensors installed in the functional area. Instantaneous light change indicates the instantaneous changes in illuminance in the functional area caused by instantaneous obstruction or enhancement of natural light and artificial lighting failures. The regional passenger flow and its changes can be quantified using the true value of regional passenger flow and the passenger flow fluctuations in the functional area (also known as local passenger flow fluctuations). The true value of regional passenger flow can be determined using the passenger flow sensors installed in the functional area and the monitoring image data of the corresponding functional area. Local passenger flow fluctuations indicate the instantaneous increase or decrease in passenger flow in the functional area. The quantification methods of the various types of information mentioned above are only examples and can be designed according to adjustment needs.

[0029] Based on the update frequency and scope of influence of the influencing factors, the multidimensional factors affecting the illuminance of functional areas designed in this application may include information with a slow update speed (such as the geographical location of the passenger station and the natural light environment), information with a medium update speed (such as weather conditions and the overall passenger flow of the passenger station), and information with a fast update speed (such as regional illuminance and its changes, and regional passenger flow and its changes). The different update speeds mentioned in this application are intended to illustrate the different acquisition frequencies, mainly following the principle that the acquisition frequency of information with a slow update speed > the acquisition frequency of information with a medium update speed > the acquisition frequency of information with a fast update speed. This invention does not specifically limit the magnitude of the acquisition frequency; for example, information with a slow update speed can be acquired quarterly, information with a medium update speed can be acquired daily or hourly, and information with a fast update speed can be acquired minutely or secondarily.

[0030] The aforementioned influencing factors, including environmental and passenger flow dimensions, are merely examples. This application does not specifically limit the dimensions of the influencing factors. Those skilled in the art can add or change other dimensions of influencing factors based on the influencing factors mentioned in this application.

[0031] Based on the service attributes and lighting requirements of passenger station hubs, each passenger station hub can be divided into four core areas: core operation area (such as ticket gates / security check areas, ticket windows / counters, etc.), core service area (such as waiting area, entrance and exit halls and basic platforms, etc.), passage area (such as transfer halls / connecting passages, stairs / escalators, other platforms / covered overpasses, etc.), and auxiliary area (such as restrooms, office area / duty room, station square / parking lot, commercial area / dining area, etc.).

[0032] As an example, cameras not only collect illuminance data but can also simultaneously extract passenger flow distribution characteristics, providing data support for coupled "illuminance-passenger flow" analysis and achieving "one device, multiple uses." The deployment of surveillance cameras within a passenger station can be configured as follows: the station is divided into multiple collection units by functional area grids. The camera density can vary between collection units corresponding to different functional areas; for example, in core service areas (such as waiting areas), there could be 200 cameras per collection unit. One camera will be deployed (meeting the requirements of the "Guidelines for the Deployment and Installation of Cameras in Railway Passenger Stations"). Fixed-point continuous deployment will be used in passageways (such as transfer corridors) (refer to the "General Technical Requirements for the Construction of Intelligent Systems for Integrated Passenger Transport Hubs"). The camera installation height will be controlled between 3 and 5 meters, with the lens tilted 15° towards the center of the area, avoiding obstructions such as pillars and seats that could affect the image frame acquisition range (see the "Code for Design of Video Security Monitoring System Engineering" (GB 50395-2007), the "Standard for Electrical Design of Civil Buildings" (GB 51348-2019), and the "Guidelines for the Deployment and Design of Cameras in Railway Passenger Stations"). Images from each area will be acquired at a default rate of 30fps, and data will be synchronously acquired every 5 seconds with the light sensor and passenger flow sensor to establish a correspondence between "image frame - natural light intensity - passenger flow data" and ensure data timestamp consistency. In addition, key areas (such as ticket halls and transfer nodes) can adopt a dual-camera backup mode ("main and secondary cameras"), while also linking with the light sensor at the same location to provide a benchmark for data calibration. The image acquisition equipment uses industrial-grade cameras with a resolution of 1080P or higher and a dynamic range (DR) of ≥120dB, supporting a frame rate of 30fps or higher (automatically boosting to 60fps during peak or inclement weather to reduce acquisition errors caused by motion blur and sudden changes in light intensity); the lens is a wide-angle fixed-focus lens, and a single device can cover an area of ​​≥50 m. 2 It balances data collection efficiency with detail accuracy.

[0033] Step S120: Based on environmental index parameters and passenger flow index parameters, determine the correction coefficients corresponding to various types of information in environmental index and passenger flow index for each functional area, and based on the correction coefficients corresponding to information related to the overall illuminance of the passenger station in environmental index and passenger flow index and the lighting design standards of each functional area, determine the illuminance standard value of each functional area during the future adjustment period.

[0034] Environmental and passenger flow parameters are quantified information within these parameters. Therefore, correction coefficients can be determined based on the quantified values ​​of each type of information, providing a quantifiable basis for illuminance adjustment calculations. Furthermore, to facilitate adaptive illuminance adjustments for different functions, this application designs different correction coefficients for each type of information across different functional areas. Specifically, this application uses any of the following methods to determine the correction coefficients for each type of information across various functional areas: ① Based on the function of each functional area, assign a corresponding functional area correction coefficient to each functional area to reflect the illuminance requirements of different functional areas. For example, this application can set a correction coefficient for each functional area. All correction coefficients are between 1.0 and 2.5, with the correction coefficient for the waiting area > the correction coefficient for the transfer passage > the correction coefficient for the equipment area. Based on the specific values ​​of various information types in the environmental and passenger flow indicators, a corresponding correction coefficient is assigned to each type of information (the range of correction coefficient values ​​can be found in Table 1), resulting in N+1 correction coefficients (N represents the total number of information types in the environmental and passenger flow indicators). Since the correction coefficients differ across areas, the correction coefficients for each type of information also differ relative to different areas. For example, the correction coefficient for the waiting area is 2.0, and the correction coefficient for the transfer passage is 1.5. Assuming the correction coefficient for the overall passenger flow of the station is 1.0, then the correction coefficients for the waiting area and the transfer passage are different for the overall passenger flow of the station (1.0 is different from 2.0 and 1.5). ② Combining the functions of the functional areas and the quantitative values ​​of various information, for each functional area, a corresponding correction coefficient is directly assigned to each type of information (the range of the correction coefficient can also be referred to Table 1), resulting in N×M (M represents the number of functional areas) correction coefficients. For example, the correction coefficient for the overall passenger flow in the waiting area is 2.0, and the correction coefficient for the overall passenger flow in the transfer passage is 1.5.

[0035] Table 1. Information Type-Quantization-Correction Coefficient Mapping Table As an example, if the weather conditions deviate (e.g., the measured illuminance is lower in cloudy or rainy weather), the weather correction coefficient will be automatically increased by 0.05 to 0.1; if passenger flow deviates (e.g., the actual passenger flow is higher than the predicted value), the overall passenger flow correction coefficient or the regional passenger flow correction coefficient will be automatically increased by 0.05 to 0.15; if there is a sudden change in illumination (e.g., a sudden decrease in illumination), the regional illuminance correction coefficient will be automatically increased by 0.1 to 0.2. That is, the magnitude of the correction coefficient can be determined based on the quantified value of the information. For example, in the case of "peak passenger flow + cloudy / rainy weather," the correction coefficient corresponding to the overall passenger flow and weather conditions of the station can be appropriately increased. Furthermore, the single adjustment range of a correction coefficient should be ≤0.2, and the cumulative adjustment range should be ≤0.3, to avoid excessive correction of the coefficient leading to subsequent illuminance fluctuations.

[0036] Natural light environment is highly dependent on the season; therefore, this application can also design a natural light environment correction coefficient that adapts to the season. The natural light environment and its correction coefficients are updated daily. For example, by utilizing a seasonal adaptation strategy, combining the inherent characteristics of sunshine duration and solar altitude angle in different seasons, a smooth transition between natural light and artificial lighting can be achieved, without illuminance gaps. The specific seasonal adaptation strategy is as follows: ① Summer: Sufficient sunshine duration and high solar altitude angle... And raise the threshold for activating supplemental lighting to utilize natural light. (i.e., natural light intensity) This application method is only activated when artificial lighting is needed to extend the off-peak hours of artificial lighting and maximize energy savings; ② Winter: Short sunshine hours and low solar altitude angle, Artificial lighting should be started 1 to 2 hours in advance for preheating, and the threshold for starting supplemental lighting for natural light utilization should be lowered to [a certain value]. (Right now The method described in this application is activated (using artificial lighting) to ensure that the regional illuminance always matches the target illuminance; ③ Spring and autumn: Sunlight conditions are balanced. and set The method described in this application is activated to utilize lighting equipment for supplemental lighting, balancing natural light utilization and lighting comfort, without requiring additional threshold adjustments; ④ Extreme weather (heavy rain / heavy snow / fog): correction coefficient And directly block the natural light utilization strategy, at which point the proportion of natural light contribution is... Turn on all artificial lighting circuits to ensure that the illuminance in the area meets the standards.

[0037] In some embodiments of the present invention, the illuminance requirements for each area are matched with mandatory regulations (such as national or industry standards such as the "Standard for Lighting Design of Buildings" (GB 50034-2013), the "Code for Lighting Design of Railways" (TB 10089-2015), and the "Technical Specification for LED Lighting in Railway Sites" (T-CET411-2024). Therefore, this application can design correction coefficients based on existing lighting design standards, environmental indicators, and passenger flow indicators to determine the illuminance standard values ​​for each functional area. Specifically, the process of obtaining a single, definitive illuminance standard value based on national / industry standards and through multiple layers of adaptive modifications is as follows: ① Determine the basic illuminance requirements for each functional area based on the station's size. For example, the station size can be divided into extra-large (annual passenger volume ≥ 50 million passengers / year), large (10 million passengers / year ≤ annual passenger volume < 50 million passengers / year), medium (2 million passengers / year ≤ annual passenger volume < 10 million passengers / year), and small (500,000 passengers / year ≤ annual passenger volume < 2 million passengers / year); ② The station's geographical location, natural light environment (adaptable to the season), weather conditions, and... The overall passenger flow of the station is the core basis for dynamically adjusting the basic illuminance requirement. The lighting design standards for each functional area are adjusted based on the correction coefficients corresponding to the natural light environment, weather conditions, and the overall passenger flow of the station. ③ For each functional area, the hardware feasibility of the lighting equipment within that area is verified based on the station's geographical location and the adjusted lighting design standards. If the verification passes, the adjusted lighting design standards are used as the illuminance standard value for that area during future adjustment periods. If the verification fails, the correction coefficients corresponding to the natural light environment and weather conditions are updated according to the correction coefficients corresponding to the station's geographical location (and the correction coefficients corresponding to the overall passenger flow of the station can also be updated) until the verification passes. In other words, this application can determine the illuminance standard value through quantitative calculation via "average illuminance standard selection → specific type information correction → hardware adaptation verification". The unique value provides a compliant and accurate core illuminance benchmark for subsequent lighting adjustments.

[0038] As an example, taking a super-large passenger station hub as an example, the process of determining the illuminance standard value is as follows: ① Refer to the average illuminance standards for each functional area in Table 2 to determine the basic illuminance requirements for the corresponding functional areas. If the average illuminance standard for a functional area in the illuminance design standard is a fixed value (e.g., the average illuminance standard of 200 lux for the waiting area in Table 2 is a fixed value), then the basic illuminance requirement for the functional area is this fixed value. If the average illuminance standard for a functional area in the illuminance design standard is a range value, then the basic illuminance requirement for the functional area can be the midpoint of the range (e.g., the average illuminance standard for the ticket gate / security check area in Table 2 is a range value of 500–700 lux, then the basic illuminance requirement for the ticket gate / security check area in a large passenger station is 600 lux). Taking the midpoint of the range when the average illuminance standard is a range value is only an example; the specific value can be designed independently.

[0039] Table 2. Basic Illumination of Major Functional Areas in Extra-Large Integrated Hubs When determining the basic illuminance requirements based on the illuminance design standards, this application may also set additional value conditions, such as the illuminance of key operating areas such as ticket gates and ticket counters not less than 500 lx during peak passenger flow periods and not less than 200 lx during off-peak passenger flow periods.

[0040] In the absence of an average illuminance standard for a specific functional area of ​​a passenger station of a certain size in the existing lighting design standards, the average illuminance standard for each functional area in a super-large passenger station can be used as a basis to adapt and adjust to obtain the basic illuminance situation for other passenger station sizes. For example, for the same functional area, the average illuminance standard for a large passenger station hub is 90% of the average illuminance standard for a super-large passenger station hub, and the average illuminance standard for a medium and small passenger station hub is 80% of the average illuminance standard for a super-large passenger station hub, so as to adapt to the passenger flow scale and spatial scale of different hubs.

[0041] ② Information Correction By overlaying natural light conditions and weather conditions onto the baseline illuminance requirement, the corrected baseline illuminance can be calculated. . Basic illuminance requirements, The climate correction coefficient is obtained based on the correction coefficients corresponding to the natural light environment and weather conditions. The climate correction coefficient can be obtained by weighting the correction coefficients corresponding to the natural light environment and the weather conditions, or it can be determined by the magnitude of the correction coefficients corresponding to the natural light environment and the weather conditions. This application does not limit the method of determining the climate correction coefficient.

[0042] As shown in Table 3, this application can design regular weather for each season: Summer ,winter , spring and autumn Extreme weather conditions in each season (such as heavy rain or heavy snow): The value can be selected based on the severity of the weather, and the adjusted basic illuminance requirement must match the basic illuminance requirement of emergency lighting.

[0043] Table 3 Climate Correction Factors and Illuminance Differences For example, When the value is within a range, the appropriate correction coefficient can be selected based on the latitude, climate zone characteristics, and seasonal attributes of the passenger station's geographical location. For example, in northern regions with abundant sunshine and a high solar altitude angle in summer, the basic illuminance requirement can be appropriately lowered to maximize energy savings. A correction factor of 0.95 is used. Southern regions experience hot and humid summers with frequent cloudy, rainy, and thunderstorm weather. While solar intensity is high, the proportion of diffused light is large, resulting in less stable actual effective natural sunlight compared to the north. Therefore, the correction factor is appropriately reduced to ensure that illuminance remains within acceptable limits during cloudy and rainy periods. A value of 0.9 is used. In northern regions, winter sunshine hours are short, the solar altitude angle is low, and natural light is significantly insufficient. Therefore, the illuminance standard needs to be increased to compensate for the lack of natural light and ensure passenger safety and visual comfort. A value of 1.1 is used; since winters in the south are relatively mild and sunshine conditions are better than in the north, the correction range is slightly lower than that in northern winters, striking a balance between energy conservation and ensuring sufficient illumination. 1.05 is acceptable.

[0044] ③ Passenger flow correction exist Based on the overall passenger flow of the station, a correction factor is added. Calculate the calibrated illuminance requirement .

[0045] This application can be designed The value is determined as follows: The overall passenger flow density of the station is determined using data from passenger flow sensors, the ticketing system, and surveillance images captured by station cameras. The overall passenger flow is then divided into peak (regional passenger flow density ≥ 80% of designed passenger flow), off-peak (30% ≤ regional passenger flow density < 80% of designed passenger flow), and low-peak (regional passenger flow density < 30% of designed passenger flow). The value is then assigned based on the peak level of the overall passenger flow at the station: Peak... Peak Low peak It is important to note the calibrated illuminance requirements. Additional value conditions mentioned above must be met, such as the illuminance in key operating areas such as ticket gates and ticket counters not being less than 500 lx during peak passenger flow periods and not less than 200 lx during off-peak passenger flow periods.

[0046] ④ Hardware verification to ensure feasibility of implementation. After obtaining the calibrated illuminance requirements Afterwards, the standard illuminance value can be determined based on the passenger station's geographical location and hardware compatibility verification. As shown in Table 4 (although the parameters in Table 4 are not directly involved in the calculation, they can be the standard values ​​for illuminance). The implementation includes hardware feasibility verification. Hardware compatibility verification follows these steps: If the passenger station is located in a frigid / cold / temperate region, verify the lighting fixtures' protection rating and operating temperature to ensure they can stably output appropriate light intensity; if the passenger station is located in a hot region, verify that the lighting fixtures' heat dissipation design meets standards to avoid insufficient heat dissipation leading to unsatisfactory actual illuminance. If the passenger station is located in a high-humidity area, check whether the moisture-proof function of the lamps meets the standards to avoid leakage or short circuit.

[0047] If the verification passes (i.e., the hardware is compatible), then If the verification fails (i.e., the hardware is not compatible), then fine-tuning will be performed. (e.g., fluctuation ±0.05) or adjust the correction factors corresponding to natural light environment, weather conditions, and overall passenger flow at the station, and recalculate. and And re-verify the determined If the verification fails after multiple adjustments to the correction factor, the preset value will be used as the standard illuminance value. .

[0048] Table 4 Lighting Equipment Parameter Requirements for Passenger Stations in Different Geographical Locations Based on the above four-step method, the standard illuminance value for the next hour can be calculated using correction factors corresponding to national / industry standards, hub type, functional area, and information related to the overall illuminance of the passenger station. The process is as follows: Taking the ticket gate in a large passenger station hub as an example, under typical scenarios of hot northern regions, normal summer weather, and peak passenger flow, the basic illuminance requirement is determined to be 650 lux based on Table 2; combined with the typical summer weather in northern regions, the required illuminance is determined to be... It is 0.95 at this time. The value is 617.5 lux; during peak passenger flow... ,at this time lux; During hardware verification, it must be ensured that the heat dissipation design can output a specific illuminance (specific illuminance = 679.25 lux - actual measured illuminance). The light intensity meets the requirements; the calibration is passed (luminous efficacy ≥ 130 lm / w, maximum illuminance ≥ 800 lux) and confirmed. lux.

[0049] The lighting system is dynamically adjusted according to the station's geographical location, natural light environment (adapting to the season), weather conditions, and overall passenger flow. Its update frequency can be matched with the frequency of acquiring key parameters (e.g., update frequency ≤ daily) to avoid fluctuations in the lighting system caused by frequent updates. Furthermore, the illuminance uniformity and glare values ​​of each functional area can be used as... The supporting technical indicators must be met simultaneously with subsequent lighting adjustments and hardware light distribution, and the color temperature requirements for each season / weather can be coordinated with the existing illuminance standards to improve the visual comfort of passengers.

[0050] Step S130: Input environmental indicator parameters, passenger flow indicator parameters, and monitoring image data into the pre-trained illuminance prediction model to obtain the illuminance prediction values ​​for each functional area during the future adjustment period. The acquired environmental indicator parameters, passenger flow indicator parameters, and monitoring image data for each functional area can be input into the illuminance prediction model together. Alternatively, for each functional area, the quantified values ​​of information related to the overall illuminance of the passenger station from the monitoring image data, environmental indicator parameters, and passenger flow indicator parameters, as well as the quantified values ​​of information related to the illuminance of that area from the environmental indicator parameters and passenger flow indicator parameters, can be used as an input dataset, thereby inputting the input datasets for all functional areas into the illuminance prediction model separately. Furthermore, when performing illuminance prediction based on the illuminance prediction model and visually acquired data, this application does not specifically limit the architecture of the illuminance prediction model; it can be as follows... Figure 2 As shown, the influence of dynamic changes in passenger flow, weather, and natural light on illuminance can be predicted by combining a time series large model (such as TimeGPT) and LSTM (time series large models are used for long-term prediction, LSTM is used for short-term prediction, and the illuminance prediction results of the model can be obtained by integrating the output of the time series large model and the output of LSTM). Illuminance prediction models can also be constructed by using genetic algorithms or ant colony algorithms.

[0051] More specifically, the process of obtaining illuminance prediction values ​​using the illuminance prediction model is as follows: extract the image brightness values ​​of each functional area from the monitoring image data of each functional area; convert the environmental index parameters, passenger flow index parameters, and image brightness values ​​into feature representations, and input the converted feature representations into the pre-trained illuminance prediction model to output the illuminance prediction values ​​of each functional area in the future adjustment period.

[0052] In some embodiments of the present invention, the images captured by the camera need to go through a technical chain of "interference cancellation - feature extraction - model mapping" to achieve accurate conversion from visual information to image brightness values. The specific process is as follows: ① Image data preprocessing To adapt to the recognition of image brightness values, this application can further perform the following combined processing on the image frames captured by the camera, based on preprocessing, to reduce image noise and deviation caused by the complex lighting environment of the passenger station hub, including: Noise Removal: Gaussian filtering algorithm is used to eliminate high-frequency noise caused by ambient light fluctuations in monitoring image data. It can also be combined with median filtering to process abnormal pixels caused by slight camera shake, ensuring stable image brightness information; Illumination Correction: Gray-world algorithm is used to correct color cast issues in images under different light source color temperatures (such as sunlight and LED lights); Region segmentation: Introducing an image semantic segmentation model (such as the U-Net model) to automatically identify sub-regions such as "passenger activity area, passageway, seat, wall, door and window" in the image. Combined with the station functional zoning map, the corresponding illuminance standards and analysis rules for each sub-region are matched to avoid confusion of the lighting characteristics of different functional areas.

[0053] ② Illuminance mapping and calculation, i.e., obtaining image brightness values ​​based on the processed image. This application can be designed to use a dual selection method of "basic calibration + high-precision prediction" to meet the accuracy requirements of different scenarios: (1) Reference-based calibration method A standard grayscale plate (with a known reflectivity, e.g., ρ=0.85) is fixedly deployed within the monitoring range of each camera to serve as the physical benchmark for illumination calculations. The average brightness value L (unit: cd / m²) of the grayscale plate area is extracted from the image. 2 ), combined with ambient light calculation formula ( This method represents the area illumination measured by the camera (in lux), and a linear mapping model of "image brightness - actual illumination" can be established. It is suitable for scenarios with high real-time requirements, with a computational delay ≤50ms and an accuracy error controllable within ±10 lux.

[0054] (2) Color conversion For areas without reference objects or with complex lighting, all sub-images are uniformly converted to the HSV color space, and the luminance channel (V channel) features are extracted as the core object for illuminance analysis, thus removing color interference. This method is applicable to core scenarios such as illuminance uniformity analysis and accurate prediction.

[0055] In addition to image brightness values, this application can also extract multi-dimensional core features of the image (such as regional average grayscale value, mean brightness histogram value, contrast, entropy value, highlight ratio, shadow ratio, edge brightness gradient, and color saturation, etc.), and combine them with auxiliary data such as concurrent passenger flow density and weather conditions, inputting them into an illumination prediction model (such as a lightweight LightNet model optimized based on the CNN-LSTM architecture) to obtain illumination prediction values. The illumination prediction model can be trained with massive "image features - measured illumination from light sensors" samples, incorporating different influencing factors and weather (sunny, rainy, snowy) scene labels for different areas of the passenger station at different times (morning peak, night), significantly improving the model's generalization ability.

[0056] The output data of the illuminance prediction model can be divided into three dimensions: 1) Time dimension: such as outputting the illuminance prediction value for the next 10 minutes, 1 hour, 6 hours, 1 day or 1 week; 2) Spatial dimension: outputting detailed prediction results by functional area (such as waiting area 1~3, transfer passage A~B); 3) Confidence dimension: outputting the confidence interval of the illuminance prediction value (such as the illuminance of the waiting area will be 220 lux in the next hour at a 95% confidence level).

[0057] As an example, taking the illumination prediction model using a hybrid architecture of "time series large model fine-tuning + LSTM model assistance", the illumination prediction model can be trained in the following way: 1) Dataset partitioning: Divide the collected multi-source heterogeneous data into training set, validation set and test set in a ratio of 7:2:1; 2) Training strategy: Use "transfer learning + incremental training", first use general time series data to train the basic capabilities of the large model, and then use multi-source heterogeneous data for fine-tuning to achieve long-term (e.g., 1 day to 1 week) prediction, and introduce LSTM model to process short-term (e.g., 10 minutes to 1 hour) high-frequency data to improve short-term prediction accuracy; subsequently, new data are added every quarter for incremental training to avoid model aging; 3) Loss function: Use mean squared error (MSE) combined with mean absolute error (MAE) as the loss function, and introduce a penalty term to increase the penalty for prediction error during peak passenger flow periods; 4) Optimization algorithm: Use AdamW optimizer, and adopt cosine annealing strategy for learning rate to improve model convergence speed and stability.

[0058] To ensure the prediction accuracy and generalization ability of the illuminance prediction model, this application can collect historical data from passenger stations to construct a multi-source heterogeneous data structure that is "all-dimensional, highly timely, and strongly correlated," covering four major categories of data: static basic features, historical operational patterns, real-time dynamic changes, and refined visual perception. All data sources and collection frequencies can be deeply matched with subsequent model training requirements and lighting adjustment algorithm logic to ensure data availability and relevance. Specifically, the multi-source heterogeneous data can include static basic data, historical operational data, and visually acquired data. Static basic data can include equipment static data (such as parameters of intelligent lighting fixtures such as rated luminous efficacy, maximum illuminance, color temperature range, and protection level), spatial static data (such as the layout of railway passenger station hub areas), and environmental static data (such as regional basic attributes (latitude, climate zone), seasonal sunshine patterns (multi-year average sunshine duration, range of solar altitude angle changes)). Static basic data does not change in the short term and does not require high-frequency collection; it can be collected once and updated synchronously only when the passenger station's functions are upgraded or lighting fixtures are replaced. Historical operational data can include the actual true illuminance values ​​of each functional area (measured illuminance from illuminance sensors and cameras). The data collected includes: intensity fusion determination, passenger flow data (overall passenger flow data of the station and statistics on local passenger flow fluctuations), lighting system energy consumption data (electricity consumption per unit area and lamp operating power, etc.), and meteorological data (average daily light intensity, weather type, and seasonal transition nodes); visual acquisition data uses cameras as the core acquisition device to collect regional image data (including brightness distribution and color characteristics of sub-areas such as passenger activity areas, passages, and walls), and through standardized deployment and intelligent algorithms, determines the illuminance of each area, regional passenger flow distribution characteristics, and instantaneous light change events (such as cloud cover and direct strong light) based on image recognition.

[0059] Static data serves as the foundational feature for model training, supporting differentiated prediction of illuminance standards for different hub types and functional areas. Historical operational data is used to uncover long-term potential correlations between illuminance and multi-dimensional factors such as season, climate, and passenger flow. Visually acquired data can serve as an important supplement and verification method for illuminance data. Introducing visually acquired data enables the illuminance prediction model to achieve refined and spatial perception of regional illuminance. However, these data are quite complex. After acquiring multi-source heterogeneous data, this application employs a three-step method of "cleaning-normalization-feature fusion" to process multi-source heterogeneous data: ① Cleaning: Outliers (such as extreme values ​​caused by sensor failures or blurred images caused by camera obstruction) can be removed using the Laida criterion, and missing values ​​can be supplemented using linear interpolation; ② Normalization: Data of different dimensions (such as passenger flow density, illuminance, and temperature) are mapped to the [0,1] interval to avoid the impact of magnitude differences on model data processing; ③ Feature fusion: Static and dynamic features in image frames are dynamically fused through an attention mechanism to highlight key features such as peak passenger flow and rainy weather.

[0060] In some embodiments of the present invention, this application can establish an illuminance prediction model update mechanism of "multi-source measurement - prediction comparison - error calibration - model iteration". It does not rely solely on camera recognition data, but integrates the dual-source illuminance values ​​collected by camera visual recognition and illuminance sensors with AI prediction values ​​for comparison and analysis. This achieves both real-time compensation for illuminance adjustment and dynamic updating and optimization of the illuminance prediction model, ensuring a high degree of matching between prediction accuracy and actual scene. The specific process is as follows: ① Dual-source measured illuminance fusion: During future adjustment periods, the real-time illuminance values ​​identified by cameras in the same area and the measured illuminance values ​​collected by illuminance sensors will be integrated after noise reduction and error calibration using a Kalman filter algorithm to generate the true illuminance value for the area. (Error ≤ 5 lux) serves as the core basis for comparative analysis and model updates, avoiding judgment errors caused by data deviations from a single device.

[0061] The integration rules can be as follows: 1) Assign weights according to the passenger flow of each area (e.g., 0.6 for densely populated areas, 0.3 for ordinary areas, and 0.1 for vacant areas) and calculate the average illuminance of the functional areas; 2) Extract the maximum and minimum illuminance values ​​of all functional areas and calculate the illuminance uniformity U0 = minimum illuminance / average illuminance (average illuminance is the average value of the average value) to identify blind spots or overly bright areas; 3) Weight the measured illuminance values ​​collected by cameras and sensors according to the assigned weights and output structured data of "real-time illuminance value - true illuminance value of the area - illuminance uniformity" to provide support for subsequent prediction and lighting adjustment.

[0062] ② Judgment of prediction and measurement error: the true value of regional illuminance Simultaneous and same-area illuminance prediction values ​​output by the illuminance prediction model Compare and calculate the relative error rate. .

[0063] ③ Error classification and handling: when When the model prediction accuracy is deemed to be up to standard, the model parameters are kept unchanged, and the true value of regional illuminance can be directly fed back to the dynamic adjustment algorithm with multi-factor linkage for real-time fine-tuning of lighting parameters. when When the model prediction is found to have a deviation, two actions are triggered simultaneously: first, real-time compensation for artificial lighting (the illuminance of lighting equipment can be adjusted according to the preset illuminance compensation value) to reduce the illuminance deviation caused by model error; second, dynamic updating of the illuminance prediction model is initiated to optimize model parameters and improve the accuracy of subsequent predictions.

[0064] If the error is caused by a hardware problem (such as camera lens contamination / grayscale plate obstruction, sensor malfunction), the error data will be automatically determined. If a device fails, the system switches to data from another normally functioning device as the basis for model updates and sends a hardware maintenance reminder to the operations and maintenance system. At the same time, it activates a hardware and algorithm collaboration mechanism to provide supplemental lighting, ensuring that model updates and lighting adjustments are not affected by hardware failures.

[0065] ④ Dynamic Update of Illuminance Prediction Model: Using the true illuminance value of the region as the core label, multi-source correlated data such as passenger flow density, meteorological data, natural light contribution ratio, and hardware gain illuminance for the corresponding time period are simultaneously incorporated to form a sample set for model updates. For large time-series models (long-term prediction), the feature weights of slowly changing factors such as climate and seasons are fine-tuned; for LSTM models (short-term prediction), the feature mapping relationships of rapidly changing factors such as local passenger flow and instantaneous changes in illuminance are updated.

[0066] The update frequency of the illuminance prediction model can be set as follows: under normal circumstances, it can be lightly iterated hourly (based on incremental training of real-time collected sample sets); when extreme weather or sudden large passenger flow is detected, it can be rapidly updated minute by minute to ensure the model's adaptability to special scenarios; and the model can be retrained quarterly based on all historical data to solidify the model's basic prediction capabilities.

[0067] In addition to the above error model calibration, periodic model calibration can also be performed. To avoid accuracy drift caused by camera aging and environmental changes, a dual calibration mechanism can be established: 1) Error calibration: Every 30 minutes, the measured values ​​of the light sensor at the same location are called up and compared with the camera recognition results and model prediction results. When the error is >10%, the mapping model parameters are automatically adjusted; 2) Periodic calibration: Every week, a professional illuminance meter (accuracy ±1 lux) is used to manually inspect each area to correct the color shift and brightness attenuation deviation of the camera and ensure the accuracy of the model in long-term operation.

[0068] Step S140: Combine the standard illuminance value and the predicted illuminance value to obtain the illuminance fusion target value for each functional area during the future adjustment period. .

[0069] This application does not specifically limit the fusion method. For example, weighted fusion can be performed using the following formula, or a Bayesian algorithm or a multi-objective optimization algorithm can be used. ; in, and These are the standard values ​​of illuminance. and illuminance prediction value The weights, and This application does not constitute a threat to the rights and interests of the parties involved. and The value can be specifically limited, for example, in normal scenarios. , Extreme / emergency scenarios It can be 0.8 to ensure the minimum standard for lighting, and Set to 0.2 to adapt to dynamic scene changes.

[0070] Step S150: For each functional area, based on the illuminance fusion target value, the natural illuminance of the area, and the correction coefficients corresponding to various information, obtain the lighting equipment adjustment strategy for the area during the future adjustment period.

[0071] The dynamic illuminance adjustment process in step S150 follows the adjustment rules of scene priority, natural light coordination, adjustment constraints, and group adaptation, specifically: ① Scene priority: Emergency scenarios (fire, power outage, passenger congestion) have the highest priority and directly trigger emergency adjustment, ignoring the comprehensive correction coefficient. and the proportion of contribution from natural light ② Natural light synergy: Adhering to the "natural light utilization strategy," meaning the higher the natural light intensity, the smaller the adjustment required for artificial lighting, avoiding energy waste; ③ Constraint mechanism: Dimming amount of lighting equipment (such as LED lamps). Limited to 0~1.0, conforming to the stepless dimming range of lighting equipment (which can be linked to the intelligent drive power supply of the lighting equipment through the PLC controller, supporting 0~100% dimming), avoiding overload or underload; ④ Group adaptation: according to the "group control" logic, calculations are performed separately for different functional areas such as waiting areas and transfer channels. This avoids local illuminance imbalance caused by overall adjustment.

[0072] In some embodiments of the present invention, for each functional region, the adjustment result generation process in step S150 may be as follows: fusing the illuminance of the region with the target value. The difference between the actual illuminance and the natural illuminance of the area is used as a theoretical adjustment value, and is based on a comprehensive correction factor. (Based on correction coefficients corresponding to various types of information) Corrected theoretical adjustment values ​​are obtained to arrive at the lighting equipment adjustment strategy for this region during future adjustment periods. The lighting equipment adjustment strategy can achieve bidirectional brightness adjustment from 0 to the rated maximum value, and the generation process of this adjustment result can be realized using an illuminance adjustment model, which can be expressed by the formula: ; in, The illuminance of each functional area within the passenger station (which can be referred to as the area illuminance value); The proportion of natural light contribution is used to quantify the effect of natural light on the body's energy expenditure. Actual contribution Values ​​in between; This indicates the natural illuminance of the corresponding functional area; The comprehensive correction coefficient is a correction coefficient obtained by coupling multiple influencing factors. It can reflect the comprehensive impact of factors such as the station's geographical location, passenger flow, weather, and light environment on lighting. This indicates the maximum illuminance of lighting equipment within a functional area. It can be the rated power of the lighting equipment or a user-defined value (such as for the core area). ≥800 lux, auxiliary area ≥300 lux); This represents the dimming level of the lighting equipment, directly corresponding to the illuminance adjustment ratio of the lighting equipment, and its value is between [0,1]. 0 means all off (0% output). 1 represents full on (100% rated output), corresponding to 0~100% of the rated illuminance of the lighting equipment. Therefore, according to... This will generate a lighting adjustment strategy for the area during future adjustment periods.

[0073] As an example, the illuminance calculated using the dynamic illuminance adjustment model... The value may not be within the range of [0,1]. Therefore, this application proposes the following adjustment strategy: if the calculated value is not within the range of [0,1]... Pick , representing the regional illuminance value (Including natural light) Illuminance fusion target value has been met. At this time, all lighting equipment is turned off or maintained at the minimum energy-saving brightness; if the calculated The calculated value from the dynamic illuminance adjustment model indicates the proportion to be adjusted according to the calculation. Adjusting the output of lighting equipment to achieve precise supplemental lighting for artificial illumination; if the calculated... Pick This indicates that the current illuminance deficit exceeds the rated output capacity of the lighting equipment. At this time, all lighting equipment is turned on, and the lighting system simultaneously sends an insufficient illuminance warning to the operation and maintenance system. Additionally, this application can also be designed for emergency scenarios such as fires or passenger congestion. Power outage emergency scenarios Custom values ​​are assigned based on the rated parameters of the emergency lighting fixture (e.g.) ≥0.5); under the scenarios of equipment failure / low passenger flow energy saving. It can be as low as 0.1 (10% output of lighting equipment, which is sufficient for basic lighting).

[0074] It should be noted that the zonal illuminance values ​​in this application It can be a calibrated fusion value of the measured illuminance from camera vision recognition and the measured illuminance from an illuminance sensor (installed in the corresponding functional area), i.e., the true value of the area illuminance. It can also be the measured illuminance value collected by the illuminance sensor or the measured illuminance value visually recognized by the camera.

[0075] In some embodiments of the present invention, a comprehensive correction coefficient is used. The correction coefficients are determined based on the corresponding correction coefficients for various information types in environmental and passenger flow indicators. Considering that different influencing factors have varying impacts on the illuminance of different functional areas, the core purpose of the correction coefficients for various information types in environmental and passenger flow indicators is to quantify the dynamic impact of each information type on the lighting requirements of each functional area within the station. Therefore, this application design uses the weighting of different dimensions to reflect the varying impacts of factors on the illuminance requirements of each functional area within the station, and combines the weights and correction coefficients of various information types to determine the correction coefficients. In order to determine the comprehensive impact of multidimensional influencing factors on lighting demand.

[0076] This application can design a process of "classification and weighting - coupled calculation - dynamic update" to determine the comprehensive correction coefficient. The details are as follows: ① Classification and weighting: Based on the classification of information types in Table 1, the correction coefficients may include slow correction coefficients (corresponding to information with slow update speed), medium correction coefficients (corresponding to information with medium update speed), and fast correction coefficients (corresponding to information with fast update speed). A combined weighting approach using the analytic hierarchy process (AHP) and entropy weighting is employed: The AHP determines the subjective weights corresponding to the correction coefficients of each information type (this can also be combined with experience from railway lighting engineering and national or industry standards); combined with environmental and passenger flow parameters, the entropy weighting method determines the objective weights corresponding to the correction coefficients of each information type. Integrating the subjective and objective weights yields the comprehensive weight for each correction coefficient. Alternatively, coupled models (such as multiple linear regression + BP neural network) can be used to analyze the interactions between factors to determine the weights.

[0077] In addition, when assigning subjective and / or objective weights in this application, the weights of the core influencing factors in a specific scenario can be increased by 10% to 20%. For example, the subjective and objective weights corresponding to the correction coefficient of the overall passenger flow of the station during peak hours can be increased by 20%, and the subjective and objective weights of the weather correction coefficient during rainy weather can be increased by 15%.

[0078] ② Calculation of the correction coefficient coupling: The calculation formula can be expressed as: ; in, The overall correction coefficient is the weighted product of all individual correction coefficients involved in the coupling; For a single correction factor (including the station's geographical location, overall passenger flow, and weather conditions, which can be dynamically filtered according to the actual scenario); The overall weight of a single correction coefficient satisfies ; To determine the number of information types involved in the coupling, the selection is dynamically based on the scenario. For example, in a typical scenario of "off-peak passenger flow + sunny weather," only the core dimensions of information with slow update speeds and moderate update speeds can be retained, while information with fast update speeds is removed. Conversely, in an extreme scenario of "peak passenger flow + heavy rain," all information types can be included. It should be noted that in the calculation... When necessary, a correction factor for functional areas can be introduced (if applicable) to reflect the different lighting requirements of different functional areas.

[0079] As an example, taking the waiting area of ​​a large passenger station as the target for lighting adjustments, several typical scenarios were selected for testing. Example of coupled calculation. Assume the overall weight allocation for each correction coefficient is as follows: geographical correction coefficient weight is 0.2, natural light environment (seasonal) correction coefficient weight is 0.15, weather correction coefficient weight is 0.25, overall passenger flow correction coefficient for the station weight is 0.2, functional area correction coefficient is 0.2, and the correction coefficient for information with fast update speed is 1 by default, as detailed below: Typical Scenario Example 1: Peak passenger flow + summer + hot region (extreme conventional scenario) Correction factor values: Geographical correction factor is 1.0, natural light environment correction factor is 0.95, weather correction factor is 1.0, overall passenger flow correction factor of the station is 1.3, and waiting area correction factor is 1.0; Coupling calculation formula: Considering that peak passenger flow may lead to localized passenger congestion in the waiting area, a correction factor for localized passenger flow fluctuations (with a value of 1.1) can be added when calculating the overall correction factor. .

[0080] Typical Scenario Example 2: Off-peak passenger flow + spring and autumn seasons + temperate regions (normal low-load scenario): Correction coefficient values: Geographical correction coefficient is 1.0, natural light environment correction coefficient is 1.0, weather correction coefficient is 0.9, overall passenger flow correction coefficient of the station is 0.8, and waiting area function coefficient is 1.0. Coupling calculation formula: While the natural illuminance of specific functional areas can be directly measured using outdoor illuminance sensors, this application also employs methods considering light refraction and reflection. and The product of these values ​​is considered the natural illuminance of the functional area. The proportion of natural light contribution... The value can be determined using data from historical operating phases, based on the true value of natural light illuminance. With regional illuminance value The proportion is determined.

[0081] As an example, in well-lit areas of passenger station hubs (such as waiting areas with glass curtain walls, under skylights, etc.), outdoor illuminance sensors and solar angle sensors are used to collect natural light illuminance (sampling frequency 1 minute / time, error ≤5 lux) and solar incidence angle (accuracy ±1°). Furthermore, the raw data collected by the sensors can be fused with illuminance data from video surveillance visual recognition using Kalman filtering to eliminate environmental interference and equipment errors, obtaining the true value of natural light illuminance. The proportion of natural light contributed The precise calculations provide core data support. Based on this historically collected data, the contribution ratio of natural light is determined. The process can be as follows: Natural light contribution ratio It employs a dual algorithm combining dynamic quantization and correction based on the solar angle (i.e., the angle of solar incidence), with the core being the true value of natural illuminance. With regional illuminance value The ratio calculation is based on the contribution ratio. And it is adjusted according to the scene based on the angle of solar incidence.

[0082] ① Calculate the basic contribution ratio If according to Compared to The proportional division of intervals allows for stepped value assignment, ensuring... The value ranges from 0 to 1.0. For example, in and When the proportion is ≥60% =0.6 (Sufficient natural light, maximizing the use of natural light and reducing artificial lighting), in and When the proportion is ≤30% =0 (natural light contribution is negligible, relying entirely on artificial lighting), and in 30% < and When the proportion is less than 60%, the value is dynamically assigned proportionally (dynamically assigned between [0, 0.6]).

[0083] ② Sunlight Angle Scene Correction: Based on the incident angle collected by the sunlight angle sensor, the contribution ratio to the baseline. Make targeted corrections to achieve the final value. Adapts to the actual lighting effects of direct / diffuse light. The correction strategy can be customized: when the incident angle is ≤30° (direct light is the main light source, and the light utilization rate is high), (Base value increased by 20%); when the incident angle is ≤60% (dominated by scattered light, low light utilization), (Base value reduced by 15%); no correction is made when 30° < incident angle < 60°. .at this time It is still constrained to the range of 0 to 1.0.

[0084] In some embodiments of the present invention, considering the potential impact of reflective coatings on the ceiling, glass curtain walls, and skylights on the illuminance of various functional areas within the passenger station, therefore, when determining... This application may also introduce This refers to the hardware gain illuminance (i.e., the gain value of the station's hardware facilities (such as skylights and reflective materials) on the area's illuminance). Specifically, for each functional area, based on the illuminance fusion target value, the area's natural illuminance, the correction coefficients corresponding to various information, and the hardware illuminance gain, the lighting equipment adjustment strategy for that area during future adjustment periods is obtained. The formula can be expressed as: ; Among them, it can be set ( This is a hardware correction factor. , This is a reference value for illuminance. The unit is lux. , .

[0085] This application employs an error calibration mechanism of "data acquisition - fusion of target values ​​- lighting adjustment" to achieve illuminance adjustment in various functional areas within the passenger station. The illuminance adjustment can be initiated by performing a full-area calibration every 30 minutes, or by setting the illuminance value for each functional area. According to The determined theoretical illuminance value (i.e. and Calibration is triggered immediately when there is a significant deviation (e.g., deviation > 10%) in the product of the two components. Failures in hardware devices such as data acquisition equipment and lighting equipment (e.g., sensor damage, lamp light decay, etc.) may cause this. According to The deviation between the determined theoretical illuminance values ​​can be used to periodically inspect the hardware equipment, and in the event of a fault, a hardware maintenance reminder can be sent directly to the operation and maintenance system, while simultaneously activating the hardware and algorithm collaboration mechanism to provide supplemental lighting.

[0086] In a specific embodiment of the present invention, the process of generating an illuminance adjustment strategy based on a real-time dynamic illuminance adjustment model and the value range of parameter settings is as follows: Waiting area of ​​extra-large passenger station ( (Lux), the scene combination is peak passenger flow + summer + hot region, with no local passenger flow fluctuations / instantaneous changes in illumination (high-speed / rapid change coefficients do not need to be superimposed), the hardware configuration includes skylight + high reflectivity wall (hardware gain needs to be taken into account), predicted illuminance value The lux (illuminance prediction model is based on natural light and passenger flow trend prediction in hot summer regions during peak periods) has the following fusion weights for the standard illuminance value and the predicted illuminance value: , ,therefore ; Comprehensive correction coefficient (Geographical correction factor 0.2 + Natural light environment correction factor) +Weather Correction Factor + Overall Passenger Flow Correction Coefficient for Passenger Station +Functional area correction factor (Results of coupled calculations) Regional illuminance values lux (the fused value of the illuminance measured by the illuminance sensor and the illuminance measured by the camera, with an error ≤ 5 lux), the proportion of natural light contribution. (In hot summer regions at noon, natural light is 50% of the target illuminance, with a sunlight angle of 25°, which is direct light and requires no additional correction.) Maximum illuminance of luminaires. lux, hardware gain illuminance (Gain of 15 lux from the waiting hall skylight + gain of 15 lux from the high reflectivity of the wall materials), the corrected formula is: The illuminance adjustment strategy is to "adjust the lighting equipment in the waiting area to 25.3% of the rated brightness" (i.e., adjust the lighting equipment in the waiting area to 25.3% of the rated brightness). This is to meet the illuminance requirements. Alternatively, it can be approximated as 800 × 25% = 200 lux using a 25% dimming ratio.

[0087] If there is localized passenger flow clustering in this scenario, a correction factor for localized passenger flow fluctuations (assumed to be 1.1) can be applied. Recalculate the dimming amount ,Right now (This means that the lighting equipment can be adjusted to 27.8% of its rated brightness).

[0088] The method proposed in this application follows the principles of "stepless adaptation + precise supplemental lighting + uniformity control" during dynamic adjustment: combined with regional grouping control logic, the illuminance dynamic adjustment model can be applied to calculate the dimming amount of lighting equipment separately for different functional areas (which can be divided into areas with advantageous lighting and areas with non-advantageous lighting). It achieves stepless dimming from 0 to 100%; during the illuminance adjustment process, it simultaneously controls the illuminance uniformity (≥0.7) and glare limitation (UGR≤19). For areas where direct light is prone to glare (such as waiting areas on the side of glass curtain walls), it links with the intelligent dimming glass to adjust the light transmittance, avoiding visual interference and local overbrightness caused by direct light.

[0089] In emergency scenarios, a centralized emergency lighting system (compliant with the mandatory requirements of GB 51309-2018) can be used, linked to the automatic fire alarm system (FAS), power monitoring system, video surveillance, and ticketing data platform to achieve rapid identification and response to emergency scenarios, with a trigger response time of ≤3 seconds. The core equipment of the lighting system uses a centralized emergency lighting power supply (rated voltage ≤DC 36V), and the supporting batteries must not contain cobalt, such as lithium cobalt oxide batteries, to ensure safety and stability under extreme conditions. Specific lighting strategies for emergency scenarios include: ① Power Outage Emergency Adjustment: Based on the functional zoning and emergency support needs of railway passenger station hubs, when the power monitoring system detects a main power outage, it sends corresponding information to the lighting system. The lighting system automatically switches to emergency lighting mode, prioritizing lighting for evacuation routes, safety exits, and key service areas (such as ticket gates / security checkpoints, waiting areas, stairs / escalators, safety exits, and evacuation routes), ensuring an illuminance of ≥50 lux (compliant with GB 17945-2024 standard). The continuous working time of emergency lighting is no less than 0.5 hours, and extended to more than 1.5 hours for passenger stations in super high-rise buildings, to meet the needs of safe evacuation of personnel.

[0090] ② Fire Emergency Adjustment: Upon receiving a fire alarm signal confirmed by the FAS (two independent detectors in the same area or one detector plus a manual alarm button), the lighting system immediately performs three actions: cutting off the normal lighting power to the fire area to avoid secondary risks; increasing the brightness of evacuation routes and safety exits to ≥100 lux to enhance escape visibility; and dynamically adjusting the directional signs via a centralized controller to guide the optimal evacuation route, avoiding the fire area. Simultaneously, the lighting system provides real-time feedback to the fire control room on the lamp's operating status, battery level, and evacuation direction adjustments. Fire scenarios require deep integration with the fire protection system, precisely controlling lighting areas according to fire compartments to avoid cross-area interference.

[0091] ③ Emergency Adjustment for Passenger Flow Congestion: Based on the spatial segmentation algorithm of video surveillance and the linkage of ticketing data, a passenger flow density classification standard is established, such as when the passenger flow density of a functional area is ≥2.0 people / m². 2 The system issues congestion warnings to the lighting system, prompting it to adjust its illumination. During adjustment, the illuminance in congested areas and surrounding access routes is increased to ≥300 lux, using warm white light (color rendering index Ra≥80) to enhance environmental visibility, assist in guiding people out of congestion, and reduce the risk of stampedes.

[0092] In an emergency, the lighting system can prioritize cutting off non-fire-fighting lighting loads to ensure sufficient emergency power capacity. After the emergency scenario is resolved (e.g., fire alarm is deactivated, passenger density drops to 1.5 people / m²), the system will automatically shut down. 2In the event of power outages (such as after mains power is restored), the lighting system automatically initiates a gradual recovery procedure, switching the lighting status from emergency mode to normal adjustment mode in 3-minute increments to avoid visual discomfort caused by sudden changes in illuminance. In addition, the lighting system supports both quarterly automatic self-checks and manual tests, monitoring for issues such as lamp malfunctions and battery degradation in real time and providing early warnings to ensure reliable emergency response.

[0093] This application also designs multi-dimensional lighting evaluation indicators as the basis for the acceptance and promotion of smart railway station lighting systems, providing a quantitative reference for the regional adjustment of lighting solutions. As shown in Table 5, the multi-dimensional lighting evaluation indicators focus on four aspects: technical performance, energy efficiency, user experience, and economy, ensuring the relevance and comprehensiveness of the evaluation. Experimental verification ensures the feasibility of the solution and solves the problem of "technology implementation."

[0094] Table 5. Evaluation System Composed of Multidimensional Lighting Evaluation Indicators When conducting experimental verification of the lighting adjustment scheme proposed in this application, different types of hubs can be selected for comparative experiments, and the secondary indicators of the evaluation system can be used as the sole quantitative standard for judging the adjustment effect.

[0095] Specifically, three different types of railway passenger station hubs were selected as test sites: 1) extra-large passenger stations (such as Shanghai Hongqiao Station) to test core service areas and high passenger flow scenarios; 2) large passenger stations (such as Wuhan Station) to test transfer channels and medium passenger flow scenarios; and 3) small and medium-sized passenger stations (such as Yichang East Station) to test auxiliary areas and low passenger flow scenarios. Three to five typical functional areas were selected as test units at each station, each with an area of ​​no less than 500 square meters. 2 This ensures the representativeness of the experiment.

[0096] Experimental platform setup: such as Figure 3 As shown, a four-layer architecture of "perception layer - transmission layer - decision layer - execution layer" is adopted. 1) Perception layer: Passenger flow sensors (accuracy ±5 people / m) are installed in the test area. 2The system consists of: 1) an illuminance sensor (accuracy ±1 lux), an industrial-grade camera (1080P resolution, DR≥120dB, 30fps frame rate, with a standard grayscale board), and a weather sensor, enabling simultaneous acquisition of multiple data sources; 2) a transmission layer: deploying 5G industrial modules and LoRa gateways to ensure data transmission latency ≤50ms; 3) a decision layer: building edge computing nodes (configured with Intel Core i7 processors and 16GB of memory) and a central controller. Image data acquired by the camera is preprocessed and basic illuminance calculated at the edge nodes to reduce transmission pressure. Illuminance prediction models can be deployed on the edge computing nodes. The central controller is responsible for global coordination and strategy optimization. It can receive structured illuminance data and illuminance prediction results uploaded by the camera, and lighting adjustment algorithms can be deployed on the central controller; 4) an execution layer: replacing the existing lamps in the test area with smart LED lamps and installing smart dimming drivers and emergency lighting modules.

[0097] Controlled experimental design: A "simultaneous control" model is adopted: at each test site, an area with the same conditions as the test unit (same area, function, and passenger flow characteristics) is selected as the control unit. The control unit uses a traditional lighting system (fixed illuminance, no intelligent adjustment), while the test unit adopts the technical system of this scheme. The test period is one year, covering all four seasons and peak passenger flow periods such as Spring Festival and summer travel seasons.

[0098] The trial phase will be conducted as follows: 1) Baseline data collection phase (month 1): Collect raw illuminance, energy consumption, passenger flow, and other data from the test unit and control unit as evaluation benchmarks; 2) Solution deployment phase (month 2): Deploy the hardware and software system of this solution in the test unit and complete model training and parameter debugging; 3) Dynamic operation phase (months 3-12): The system will operate normally, and technical performance indicators (illuminance compliance rate, uniformity, etc.), energy consumption indicators (energy consumption per unit area, energy saving rate, etc.), and user experience indicators (satisfaction, complaint rate, etc.) of the two types of units will be collected in real time, and an operation report will be generated monthly; 4) Extreme scenario testing phase: Under extreme scenarios such as heavy rain, heavy snow, and peak passenger flow (120% above design passenger flow), the system's adjustment capability and stability will be specifically tested.

[0099] After obtaining the evaluation results for each evaluation indicator in the experiment, statistical software (such as SPSS) can be used to compare the differences between the evaluation indicators in the method of this application (i.e., the experimental unit) and the traditional method (i.e., the control unit) (the significance of the difference is judged by t-test, with P<0.05 indicating a significant difference). The scheme is considered effective when each evaluation indicator of the method of this application reaches the corresponding target value. Alternatively, a combination of Analytic Hierarchy Process (AHP) and fuzzy comprehensive evaluation can be used to evaluate the differences between the method of this application and the traditional scheme. For example, the weights of the first-level indicators and the weights of the second-level indicators in Table 4 can be determined by AHP; a fuzzy evaluation matrix can be constructed using fuzzy comprehensive evaluation, and an evaluation group composed of multiple participants such as railway station operators, lighting engineers, and passenger representatives can be formed to perform fuzzy scoring on each indicator; the comprehensive evaluation results of the method of this application and the traditional lighting scheme can be obtained through matrix operation, and the lighting scheme can be divided into four levels: "Excellent (≥85 points), Good (70-84 points), Qualified (60-69 points), and Unqualified (<60 points)" based on the comprehensive evaluation results.

[0100] The test results were evaluated from four dimensions: In terms of technical performance, the illuminance compliance rate of the test unit was ≥95%, which was more than 20% higher than that of the control unit; in terms of energy efficiency, the energy saving rate of the test unit was ≥30%, and the energy consumption per unit area was reduced to below 5kWh / ㎡·d; in terms of user experience, the passenger satisfaction score of the test unit was ≥8.5 points, and the glare complaint rate was ≤0.5%; in terms of economics, the investment payback period of the test unit was ≤3 years, which met the expected goals.

[0101] This solution, through a complete technical chain of "standardization-prediction-adjustment-hardware-verification," can systematically address the problem of lighting in railway passenger station hubs being affected by multiple factors, achieving precision, intelligence, and energy efficiency in the lighting system. Furthermore, experimental verification shows that the proposed solution meets the expected goals in terms of technical performance, energy efficiency, user experience, and economy, demonstrating significant potential for widespread adoption.

[0102] Corresponding to the above method, the present invention also provides an illumination adjustment strategy generation system for railway passenger station hubs. The system includes a computer device, which includes a processor and a memory. The memory stores computer programs / instructions, and the processor is used to execute the computer programs / instructions stored in the memory. When the computer programs / instructions are executed by the processor, the system implements the steps of the method described above.

[0103] This invention also provides a computer-readable storage medium storing a computer program / instructions thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0104] This invention also provides a computer program product storing a computer program / instructions thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. The computer program product can be a tangible product, such as random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of product known in the art.

[0105] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0106] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0107] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0108] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for generating illuminance adjustment strategies for railway passenger station hubs, characterized in that, The method includes the following steps: Acquire environmental index parameters, passenger flow index parameters, and monitoring image data of various functional areas of the railway passenger station; among them, environmental index and passenger flow index both include information types related to the overall illuminance of the passenger station and information types related to the illuminance of functional areas; Based on the environmental index parameters and the passenger flow index parameters, the correction coefficients corresponding to various types of information in the environmental index and passenger flow index are determined for each functional area. Based on the correction coefficients corresponding to information related to the overall illuminance of the passenger station and the lighting design standards of each functional area, the illuminance standard values ​​of each functional area are determined for the future adjustment period. By inputting environmental index parameters, passenger flow index parameters, and monitoring image data into a pre-trained illuminance prediction model, the illuminance prediction values ​​for each functional area during future adjustment periods are obtained. By combining the standard illuminance value and the predicted illuminance value, the illuminance fusion target value for each functional area during the future adjustment period can be obtained; For each functional area, the lighting equipment adjustment strategy for that area during the future adjustment period is obtained based on the illuminance fusion target value, the natural illuminance of that area, and the correction coefficients corresponding to various types of information.

2. The method according to claim 1, characterized in that, The environmental indicators related to the overall illuminance of the passenger station include the passenger station's geographical location, natural light environment, and weather conditions; the passenger flow indicators related to the overall illuminance of the passenger station include the overall passenger flow of the passenger station. The environmental indicators related to functional area illuminance include area illuminance and its changes; the passenger flow indicators related to functional area illuminance include area passenger flow and its changes. The information in the category of natural light environment includes natural light intensity.

3. The method according to claim 2, characterized in that, The determination of the illuminance standard values ​​for each functional area during future adjustment periods, based on correction coefficients corresponding to information related to the overall illuminance of the passenger station and lighting design standards for each functional area, includes: The lighting design standards for each functional area are adjusted based on the correction coefficients corresponding to the natural light environment, weather conditions, and overall passenger flow of the station. For each functional area, the hardware feasibility of the lighting equipment in the area is verified based on the station's geographical location and the adjusted lighting design standards. If the verification passes, the adjusted lighting design standards will be used as the illuminance standard value for the area during the future adjustment period. If the verification fails, the correction coefficients corresponding to the natural light environment, weather conditions, and overall passenger flow of the station will be updated according to the correction coefficients corresponding to the station's geographical location.

4. The method according to claim 1, characterized in that, The process of inputting environmental index parameters, passenger flow index parameters, and monitoring image data into a pre-trained illuminance prediction model to obtain illuminance prediction values ​​for each functional area during future adjustment periods includes: Image brightness values ​​are extracted from the monitoring image data of each functional area; Environmental index parameters, passenger flow index parameters, and image brightness values ​​are converted into feature representations, and the converted feature representations are input into a pre-trained illuminance prediction model to output the illuminance prediction values ​​for each functional area during future adjustment periods.

5. The method according to claim 4, characterized in that, The extraction of image brightness values ​​from the monitoring image data of each functional area includes: The monitoring image data is denoised using a Gaussian filtering algorithm, and the grayscale world algorithm is used to correct the color cast of the denoised monitoring image data. For each functional area, an image semantic segmentation algorithm is used to divide the corrected monitoring image data of the area into multiple sub-images, and all sub-images are converted to the HSV color space. The luminance channel features are then extracted from the HSV color space to obtain the image luminance value of the area.

6. The method according to claim 1, characterized in that, The lighting equipment adjustment strategy for the area during future adjustment periods is obtained based on the illuminance fusion target value, the natural illuminance of the area, and correction coefficients corresponding to various information, including: The difference between the target illuminance value and the natural illuminance of the area is used as the theoretical adjustment value. The theoretical adjustment value is then corrected based on a comprehensive correction coefficient to obtain the lighting equipment adjustment strategy for the area during future adjustment periods. The comprehensive correction coefficient is determined based on the correction coefficients corresponding to various information in environmental indicators and passenger flow indicators.

7. The method according to claim 6, characterized in that, Before correcting the theoretical adjustment value, the method further includes subtracting the regional hardware illuminance gain from the theoretical adjustment value to update the theoretical adjustment value.

8. A system for generating illumination adjustment strategies for railway passenger station hubs, comprising a processor, a memory, and a computer program / instructions stored in the memory, characterized in that, The processor is configured to execute the computer program / instructions, and when the computer program / instructions are executed, the system implements the steps of the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.

Images