Subway ventilation air conditioner monitoring system

Abstract

The invention relates to the technical field of air conditioner monitoring, in particular to a subway ventilation air conditioner monitoring system. The method comprises the following modules: a working state identification module S1, which is used for carrying out multi-parameter real-time monitoring on a target subway ventilation air conditioner to obtain an air conditioner operation parameter set; recognizing the working state of the target subway ventilation air conditioner according to the air conditioner operation parameter set to obtain real-time working state data of the air conditioner; the passenger flow monitoring module S2 is used for carrying out panoramic shooting on the target subway carriage to obtain a real-time panoramic image of the carriage; passenger density distribution calculation is conducted on a target subway carriage according to the real-time panoramic image of the carriage, carriage passenger density distribution data are obtained, and the target subway carriage is the subway carriage where the target subway ventilation air conditioner is located. The problems that a traditional ventilation air conditioner monitoring system is poor in flexibility, slow in response, single in parameter regulation and control and waste in energy are effectively solved, and the comfort level of subway passengers is remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of air conditioning monitoring, and in particular to a subway ventilation and air conditioning monitoring system. Background Art

[0002] As an important means of transportation in modern cities, the comfort and safety of the subway's internal environment are crucial to the passenger experience. The ventilation and air-conditioning system is the core of the subway's internal environment control, which not only affects the direct feelings of passengers, but also affects the efficiency and safety of subway operations. However, the existing subway ventilation and air-conditioning monitoring methods have some problems that need to be solved.

[0003] Traditional ventilation and air conditioning monitoring systems usually operate in a fixed mode and lack sensitivity and adaptability to environmental changes. When faced with morning and evening rush hours, extreme weather or emergencies, these systems often fail to respond quickly and effectively, resulting in a poor passenger experience.

[0004] The complexity of the subway internal environment requires that the monitoring system must be highly flexible and accurate. Existing monitoring methods often use a single environmental parameter for regulation, ignoring the mutual influence and comprehensive effects of multiple factors. For example, the passenger density and temperature factors in the carriage will affect the working effect of the air-conditioning system, but existing systems rarely integrate these factors for intelligent regulation.

[0005] In addition, subway ventilation and air conditioning systems are often not designed with full consideration of passenger flow changes at different times and in different areas. During peak hours in the morning and evening, the number of passengers surges, while during off-peak hours, there are fewer passengers. The fixed-mode air conditioning system cannot dynamically adjust according to the actual passenger flow, resulting in energy waste and poor passenger experience. Summary of the invention

[0006] Based on this, it is necessary for the present invention to provide a subway ventilation and air-conditioning monitoring system to solve at least one of the above technical problems.

[0007] To achieve the above purpose, a subway ventilation and air conditioning monitoring system includes the following modules:

[0008] The working state identification module S1 is used to perform multi-parameter real-time monitoring of the target subway ventilation and air conditioning to obtain an air conditioning operation parameter set; the working state of the target subway ventilation and air conditioning is identified according to the air conditioning operation parameter set to obtain real-time working state data of the air conditioning;

[0009] The passenger flow monitoring module S2 is used to perform panoramic photography of the target subway car to obtain a real-time panoramic image of the car; calculate the passenger density distribution of the target subway car according to the real-time panoramic image of the car to obtain passenger density distribution data of the car, wherein the target subway car is the subway car where the target subway ventilation and air conditioning is located;

[0010] The working state adjustment module S3 is used to adjust the first working state of the target subway ventilation air conditioner according to the real-time working state data of the air conditioner and the passenger density distribution data of the carriage to obtain the first working state data;

[0011] The working state prediction module S4 is used to collect passenger flow time series data of the target subway car to obtain a historical passenger flow time series data set; obtain real-time timestamp data; and predict the target working state of the target subway ventilation air conditioner based on the historical passenger flow time series data set according to the real-time timestamp data and the first working state data to obtain the second air conditioner target state data;

[0012] The intelligent decision-making execution module S5 is used to generate a parameter adjustment instruction set for the target subway ventilation and air conditioning according to the first working state data and the second air conditioning target state data to obtain an air conditioning parameter adjustment instruction set; and perform intelligent actuator linkage control operations on the target subway ventilation and air conditioning according to the air conditioning parameter adjustment instruction set.

[0013] The present invention can obtain the air-conditioning operation status and passenger density distribution data in the carriage in real time through multi-parameter real-time monitoring and panoramic camera technology. This real-time monitoring enables the system to quickly respond to changes in peak hours, extreme weather or emergencies in the morning and evening, significantly improving the flexibility and response speed of the system. By comprehensively considering the air-conditioning operation parameters and passenger density distribution, the ventilation and air-conditioning can be accurately regulated. This method avoids the limitations of single parameter regulation and ensures that the ventilation and air-conditioning can provide the most suitable temperature and air quality under different passenger densities and environmental conditions, thereby improving the comfort of passengers. The present invention can not only perform dynamic regulation based on real-time data, but also predict the working status through historical passenger flow time series data. This prediction capability enables the system to automatically reduce air-conditioning energy consumption during off-peak hours and time periods with low passenger flow, thereby achieving the purpose of energy saving. At the same time, when the passenger flow increases, the system can timely improve the air-conditioning efficiency and avoid energy waste. Through the intelligent decision-making execution module, a parameter adjustment instruction set can be generated according to real-time data and prediction results to achieve linkage control of the subway ventilation and air-conditioning system, thereby improving the coordination and efficiency of regulation. In summary, the present invention effectively solves the problems of poor flexibility, slow response, single parameter control and energy waste in traditional ventilation and air-conditioning monitoring systems through real-time monitoring, intelligent prediction and precise control, and significantly improves the comfort of subway passengers. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0015] Figure 1 A schematic diagram of the module flow of a subway ventilation and air conditioning monitoring system according to an embodiment is shown.

[0016] Figure 2 A detailed step flow diagram of S3 of an embodiment is shown.

[0017] Figure 3 A detailed step flow diagram of S4 of an embodiment is shown. DETAILED DESCRIPTION

[0018] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0019] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0020] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0021] To achieve this, please refer to Figures 1 to 3 The present invention provides a subway ventilation and air conditioning monitoring system, comprising the following modules:

[0022] The working state identification module S1 is used to perform multi-parameter real-time monitoring of the target subway ventilation and air conditioning to obtain an air conditioning operation parameter set; the working state of the target subway ventilation and air conditioning is identified according to the air conditioning operation parameter set to obtain real-time working state data of the air conditioning;

[0023] The passenger flow monitoring module S2 is used to perform panoramic photography of the target subway car to obtain a real-time panoramic image of the car; calculate the passenger density distribution of the target subway car according to the real-time panoramic image of the car to obtain passenger density distribution data of the car, wherein the target subway car is the subway car where the target subway ventilation and air conditioning is located;

[0024] The working state adjustment module S3 is used to adjust the first working state of the target subway ventilation air conditioner according to the real-time working state data of the air conditioner and the passenger density distribution data of the carriage to obtain the first working state data;

[0025] The working state prediction module S4 is used to collect passenger flow time series data of the target subway car to obtain a historical passenger flow time series data set; obtain real-time timestamp data; and predict the target working state of the target subway ventilation air conditioner based on the historical passenger flow time series data set according to the real-time timestamp data and the first working state data to obtain the second air conditioner target state data;

[0026] The intelligent decision-making execution module S5 is used to generate a parameter adjustment instruction set for the target subway ventilation and air conditioning according to the first working state data and the second air conditioning target state data to obtain an air conditioning parameter adjustment instruction set; and perform intelligent actuator linkage control operations on the target subway ventilation and air conditioning according to the air conditioning parameter adjustment instruction set.

[0027] The present invention can obtain the air-conditioning operation status and passenger density distribution data in the carriage in real time through multi-parameter real-time monitoring and panoramic camera technology. This real-time monitoring enables the system to quickly respond to changes in morning and evening peaks, extreme weather or emergencies, significantly improving the flexibility and response speed of the system. By comprehensively considering the air-conditioning operation parameters and passenger density distribution, the ventilation and air-conditioning can be accurately regulated. This method avoids the limitations of single parameter regulation and ensures that the ventilation and air-conditioning can provide the most suitable temperature and air quality under different passenger densities and environmental conditions, thereby improving the comfort of passengers. The present invention can not only perform dynamic regulation based on real-time data, but also predict the working status through historical passenger flow time series data. This prediction capability enables the system to automatically reduce air-conditioning energy consumption during off-peak hours and time periods with low passenger flow, thereby achieving the purpose of energy saving. At the same time, when the passenger flow increases, the system can timely improve the air-conditioning efficiency and avoid energy waste. Through the intelligent decision-making execution module, a parameter adjustment instruction set can be generated according to real-time data and prediction results to achieve linkage control of the subway ventilation and air-conditioning system, thereby improving the coordination and efficiency of regulation. In summary, the present invention effectively solves the problems of poor flexibility, slow response, single parameter control and energy waste in traditional ventilation and air-conditioning monitoring systems through real-time monitoring, intelligent prediction and precise control, and significantly improves the comfort of subway passengers.

[0028] In this embodiment, multiple high-precision sensors are installed in the subway car, including temperature, humidity, wind speed and pressure sensors. These sensors transmit the key parameters of the air-conditioning system monitored in real time to the central processing unit through a specially designed data acquisition module. For example, a temperature sensor is used to monitor the inlet and outlet temperatures of the evaporator and condenser in real time, and then the heat exchange efficiency of the evaporator and condenser is calculated based on these temperature differences and air flow. These data are recorded once a minute and sent to the data processing center through a transmission device. Secondly, through the recognition mechanism, the current working state of the air conditioner, such as cooling, heating or ventilation, is automatically identified and classified. For example, the operating frequency, current value and voltage value of the compressor can be monitored in real time, and the current operating state of the air conditioner can be judged by these parameters. At the same time, a high-definition panoramic camera is installed in the car to capture the image information in the car in real time and obtain a real-time panoramic image of the car. Based on the real-time panoramic image of the car, computer vision technology is used to perform real-time identification and density distribution calculation of passengers to obtain the passenger position and density information in the car. In addition, through sensors installed on subway platforms and cars, historical passenger flow time series data sets are collected, and accurate real-time timestamp data is obtained through the network time protocol (NTP) synchronization service. Based on the historical passenger flow time series data set, real-time timestamp data and previously acquired air conditioning working status data, time series analysis and machine learning are used to predict passenger flow and predict the passenger flow in the next time period. According to the real-time working status data of the air conditioner and the passenger density distribution data in the carriage, combined with the first working status data and the preset second air conditioning target status data, the air conditioning parameter adjustment instruction set is generated using the preset rules. Finally, the generated adjustment instructions are transmitted to the air conditioning system control unit to realize intelligent linkage control operations and instantly adjust the operating status of the air conditioning system.

[0029] Preferably, the working state identification module S1 includes:

[0030] S11: Perform real-time multi-parameter monitoring on the target subway ventilation and air conditioning to obtain an air conditioning operation parameter set, wherein the air conditioning operation parameter set includes a compressor operation frequency parameter, an operating current parameter, an operating voltage parameter, an evaporator heat exchange efficiency parameter, and a condenser heat exchange efficiency parameter;

[0031] Specifically, a sensor can be used to monitor the operating frequency of the compressor in the target subway ventilation and air conditioning in real time. A current sensor is used to monitor the current value in the target subway ventilation and air conditioning in real time. A voltage sensor is used to monitor the voltage value in the target subway ventilation and air conditioning in real time. A temperature sensor is used to monitor the inlet and outlet temperatures of the evaporator in real time. The heat exchange efficiency of the evaporator is calculated based on the temperature difference and the air flow rate. For example, the temperature and air flow data of the evaporator are recorded once a minute, and the efficiency is calculated. A temperature sensor is used to monitor the inlet and outlet temperatures of the condenser in real time. The heat exchange efficiency of the condenser is calculated based on the temperature difference and the refrigerant flow rate. For example, the temperature and refrigerant flow data of the condenser are recorded once a minute, and the efficiency is calculated. Finally, the compressor operating frequency parameter, the operating current parameter, the operating voltage parameter, the evaporator heat exchange efficiency parameter and the condenser heat exchange efficiency parameter are obtained to form an air conditioning operating parameter set.

[0032] S12: performing power integral calculation on the target subway ventilation air conditioner according to the preset unit time according to the working current parameter and the working voltage parameter to obtain the air conditioner energy consumption data per unit time;

[0033] Specifically, the real-time power P(t) in each time period can be calculated according to the real-time working current parameter I(t) and working voltage parameter V(t) according to a preset unit time (for example, every second or every minute). For example, the power calculation formula is P(t) = I(t) × V(t). The power data in each time period is integrated to calculate the energy consumption per unit time. For example, assuming that the power integration is performed once a minute, the calculation method is Among them, t0 is the starting time point, Δt is the time interval, E is the energy consumption per unit time, and t is time, thus obtaining the air conditioner energy consumption per unit time data.

[0034] S13: If the evaporator heat exchange efficiency parameter is less than the preset heat transfer efficiency critical value, and the condenser heat exchange efficiency parameter is greater than or equal to the preset heat transfer efficiency critical value, the target subway ventilation air conditioner is marked as a cooling mode, and the air conditioner working mode data is obtained;

[0035] Specifically, a preset heat transfer efficiency critical value may be set according to the equipment manufacturer's recommendations and actual operating experience. If the evaporator heat exchange efficiency parameter is less than the preset heat transfer efficiency critical value, and the condenser heat exchange efficiency parameter is greater than or equal to the preset heat transfer efficiency critical value, the target subway ventilation air conditioner is marked as a cooling mode, and the cooling mode data is recorded as the air conditioner working mode data.

[0036] S14: If the evaporator heat exchange efficiency parameter is greater than or equal to a preset heat transfer efficiency critical value, and the condenser heat exchange efficiency parameter is less than a preset heat transfer efficiency critical value, the target subway ventilation air conditioner is marked as a heating mode, and the air conditioner working mode data is obtained;

[0037] Specifically, a preset heat transfer efficiency critical value may be set according to the equipment manufacturer's recommendations and actual operating experience. If the evaporator heat exchange efficiency parameter is greater than or equal to the preset heat transfer efficiency critical value, and the condenser heat exchange efficiency parameter is less than the preset heat transfer efficiency critical value, the target subway ventilation air conditioner is marked as a heating mode, and the heating mode data is recorded as the air conditioner working mode data.

[0038] S15: performing a multi-dimensional threshold comparison on the compressor operating frequency parameter, the air conditioner unit time energy consumption data, and a preset air conditioner operating intensity reference table to obtain air conditioner operating intensity level data, wherein the air conditioner operating intensity level data is any one of low load operating data, medium load operating data, and high load operating data;

[0039] Specifically, it can be assumed that the compressor operating frequency parameter is 2800RPM. The air conditioner energy consumption per unit time is 45kW. The preset air conditioner operating intensity reference table is as follows:

[0040] Low load operation data: The compressor operating frequency range is 2000-2500RPM, and the energy consumption range is 30-40kW.

[0041] Medium load operation data: The compressor operating frequency range is 2501-3000RPM, and the energy consumption range is 40-60kW.

[0042] High load operation data: The compressor operating frequency range is 3001-3500RPM, and the energy consumption range is 60-80kW.

[0043] Compare the compressor operating frequency parameters and air conditioner unit time energy consumption data obtained by real-time monitoring with the range in the preset air conditioner operation intensity reference table. The compressor operating frequency of 2800RPM falls into the medium load operation data range (2501-3000RPM). The air conditioner unit time energy consumption of 45kW falls into the medium load operation data range (40-60kW). Based on the above judgments, the air conditioner operation intensity level data obtained by multi-dimensional threshold comparison is medium load operation data.

[0044] S16: Determine the working state of the target subway ventilation air conditioner according to the air conditioner working mode data and the air conditioner operation intensity level data, and obtain the real-time working state data of the air conditioner, wherein the real-time working state data of the air conditioner is any one of low-load cooling state data, medium-load cooling state data, high-load cooling state data, low-load heating state data, medium-load heating state data and high-load heating state data.

[0045] Specifically, it can be assumed that the air conditioner working mode data is a cooling mode, and the air conditioner operation intensity level data is medium load operation data. Therefore, the air conditioner real-time working state data is medium load cooling state data.

[0046] The present invention provides detailed real-time operating data for the subway ventilation and air-conditioning system by comprehensively monitoring multiple key parameters, such as compressor operating frequency, operating current, and operating voltage, thereby enhancing the system's control accuracy and response speed. The energy efficiency analysis function of this module not only evaluates the energy efficiency of the air conditioner, but also helps operators optimize energy use and achieve energy conservation and emission reduction. The intelligent pattern recognition capability ensures that the air conditioner automatically switches to the most appropriate working state according to actual needs, and can quickly adapt to environmental changes whether it is cooling or heating. The load level assessment further refines the intensity of air conditioning operation, provides data support for refined control, and balances passenger comfort and energy efficiency. The real-time status feedback mechanism ensures that the intelligent decision-making execution module can be controlled according to accurate air conditioning working status data, thereby optimizing the passenger experience, ensuring that the temperature in the subway car is always in the best state, and providing a comfortable riding environment in both hot and cold modes.

[0047] Preferably, the passenger flow monitoring module S2 includes:

[0048] S21: Perform panoramic photography of the target subway carriage to obtain a real-time panoramic image of the carriage;

[0049] Specifically, a panoramic camera can be installed on the central ceiling of the target subway car to capture the entire scene in the car. The camera is set to take an image once a minute. After receiving each frame of the image, it is stored as a real-time panoramic image of the car.

[0050] S22: collecting a floor plan of the passenger area of ​​the target subway car to obtain a floor plan of the passenger area;

[0051] Specifically, the passenger area floor plan may be obtained from the subway manufacturer.

[0052] S23: Identify and annotate the passengers in the real-time panoramic image of the carriage to obtain a panoramic image annotated with passenger instances;

[0053] Specifically, the passenger detection and recognition algorithm in computer vision technology can be applied to identify and annotate passengers, such as the target detection algorithm based on deep learning (such as YOLO, Faster R-CNN). Once the passenger is detected, each passenger is instantiated and annotated. The annotation content usually includes the passenger's location coordinates, posture information, and behavior status (such as standing or sitting). Each passenger instance annotation is added to the original panoramic image to generate a passenger instance annotated panoramic image.

[0054] S24: Acquire the actual length data and the actual width data of the carriage;

[0055] Specifically, the actual length data and the actual width data of the carriage can be obtained from the subway manufacturer.

[0056] S25: performing passenger space projection mapping on the passenger area floor plan according to the passenger instance annotated panoramic image to obtain a passenger space distribution projection map;

[0057] Specifically, image processing software such as OpenCV can be used. A geometric projection algorithm is applied to map the position of each passenger in the passenger instance annotated panoramic image to the passenger area floor plan. According to the actual measured length and width data of the carriage, the passenger position is projected within the correct spatial range and is consistent with the actual spatial position on the floor plan, thereby obtaining a passenger spatial distribution projection map. The mapped passenger spatial distribution projection map shows the distribution of each passenger on the floor plane, including its relative position and density distribution.

[0058] S26: Calculate the passenger density distribution of the target subway car based on the passenger space distribution projection map according to the actual length data and the actual width data of the car, and obtain the passenger density distribution data of the car.

[0059] Specifically, please refer to sub-step S26 for the specific processing process of this embodiment.

[0060] The present invention realizes real-time monitoring without blind spots in the carriage through panoramic camera technology, provides comprehensive visual data for passenger flow analysis, and greatly enhances the breadth and depth of data collection. The application of image recognition technology makes passenger identification more accurate, and real-time labeling ensures the accurate capture of passenger quantity and distribution, thereby improving the accuracy of passenger density data. By mapping these data to the floor plan of the passenger area, a passenger space distribution projection map is generated, which not only provides an intuitive reference for further analysis of passenger density, but also makes the calculation of density distribution more accurate, and provides a decision-making basis based on actual passenger distribution for the regulation of the air-conditioning system. In addition, the passenger flow monitoring module can help operators dynamically adjust operating strategies according to changes in passengers in the carriage, especially during peak hours, and can timely strengthen air conditioning cooling or heating to effectively cope with passenger flow pressure.

[0061] Preferably, S26 includes:

[0062] S261: gridding the passenger space distribution projection map according to the actual length data and the actual width data of the carriage according to the preset passenger area grid division standard to obtain a passenger space distribution projection grid map, wherein the passenger space distribution projection grid map includes a plurality of grid areas, and one grid area represents a floor partition of the passenger area;

[0063] Specifically, the Python programming language combined with the OpenCV library can be used to load and convert the passenger space distribution projection map into a grayscale image. According to the preset passenger area grid division standard and the actual length and width data of the carriage, the number and size of the grids to be divided are calculated. Using OpenCV image processing functions such as cv2.resize() and cv2.rectangle(), the passenger space distribution projection map is pixelated, and a gridded passenger space distribution projection grid map is generated according to the preset grid division standard.

[0064] S262: performing passenger individual space modeling on the passenger space distribution projection grid map to obtain a passenger individual space model map;

[0065] Specifically, the Python programming language and related spatial analysis libraries can be used to load the passenger spatial distribution projection grid map and convert the passenger location data in each grid area into spatial coordinates. Based on the passenger location data, spatial statistical methods such as kernel density estimation (KDE) or simple point density calculation are used to perform individual spatial modeling of passengers in each grid area. The passenger individual spatial model map is output to show the distribution density and location of passengers in each grid area.

[0066] S263: performing spatial overlap analysis on the floor partitions of each passenger area based on the passenger individual space model diagram to obtain a number of regional overlap coefficients;

[0067] Specifically, GIS software can be used to load the partition boundary data in the passenger individual space model map and the passenger space distribution projection grid map. The area intersection ratio of each pair of partitions is calculated to determine the spatial overlap between them. Based on the calculation results, several regional overlap coefficients are generated to indicate the degree of spatial overlap between each pair of partitions.

[0068] S264: Counting passengers for each floor partition of the passenger area based on the passenger space distribution projection grid map to obtain passenger quantity data for a number of partitions;

[0069] Specifically, the passenger spatial distribution projection grid map can be loaded and converted into a grayscale image using the Python programming language and the OpenCV library. Appropriate thresholding methods are applied to each grid area to segment passengers from the background, and passengers are counted using the connected region analysis method. Morphological processing, such as dilation and erosion, is combined to eliminate noise and ensure accurate passenger counting. The passenger quantity data for each floor partition of the passenger loading area is output, and finally the passenger quantity data for several partitions are obtained.

[0070] S265: If the passenger quantity data of the partition is greater than or equal to the preset passenger quantity threshold, multiplying the preset first density weight parameter by the corresponding area overlap coefficient to obtain the passenger density data of the partition;

[0071] Specifically, if the partitioned passenger quantity data is greater than or equal to the preset passenger quantity threshold, the preset first density weight parameter can be set according to the design document or actual needs, and the corresponding partitioned passenger quantity data is multiplied by the corresponding area overlap coefficient to obtain the partitioned passenger density data.

[0072] S266: If the passenger quantity data of the partition is less than the preset passenger quantity threshold, multiplying the preset second density weight parameter by the corresponding area overlap coefficient to obtain the passenger density data of the partition, wherein the preset first density weight parameter is greater than the preset second density weight parameter;

[0073] Specifically, if the partitioned passenger quantity data is less than the preset passenger quantity threshold, the preset second density weight parameter can be set according to the design document or actual needs, and the corresponding partitioned passenger quantity data is multiplied by the corresponding area overlap coefficient to obtain the partitioned passenger density data.

[0074] S267: Extract partition identification from the passenger space distribution projection map to obtain a partition identification data set, wherein the partition identification data set includes a plurality of partition identification data, and spatially map and associate the partition passenger density data with the partition identification data to obtain the carriage passenger density distribution data.

[0075] Specifically, image processing and computer vision techniques can be used to extract partition identification. Such as OpenCV library and Python programming language. Load the passenger space distribution projection map, perform appropriate image segmentation and pixel classification on the passenger area to extract the identification information of each passenger area. Color segmentation, shape matching or pixel density-based methods can be used to extract partition identification, and output several partition identification data to represent the identification information of each passenger area floor partition. Match the passenger density data of each partition with its corresponding partition identification. Determine the location and range of each partition identification so that the passenger density data can be correctly mapped into the space. Use image analysis and data association methods to ensure that the passenger density data is consistent with the actual location of the passenger area. Output the passenger density distribution data of the carriage, that is, the correspondence between the passenger density data of each partition and its spatial position.

[0076] The present invention divides the carriage into multiple passenger area grids by using grid processing, laying the foundation for accurate monitoring of passenger distribution. The application of individual space modeling further enhances the understanding of the space occupied by each passenger and provides accurate individual data for evaluating passenger density. The introduction of spatial overlap analysis enables the system to accurately identify crowded areas and effectively evaluate even in high-density passenger environments. By real-time statistics of the number of passengers in each partition, data support is provided for the instant regulation of the air-conditioning system. By using different calculation methods according to the number of passengers, the degree of crowding in the carriage is reasonably reflected. The generation of partition passenger density data combines the number of passengers and spatial overlap, providing accurate environmental control data for air-conditioning regulation strategies. By combining density data with partition identification, a detailed carriage passenger density distribution map is created, providing a comprehensive view of carriage use. In summary, the present invention optimizes air-conditioning regulation, improves the overall comfort of passengers, reduces energy waste through precise regulation, improves resource utilization efficiency, and overall enhances the intelligence of subway operations and the quality of passenger experience.

[0077] Preferably, the working state adjustment module S3 includes:

[0078] S31: Count the number of partitions of the passenger density distribution data in the carriage to obtain the total number of partitions;

[0079] Specifically, the number of different partitions in the passenger density distribution data of the carriage can be counted according to the partition identification, so as to obtain the total number of partitions. The statistical method can be manual counting or automatic counting using data processing software such as Excel. For example, suppose that the passenger density distribution data is stored in a table according to the partition identification. Open the data table, observe and calculate the number of different partition identifications in the data set. For example, there are 10 different partition identifications in the carriage, each identification represents an independent passenger area, so the total number of partitions is 10.

[0080] S32: Compare the passenger density data of each partition in the carriage passenger density distribution data with the preset passenger density threshold, and record the number of data of the partition passenger density data being greater than or equal to the preset passenger density threshold, to obtain the number data of high passenger density partitions;

[0081] Specifically, a preset passenger density threshold can be set to determine high passenger density zones. Review the carriage passenger density distribution data to determine the passenger density data for each zone. Compare the passenger density of each zone with the preset passenger density threshold. Record the number of zones with passenger density greater than or equal to the preset threshold, and finally obtain the number of high passenger density zones. For example, assume that the preset passenger density threshold is 3 people per square meter. Check the passenger density distribution data and calculate that there are 5 zones with passenger density greater than or equal to 3 people / m2. 2 These five zones are marked as high passenger density zones.

[0082] S33: Calculate the density ratio of the high passenger density partition quantity data and the total partition quantity data to obtain high density partition proportion data;

[0083] Specifically, the total number of partitions obtained in step S31 can be assumed to be 10 partitions in total. Then, according to the number of high passenger density partitions obtained in step S32, it is assumed that the passenger density of 5 partitions exceeds or equals the preset threshold. The proportion of high passenger density partitions is calculated, that is, the ratio of the number of high-density partitions to the total number of partitions, and finally the high-density partition proportion data is obtained. The formula for calculating the proportion of high passenger density partitions is:

[0084] S34: Performing multi-point temperature sensing monitoring on the target subway car to obtain real-time temperature distribution data of the car;

[0085] Specifically, multiple temperature sensors can be installed in the target subway car to ensure that the sensor locations can cover different areas of the entire car. Set the sensor collection frequency and data transmission mode, for example, automatically collect temperature data at regular intervals and transmit it to the monitoring system, so as to obtain real-time temperature distribution data of the car.

[0086] S35: According to the real-time working state data of the air conditioner, the high-density partition proportion data and the real-time temperature distribution data of the carriage, the ventilation air conditioner of the target subway is regulated in a first working state to obtain the first working state data.

[0087] Specifically, please refer to sub-step S35 for the specific processing process of this embodiment.

[0088] The present invention quickly obtains the total number of compartments in the car through partition density statistics. Secondly, the identification of high-density areas enables the system to perform air conditioning control in a targeted manner to optimize passenger comfort. Density ratio analysis further provides a macroscopic indicator of congestion, making control decisions more in line with actual needs. The application of real-time temperature monitoring technology provides real-time data on temperature distribution in the car for air-conditioning control, ensuring the timeliness of control. Combining these data, the system can make more reasonable control decisions and achieve accurate adjustment of the air-conditioning working state. Intelligent control effectively avoids energy waste and improves energy efficiency, while ensuring that areas with high passenger density receive appropriate air-conditioning attention, thereby improving the overall comfort of passengers. The dynamic response capability of the present invention ensures a rapid response to environmental changes in the car and maintains the consistency of passenger experience.

[0089] Preferably, S35 includes:

[0090] S351: Calculate the average temperature of the target subway carriage according to the real-time temperature distribution data of the carriage to obtain the average temperature data inside the carriage;

[0091] Specifically, the real-time temperature distribution data of the car can be weighted averaged to obtain the average temperature data inside the car. For example, assuming that the car is divided into 5 areas, the temperature data measured are 20℃, 22℃, 24℃, 21℃, and 23℃, and the corresponding area ratio is 1:2:1:1:2, then the average temperature is calculated as: Among them, 7 is the total number of areas, and the average temperature data inside the car is finally obtained through calculation.

[0092] S352: Perform fuzzy logic matching on the high-density partition proportion data and the average temperature data inside the vehicle compartment with a preset multi-dimensional state mapping table to obtain first air conditioning target state data;

[0093] Specifically, the preset multi-dimensional state mapping table contains three states: low load, medium load and high load. The determination of these states depends on the combination of the high-density partition ratio and the average temperature. Fuzzy logic matching is performed based on the actual data obtained previously that the high-density partition ratio is 50% (high load) and the average temperature data inside the car is 18.57°C. The preset multi-dimensional state mapping table stipulates that when the high-density partition ratio is greater than or equal to 50% and the average temperature is between 18°C ​​and 20°C, the target state is high load, and high load is recorded as the first air conditioning target state data.

[0094] S353: If the first air conditioner target state data is consistent with the air conditioner real-time working state data, there is no need to perform control operations on the target subway ventilation air conditioner, and the air conditioner real-time working state data is used as the first working state data;

[0095] Specifically, the states of the first air conditioner target state data and the air conditioner real-time working state data can be compared. If the states of the two are consistent, that is, the target state data is the same as the real-time working state data, no further control operation is required. The real-time working state data of the air conditioner is directly used as the first working state data.

[0096] S354: If the first air conditioner target state data and the air conditioner real-time working state data are inconsistent, performing state difference analysis based on the air conditioner real-time working state data and the first air conditioner target state data to obtain first working state difference data;

[0097] Specifically, the states of the first air-conditioning target state data and the real-time working state data of the air-conditioning can be compared. If the states of the two are inconsistent, that is, the target state data is different from the real-time working state data, it is necessary to further analyze the specific content and impact of these differences. Appropriate analysis tools or methods can be used, such as data visualization software and statistical analysis tools. Compare and analyze the real-time working state data with the target state data to find out the cause of the state difference. Compare the real-time cabin temperature with the preset temperature. Analyze the difference between the real-time energy consumption and the expected energy consumption. Check whether the various parameters of the air-conditioning system (such as compressor operating frequency, wind speed setting) meet expectations, and generate the first working state difference data based on the analysis results;

[0098] S355: generating a conversion strategy for the target subway ventilation and air conditioning according to the first working state difference data to obtain state conversion strategy data;

[0099] Specifically, appropriate data analysis tools (such as Python programming language and related data analysis libraries) can be used to process and analyze the first working state difference data. The difference or change trend between each parameter can be calculated. Based on the results of the data analysis, a set of state transition strategies is designed to determine the conversion strategy of the target subway ventilation and air conditioning according to the difference data of the current state. For example, if the temperature rises above a certain threshold, the air conditioning cooling operation is triggered. The designed state transition strategy is converted into a machine-understandable data format. This can be a set of rules, conditional statements, or a decision function based on a machine learning model, and finally the state transition strategy data is obtained.

[0100] S356: Perform control operations on the target subway ventilation and air conditioning according to the state transition strategy data, and use the first air conditioning target state data as the first working state data.

[0101] Specifically, the policy information can be loaded from the state transition policy data. These policies include determining the control measures that need to be taken based on the current working state of the air conditioner and environmental conditions. The policy data includes, but is not limited to, logical rules, conditional statements, and threshold settings, which are used to determine when and how to adjust the working mode or parameters of the air conditioning system. According to the conditions and rules defined in the policy data, determine the current action that should be taken. According to the control operation performed, control the working state of the target subway ventilation air conditioner. According to the requirements of the strategy, adjust the parameters or working mode of the air conditioner to ensure that the air conditioner is in an appropriate state. Update the target state data of the regulated target air conditioner to the first working state data.

[0102] The present invention provides an accurate temperature reference for air conditioning control by calculating the average temperature of the compartment based on the real-time temperature distribution data of the compartment. Intelligent state matching uses fuzzy logic, combines the proportion of high-density partitions and the average temperature, and compares it with the multi-dimensional state mapping table to intelligently generate air conditioning target state data, thereby enhancing the adaptability and flexibility of control. Energy-saving optimization design reduces unnecessary control when the target state is consistent with the real-time state, thereby reducing energy consumption. Through state difference analysis and conversion strategy generation, it quickly responds to changes in the working state of the air conditioner, ensures passenger comfort and system efficiency, and improves the real-time state adjustment capability of the system. Dynamic control strategies and continuous state update mechanisms enable the system to continuously monitor and optimize the working state of the air conditioner, and maintain the continuity and stability of control.

[0103] Preferably, the working state prediction module S4 includes:

[0104] S41: collecting passenger flow time series data of the target subway car to obtain a historical passenger flow time series data set;

[0105] Specifically, passenger flow monitoring equipment, such as infrared sensors, cameras, or counters, can be installed in the target subway cars. Ensure that the equipment can accurately identify and record the data of passengers entering and leaving the cars. Set the time period and frequency of data collection, such as data records of passengers entering and leaving the cars every hour, every day, or at a specific time period. Collect passenger entry and exit information in real time through monitoring equipment, and store the data in a database or data file. Record the collected passenger flow time series data in time series to obtain a historical passenger flow time series data set.

[0106] S42: extracting subway station names from the historical passenger flow time series data set to obtain a number of subway station name data;

[0107] Specifically, assuming that each record in the historical passenger flow time series dataset contains a timestamp and the number of passengers, and the text description also contains the subway station name information, you can use Python's regular expression module to extract the subway station name from the text description, such as "Subway Station A", "Subway Station B", record the extracted data, and finally get several subway station name data.

[0108] S43: segmenting the historical passenger flow time series data set according to the subway station name data to obtain a plurality of groups of station passenger flow time series data sets, wherein a group of station passenger flow time series data sets represents a single station passenger flow time series data set;

[0109] Specifically, suppose that the historical passenger flow time series dataset contains the passenger entry and exit data of multiple subway stations, such as "Subway Station A" and "Subway Station B". You can use Python's Pandas library to group the data and split the dataset into different DataFrames according to the station name. Each DataFrame represents the passenger flow time series data of a station. The split single-station passenger flow time series datasets are stored in their respective data files or database tables, and finally several groups of station passenger flow time series datasets are obtained.

[0110] S44: constructing a passenger flow prediction model for the corresponding subway station according to the station passenger flow time series data set to obtain a station passenger flow prediction model, and traversing several groups of station passenger flow time series data sets to obtain a station passenger flow prediction model set;

[0111] Specifically, please refer to sub-step S44 for the specific processing process of this embodiment.

[0112] S45: Obtain real-time timestamp data;

[0113] Specifically, the Python programming language can be used as an example to obtain real-time timestamp data through the datetime module.

[0114] S46: Predicting a target working state of the target subway ventilation air conditioner according to the station passenger flow prediction model set, the real-time timestamp data and the first working state data, and obtaining second air conditioner target state data.

[0115] Specifically, please refer to sub-step S46 for the specific processing process of this embodiment.

[0116] The present invention provides a rich data foundation for the system through historical data analysis, making the identification and understanding of passenger flow patterns more accurate. The refined passenger flow prediction technology constructs customized prediction models for different subway stations, improving the pertinence and effectiveness of the prediction service. The construction of the model set further enhances the system's responsiveness to changes in passenger flow at different stations, ensuring the comprehensiveness and adaptability of the prediction. The real-time data fusion technology ensures the continuous updating of the prediction model, improving the timeliness and accuracy of the prediction results. The air conditioning working status prediction function provides a scientific basis for system regulation and enhances the predictability of regulation. The present invention significantly improves energy efficiency and reduces energy waste by rationally allocating air conditioning resources. In addition, the passenger flow prediction model's ability to cope with peak hours and its enhanced adaptive capabilities have significantly improved the intelligence level of the system.

[0117] Preferably, S44 includes:

[0118] S441: performing time series segmentation on the site passenger flow time series data set according to date to obtain a plurality of groups of passenger flow time series data sets with the same date;

[0119] Specifically, you can use the pandas library in Python to load the site passenger flow time series dataset. Make sure the timestamp column is of datetime type to facilitate date segmentation. Split the time series according to the date to obtain several groups of passenger flow time series datasets with the same date.

[0120] S442: Calculate the time span of the passenger flow time series data set on the same date to obtain passenger flow time span data;

[0121] Specifically, Python and the pandas library can be used to calculate the time span. For each passenger flow time series data set on the same date, the time span between the earliest and latest timestamps is calculated to obtain the passenger flow time span data.

[0122] S443: Divide the passenger flow time span data into time periods according to a preset time granularity to obtain a plurality of passenger flow time period data;

[0123] Specifically, a preset time granularity, such as 30 minutes or 1 hour, may be determined. According to the preset time granularity, the passenger flow time span data is divided into several time periods using Python and pandas libraries, thereby obtaining several passenger flow time period data.

[0124] S444: Based on the passenger flow time series data set on the same date, total passenger flow statistics are performed for the time period corresponding to the passenger flow time period data to obtain total passenger flow data, and the passenger flow time period data is used to time-mark the total passenger flow data to obtain marked total passenger flow data;

[0125] Specifically, data processing software (such as Excel) or programming tools (such as Python's pandas library) can be used to load the passenger flow time series data set and passenger flow time period data on the same date. The passenger flow data on the same date are grouped according to the pre-divided time periods, and the total passenger flow in each time period is calculated to obtain the total passenger flow data. The calculated total passenger flow data is associated with the corresponding time period, and a time label is added to the total passenger flow data for each time period to obtain the labeled total passenger flow data.

[0126] S445: traverse several groups of passenger flow time series data sets on the same date, execute S442-S444, and obtain a labeled passenger flow total data set, wherein the labeled passenger flow total data set includes several labeled passenger flow total data;

[0127] Specifically, the data sets of each date are traversed one by one, and processed according to the time series segmentation method of S442, the time period division method of S443 and the total passenger flow statistics and labeling method of S444, and the labeled total passenger flow data obtained from each date are aggregated into a comprehensive data set to form a labeled total passenger flow data set.

[0128] S446: Use the labeled total passenger flow data set to perform model training on a preset neural network architecture to obtain a site passenger flow prediction model.

[0129] Specifically, a deep learning framework (such as TensorFlow or PyTorch) can be used as a preset neural network architecture for model training. The time period data and total passenger flow data in the labeled total passenger flow data set are sorted as the input features and labels required for model training. A neural network with an input layer, several hidden layers, and an output layer is designed, and appropriate activation functions and loss functions are selected. The prepared data is input into the neural network for training to obtain a station passenger flow prediction model.

[0130] The present invention improves the efficiency and accuracy of analyzing data on the same date through the implementation of time series analysis. The identification of the time span of passenger flow and the division of time periods provide a detailed time framework for passenger flow management, which helps the system identify peak and trough periods. The statistics and time annotation of the total passenger flow enrich the historical data, providing data support for the training of the prediction model. By constructing a data set containing the annotated total passenger flow, the model training is optimized, so that the site passenger flow prediction model can learn and predict the passenger flow pattern more accurately, significantly improving the prediction accuracy. In addition, the adaptability of the model has been enhanced, and it can adapt to the changes in passenger flow on different dates and time periods, and better serve the actual operational needs.

[0131] Preferably, S46 includes:

[0132] S461: Acquire target train track data, wherein the target train track data includes a plurality of subway station name data, and the target train is the train where the target subway car is located;

[0133] Specifically, the train's running track and stop sites can be recorded in real time by a train control system or a GPS device installed on the train. The recorded data is exported, and subway station name data is extracted from it. The extracted subway station name data is sorted and deduplicated to form a target train track data list.

[0134] S462: Dynamic state monitoring of the target subway car is performed to obtain dynamic state data of the car;

[0135] Specifically, necessary sensors, such as acceleration sensors and speed sensors, can be installed in the carriage to ensure that the dynamic state of the carriage can be monitored in real time. The dynamic state data of the carriage (such as acceleration, speed and position) is collected through the sensors to obtain the dynamic state data of the carriage.

[0136] S463: If the dynamic state data of the carriage shows that the target subway carriage is in a moving state, then the name data of the subway station of the previous station is obtained; if the dynamic state data of the carriage shows that the target subway carriage is in a stationary state, then the name data of the current subway station is obtained;

[0137] Specifically, by analyzing the dynamic state data of the car (such as acceleration and speed), it can be determined whether the car is in a moving state or a stationary state. If the acceleration and speed data are large, it means that the car is in a moving state; if the acceleration and speed data are close to zero, it means that the car is in a stationary state. If the car is in a moving state, the name data of the subway station where it stopped most recently before the current time point is extracted from the trajectory data as the name data of the previous subway station. If the car is in a stationary state, the name data of the subway station corresponding to the current time point is extracted from the trajectory data as the name data of the current subway station.

[0138] S464: extracting the name of the next subway station from the target subway car based on the target train trajectory data and the previous subway station name data or the current subway station name data, and obtaining the next subway station name data;

[0139] Specifically, the track recording system can be used to query the current direction of the train, and by combining the target train track data, the previous station or current station where the train stopped can be determined by comparing the track data at the current time point. According to the train's direction of travel and the route map, the name of the next station of the current station or previous station can be found.

[0140] S465: extracting a corresponding model from the station passenger flow prediction model set according to the next station subway station name data to obtain a target station passenger flow prediction model;

[0141] Specifically, a database query tool can be used to search for a prediction model that matches the name of the next subway station in the site passenger flow prediction model set, extract the target site passenger flow prediction model from the database, and load it into the system.

[0142] S466: Input the real-time timestamp data into the target site passenger flow prediction model to perform passenger flow prediction, and obtain the site predicted passenger flow data;

[0143] Specifically, the real-time timestamp data can be input into the target site passenger flow prediction model through an API interface or a data transmission tool. The prediction model is called to perform calculations, and the site passenger flow prediction results are generated based on the input timestamp data. The output results of the prediction model are extracted to obtain the site predicted passenger flow data.

[0144] S467: monitoring the real-time passenger flow of the target subway carriage to obtain the real-time passenger flow data of the carriage, and adding the station predicted passenger flow data and the real-time passenger flow data of the carriage to obtain the total passenger flow prediction data of the carriage;

[0145] Specifically, the passenger flow in the carriage can be continuously monitored by a real-time passenger flow monitoring device installed in the carriage. The real-time passenger flow data of the carriage is extracted from the monitoring device. The station predicted passenger flow data and the real-time passenger flow data of the carriage are added to obtain the total passenger flow prediction data of the carriage.

[0146] S468: Match the total passenger flow prediction data of the carriage with the preset air-conditioning working state threshold table to obtain the second air-conditioning target state data.

[0147] Specifically, a preset air conditioning working state threshold table can be loaded from the database. The threshold table contains the air conditioning working states corresponding to different passenger flows. Using a database query system or a matching algorithm, the total passenger flow prediction data of the carriage is input into the threshold table for matching search. The search process includes traversing the threshold table to find the closest match to the total passenger flow prediction data of the carriage. Based on the result of the matching search, the corresponding air conditioning working state is determined, and the second air conditioning target state data is generated. For example, if the total passenger flow prediction data of the carriage falls within the threshold range of the medium load cooling state, the medium load cooling state is generated as the second air conditioning target state data.

[0148] The present invention enables the system to grasp the train operation status and passing stations in real time through precise trajectory tracking, providing detailed background information for passenger flow prediction. Dynamic state monitoring gives the system the ability to distinguish between the dynamic and static states of the carriages, ensuring that the data processing strategy is appropriate. Intelligent identification of boarding and alighting stations and matching with the target station prediction model further improve the accuracy and relevance of the prediction. Through real-time passenger flow prediction and data integration technology, combined with real-time timestamps, the system is provided with a comprehensive and close-to-actual passenger flow prediction, laying the foundation for instant regulation. Through intelligent matching of the air-conditioning working state, the timeliness and effectiveness of air-conditioning regulation are optimized based on the predicted passenger flow and preset thresholds. In summary, the present invention enhances the system's ability to respond to changes in train status and fluctuations in passenger flow. In addition, through precise prediction and intelligent regulation, the system optimizes energy utilization, reduces waste, and improves energy efficiency, reflecting the dual considerations of environmental friendliness and cost-effectiveness.

[0149] Preferably, the intelligent decision execution module S5 includes:

[0150] S51: performing state difference analysis on the first working state data and the second air conditioning target state data to obtain second working state difference data;

[0151] Specifically, the first working state data and the second air conditioner target state data can be loaded using Python's Pandas library, and the first working state data and the second air conditioner target state data can be compared item by item. A difference analysis algorithm is applied to calculate the difference value of each data item to obtain the second working state difference data.

[0152] S52: Identify the equipment adjustment requirement according to the second working state difference data, and obtain the name data of the equipment to be adjusted;

[0153] Specifically, thresholds and rules for identifying adjustment requirements can be set according to system requirements and device performance. A rule engine or a logic judgment algorithm is used to identify the device that needs to be adjusted according to the second working state difference data. According to the identification result, the corresponding device name is matched from the device information database.

[0154] S53: generating a parameter adjustment strategy according to the second working state difference data and the name data of the device to be adjusted to obtain a state adjustment parameter strategy;

[0155] Specifically, by analyzing the second working state difference data and the name data of the device to be adjusted, the specific device that needs to be adjusted and the direction and amplitude of the adjustment can be identified. For example, according to the difference data, the air-conditioning temperature exceeds the threshold, and the condenser and compressor need to be adjusted. Set a specific adjustment strategy, such as increasing the heat removal efficiency of the condenser and adjusting the operating frequency of the compressor to quickly reduce the cabin temperature. Generate a detailed adjustment plan, including the equipment operation sequence and parameter adjustment range, to obtain the state adjustment parameter strategy.

[0156] S54: Serializing the state adjustment parameter strategy into instructions to obtain an air conditioning parameter adjustment instruction set;

[0157] Specifically, each device adjustment operation involved in the state adjustment parameter strategy, such as the adjustment parameters of the condenser and compressor, can be converted into instructions that the computer can understand. The JSON format is selected as an example to ensure that all adjustment instructions are clearly encoded and recorded. A complete instruction set containing all adjustment instructions is generated, including the specific adjustment values ​​and operation sequence of each device, and finally the air conditioning parameter adjustment instruction set is obtained.

[0158] S55: Perform intelligent actuator linkage control operations on the target subway ventilation and air conditioning according to the air conditioning parameter adjustment instruction set.

[0159] Specifically, the air conditioning parameter adjustment instruction set can be sent to the air conditioning system control unit using a suitable communication protocol, such as through a local area network or a dedicated communication line. The control unit converts the received air conditioning parameter adjustment instruction set into an executable control signal according to a preset parsing algorithm. According to the operation steps and adjustment parameters described in detail in the instruction, the actuator inside the air conditioning system is controlled to perform the corresponding adjustment operation.

[0160] The present invention can quickly identify the difference between the working state and the target state through state difference analysis, providing an accurate basis for decision-making. The pertinence and efficiency of the adjustment measures are ensured through precise equipment adjustment identification. The intelligent parameter adjustment strategy reduces human operation and improves the scientificity and rationality of the adjustment. The orderly execution of the adjustment instructions is ensured through instruction serialization, avoiding execution confusion. The linkage control operation realizes precise control of the air-conditioning system through the intelligent actuator, improving the response speed and adjustment accuracy. The application of these technologies optimizes energy management and reduces energy waste, while improving passenger experience and ensuring suitable temperature and environment in different compartment areas. The stability and reliability of the system are improved through orderly instruction execution and precise equipment adjustment. Through automated control, human intervention is reduced, operational errors are reduced, and efficient and reliable control is achieved. Through continuous learning ability, it can continuously optimize the adjustment strategy according to historical data, strengthen the closed-loop control of prediction and response, and ensure the efficient coordination of the entire system.

[0161] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0162] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A subway ventilation and air conditioning monitoring system, characterized in that: Includes the following modules: The working state identification module S1 is used to perform multi-parameter real-time monitoring of the target subway ventilation and air conditioning to obtain an air conditioning operation parameter set; the working state of the target subway ventilation and air conditioning is identified according to the air conditioning operation parameter set to obtain real-time working state data of the air conditioning; The passenger flow monitoring module S2 is used to perform panoramic photography of the target subway car to obtain a real-time panoramic image of the car; calculate the passenger density distribution of the target subway car according to the real-time panoramic image of the car to obtain passenger density distribution data of the car, wherein the target subway car is the subway car where the target subway ventilation and air conditioning is located; The working state adjustment module S3 is used to adjust the first working state of the target subway ventilation air conditioner according to the real-time working state data of the air conditioner and the passenger density distribution data of the carriage to obtain the first working state data; The working state prediction module S4 is used to collect passenger flow time series data of the target subway car to obtain a historical passenger flow time series data set; obtain real-time timestamp data; and predict the target working state of the target subway ventilation air conditioner based on the historical passenger flow time series data set according to the real-time timestamp data and the first working state data to obtain the second air conditioner target state data; The intelligent decision-making execution module S5 is used to generate a parameter adjustment instruction set for the target subway ventilation and air conditioning according to the first working state data and the second air conditioning target state data to obtain an air conditioning parameter adjustment instruction set; and perform intelligent actuator linkage control operations on the target subway ventilation and air conditioning according to the air conditioning parameter adjustment instruction set.

2. The subway ventilation and air conditioning monitoring system according to claim 1, characterized in that: The working state identification module S1 includes: S11: Performing real-time multi-parameter monitoring on the target subway ventilation and air conditioning to obtain an air conditioning operation parameter set, wherein the air conditioning operation parameter set includes a compressor operation frequency parameter, an operating current parameter, an operating voltage parameter, an evaporator heat exchange efficiency parameter, and a condenser heat exchange efficiency parameter; S12: performing power integral calculation on the target subway ventilation air conditioner according to the preset unit time according to the working current parameter and the working voltage parameter to obtain the air conditioner energy consumption data per unit time; S13: If the evaporator heat exchange efficiency parameter is less than the preset heat transfer efficiency critical value, and the condenser heat exchange efficiency parameter is greater than or equal to the preset heat transfer efficiency critical value, the target subway ventilation air conditioner is marked as a cooling mode, and the air conditioner working mode data is obtained; S14: If the evaporator heat exchange efficiency parameter is greater than or equal to a preset heat transfer efficiency critical value, and the condenser heat exchange efficiency parameter is less than a preset heat transfer efficiency critical value, the target subway ventilation air conditioner is marked as a heating mode, and the air conditioner working mode data is obtained; S15: performing a multi-dimensional threshold comparison on the compressor operating frequency parameter, the air conditioner unit time energy consumption data, and a preset air conditioner operating intensity reference table to obtain air conditioner operating intensity level data, wherein the air conditioner operating intensity level data is any one of low load operating data, medium load operating data, and high load operating data; S16: Determine the working state of the target subway ventilation air conditioner according to the air conditioner working mode data and the air conditioner operation intensity level data, and obtain the real-time working state data of the air conditioner, wherein the real-time working state data of the air conditioner is any one of low-load cooling state data, medium-load cooling state data, high-load cooling state data, low-load heating state data, medium-load heating state data and high-load heating state data.

3. The subway ventilation and air conditioning monitoring system according to claim 1, characterized in that: The passenger flow monitoring module S2 includes: S21: Perform panoramic photography of the target subway carriage to obtain a real-time panoramic image of the carriage; S22: collecting a floor plan of the passenger area of ​​the target subway car to obtain a floor plan of the passenger area; S23: Identify and annotate the passengers in the real-time panoramic image of the carriage to obtain a panoramic image annotated with passenger instances; S24: Acquire the actual length data and the actual width data of the carriage; S25: performing passenger space projection mapping on the passenger area floor plan according to the passenger instance annotated panoramic image to obtain a passenger space distribution projection map; S26: Calculate the passenger density distribution of the target subway car based on the passenger space distribution projection map according to the actual length data and the actual width data of the car, and obtain the passenger density distribution data of the car.

4. The subway ventilation and air conditioning monitoring system according to claim 3 is characterized in that: S26 includes: S261: gridding the passenger space distribution projection map according to the actual length data and the actual width data of the carriage according to the preset passenger area grid division standard to obtain a passenger space distribution projection grid map, wherein the passenger space distribution projection grid map includes a plurality of grid areas, and one grid area represents a floor partition of the passenger area; S262: performing passenger individual space modeling on the passenger space distribution projection grid map to obtain a passenger individual space model map; S263: performing spatial overlap analysis on the floor partitions of each passenger area based on the passenger individual space model diagram to obtain a number of regional overlap coefficients; S264: Counting passengers for each floor partition of the passenger area based on the passenger space distribution projection grid map to obtain passenger quantity data for a number of partitions; S265: If the passenger quantity data of the partition is greater than or equal to the preset passenger quantity threshold, multiplying the preset first density weight parameter by the corresponding area overlap coefficient to obtain the passenger density data of the partition; S266: If the passenger quantity data of the partition is less than the preset passenger quantity threshold, multiplying the preset second density weight parameter by the corresponding area overlap coefficient to obtain the passenger density data of the partition, wherein the preset first density weight parameter is greater than the preset second density weight parameter; S267: Extract partition identification from the passenger space distribution projection map to obtain a partition identification data set, wherein the partition identification data set includes a plurality of partition identification data, and spatially map and associate the partition passenger density data with the partition identification data to obtain the carriage passenger density distribution data.

5. The subway ventilation and air conditioning monitoring system according to claim 1, characterized in that: The working state adjustment module S3 includes: S31: Count the number of partitions of the passenger density distribution data in the carriage to obtain the total number of partitions; S32: Compare the passenger density data of each partition in the carriage passenger density distribution data with the preset passenger density threshold, and record the number of data of the partition passenger density data being greater than or equal to the preset passenger density threshold, to obtain the number data of high passenger density partitions; S33: Calculate the density ratio of the high passenger density partition quantity data and the total partition quantity data to obtain high density partition proportion data; S34: Performing multi-point temperature sensing monitoring on the target subway car to obtain real-time temperature distribution data of the car; S35: According to the real-time working state data of the air conditioner, the high-density partition proportion data and the real-time temperature distribution data of the carriage, the ventilation air conditioner of the target subway is regulated in a first working state to obtain the first working state data.

6. The subway ventilation and air conditioning monitoring system according to claim 5, characterized in that: S35 includes: S351: Calculate the average temperature of the target subway carriage according to the real-time temperature distribution data of the carriage to obtain the average temperature data inside the carriage; S352: Perform fuzzy logic matching on the high-density partition proportion data and the average temperature data inside the vehicle compartment with a preset multi-dimensional state mapping table to obtain first air conditioning target state data; S353: If the first air conditioner target state data is consistent with the air conditioner real-time working state data, there is no need to perform control operations on the target subway ventilation air conditioner, and the air conditioner real-time working state data is used as the first working state data; S354: If the first air conditioner target state data and the air conditioner real-time working state data are inconsistent, performing state difference analysis based on the air conditioner real-time working state data and the first air conditioner target state data to obtain first working state difference data; S355: generating a conversion strategy for the target subway ventilation and air conditioning according to the first working state difference data to obtain state conversion strategy data; S356: Perform control operations on the target subway ventilation and air conditioning according to the state transition strategy data, and use the first air conditioning target state data as the first working state data.

7. The subway ventilation and air conditioning monitoring system according to claim 1, characterized in that: The working state prediction module S4 includes: S41: collecting passenger flow time series data of the target subway car to obtain a historical passenger flow time series data set; S42: extracting subway station names from the historical passenger flow time series data set to obtain a number of subway station name data; S43: segmenting the historical passenger flow time series data set according to the subway station name data to obtain a plurality of groups of station passenger flow time series data sets, wherein a group of station passenger flow time series data sets represents a single station passenger flow time series data set; S44: constructing a passenger flow prediction model for the corresponding subway station according to the station passenger flow time series data set to obtain a station passenger flow prediction model, and traversing several groups of station passenger flow time series data sets to obtain a station passenger flow prediction model set; S45: Obtain real-time timestamp data; S46: Predicting a target working state of the target subway ventilation air conditioner according to the station passenger flow prediction model set, the real-time timestamp data and the first working state data, and obtaining second air conditioner target state data.

8. The subway ventilation and air conditioning monitoring system according to claim 7, characterized in that: S44 includes: S441: performing time series segmentation on the site passenger flow time series data set according to date to obtain a plurality of groups of passenger flow time series data sets with the same date; S442: Calculate the time span of the passenger flow time series data set on the same date to obtain passenger flow time span data; S443: Divide the passenger flow time span data into time periods according to a preset time granularity to obtain a plurality of passenger flow time period data; S444: Based on the passenger flow time series data set on the same date, total passenger flow statistics are performed for the time period corresponding to the passenger flow time period data to obtain total passenger flow data, and the passenger flow time period data is used to time-mark the total passenger flow data to obtain marked total passenger flow data; S445: traverse several groups of passenger flow time series data sets on the same date, execute S442-S444, and obtain a labeled passenger flow total data set, wherein the labeled passenger flow total data set includes several labeled passenger flow total data; S446: Use the labeled total passenger flow data set to perform model training on a preset neural network architecture to obtain a site passenger flow prediction model.

9. The subway ventilation and air conditioning monitoring system according to claim 7, characterized in that: S46 includes: S461: Acquire target train track data, wherein the target train track data includes a plurality of subway station name data, and the target train is the train where the target subway car is located; S462: Dynamic state monitoring of the target subway car is performed to obtain dynamic state data of the car; S463: If the dynamic state data of the carriage shows that the target subway carriage is in a moving state, then the name data of the subway station of the previous station is obtained; if the dynamic state data of the carriage shows that the target subway carriage is in a stationary state, then the name data of the current subway station is obtained; S464: extracting the name of the next subway station from the target subway car based on the target train trajectory data and the previous subway station name data or the current subway station name data, and obtaining the next subway station name data; S465: extracting a corresponding model from the station passenger flow prediction model set according to the next station subway station name data to obtain a target station passenger flow prediction model; S466: Input the real-time timestamp data into the target site passenger flow prediction model to perform passenger flow prediction, and obtain the site predicted passenger flow data; S467: monitoring the real-time passenger flow of the target subway carriage to obtain the real-time passenger flow data of the carriage, and adding the station predicted passenger flow data and the real-time passenger flow data of the carriage to obtain the total passenger flow prediction data of the carriage; S468: Match the total passenger flow prediction data of the carriage with the preset air-conditioning working state threshold table to obtain the second air-conditioning target state data.

10. The subway ventilation and air conditioning monitoring system according to claim 1, characterized in that: The intelligent decision execution module S5 includes: S51: performing state difference analysis on the first working state data and the second air conditioning target state data to obtain second working state difference data; S52: Identify the equipment adjustment requirement according to the second working state difference data, and obtain the name data of the equipment to be adjusted; S53: generating a parameter adjustment strategy according to the second working state difference data and the name data of the device to be adjusted to obtain a state adjustment parameter strategy; S54: Serializing the state adjustment parameter strategy into instructions to obtain an air conditioning parameter adjustment instruction set; S55: Perform intelligent actuator linkage control operations on the target subway ventilation and air conditioning according to the air conditioning parameter adjustment instruction set.

Images