A remote intelligent monitoring method for the health status of rail transit vehicle air conditioning systems

Abstract

The present invention discloses a remote intelligent monitoring method for the health status of rail transit vehicle air-conditioning systems, which relates to the field of intelligent control technology. The method initializes the rail transit air-conditioning system and collects data in real time, which is then sent to an edge computing device via wireless communication. After the edge computing preprocesses the data, it compares it with preset parameters to calculate the current state of the air-conditioning system. Based on this state, a physical model is constructed to perform fault deduction. The operation of the air-conditioning system is simulated through heat load prediction and pressure change analysis, and the fault deduction report is generated by comparing it with real-time data. The report is input into a machine learning algorithm for analysis, and the fault type and source are determined based on historical fault data and the operating mode of the air-conditioning system. Combined with remote intelligent adaptive control, the system parameters are optimized and the air-conditioning system status is updated. The present invention integrates real-time data collection, modeling deduction, and machine learning to improve the accuracy and response speed of fault diagnosis, thereby enhancing the reliability and intelligence of rail transit air-conditioning systems.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a method for remote intelligent monitoring of the health status of an air-conditioning system of a rail transit vehicle. Background Art

[0002] With the development of rail transit, air conditioning systems have become important for ensuring passenger comfort and stable operation of rail transit vehicles. However, existing air conditioning systems mainly rely on periodic manual inspections or simple fault alarms, making it difficult to achieve comprehensive monitoring. Especially in complex high and low temperature environments, equipment aging or sudden failures may cause the air conditioning system to lose efficiency and shorten its service life.

[0003] In particular, traditional technologies for fault prediction and source location in air conditioning systems fail to fully account for the dynamic and volatile nature of the system, resulting in poor fault prediction accuracy and an inability to dynamically adjust early warning models based on real-time data. This static approach prevents the air conditioning system from quickly responding to and effectively repairing complex or sudden faults, impacting system stability and passenger comfort. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a remote intelligent monitoring method for the health status of rail transit vehicle air-conditioning systems to solve the problems of real-time fault diagnosis and precise control of rail transit air-conditioning systems.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a remote intelligent monitoring method for the health status of a rail transit vehicle air conditioning system, which includes initializing the rail transit air conditioning system and collecting data in real time, and sending the data to edge computing using wireless communication;

[0008] Pre-process the received data, compare it with the preset working parameters, and calculate the current working status of the air conditioning system;

[0009] Build a physical model to simulate the current working status of the air conditioning system, simulate the operation of the air conditioning system through heat load prediction and pressure change analysis, compare it with real-time data, and generate a fault simulation report;

[0010] The fault simulation report is input into the machine learning algorithm for analysis, and the fault type and source are determined based on historical fault data and the operating mode of the air conditioning system;

[0011] Through analysis using machine learning algorithms and combined with remote intelligent adaptive control, the current working status of the air-conditioning system is updated.

[0012] As a preferred solution of the method for remote intelligent monitoring of the health status of the rail transit vehicle air-conditioning system of the present invention, the rail transit air-conditioning system is initialized and data is collected in real time, and sent to edge computing using wireless communication, including:

[0013] The air-conditioning system is initialized using an embedded control platform. The temperature, humidity and refrigerant data are collected in real time through distributed sensors and stored in the control unit cache, and then transmitted to edge computing using narrowband Internet of Things.

[0014] As a preferred solution of the method for remote intelligent monitoring of the health status of rail transit vehicle air-conditioning systems according to the present invention, the received data is pre-processed and compared with preset operating parameters to calculate the current operating status of the air-conditioning system. The specific steps are as follows:

[0015] Normalize temperature, humidity, and refrigerant data to extract the preset operating parameters built into edge computing;

[0016] Deviation calculation is used to output deviation results and mark anomalies, and the health status and abnormal status of the air-conditioning system are evaluated in combination with parameter weights.

[0017] As a preferred embodiment of the method for remote intelligent monitoring of the health status of rail transit vehicle air-conditioning systems of the present invention, a physical model is constructed to simulate the current working status of the air-conditioning system, and the operation of the air-conditioning system is simulated by heat load prediction and pressure change analysis, and the system operation is compared with real-time data to generate a fault simulation report. The specific steps are as follows:

[0018] Build a dynamic physical model, couple the heat load and refrigerant flow changes, and generate a dynamic temperature prediction model;

[0019] Update dynamic temperature prediction parameters through real-time data, correct changes in heat load and refrigerant pressure, and optimize the dynamic temperature prediction model in real time;

[0020] Combined with real-time temperature data collection, the temperature deviation trend is analyzed, the operating status is determined by the refrigerant pressure change, the fault is classified and located, and a fault deduction report is generated.

[0021] As a preferred solution of the method for remote intelligent monitoring of the health status of rail transit vehicle air-conditioning systems according to the present invention, a dynamic physical model is constructed to couple the heat load and refrigerant flow changes to generate a dynamic temperature prediction model. The specific steps are as follows:

[0022] By identifying the heat load, refrigerant pressure and refrigerant flow rate under the current working state, combined with the air specific heat capacity and the cabin volume, the core thermodynamic relationship of cabin temperature change is constructed;

[0023] Based on the core relationship of thermodynamics, by dynamically balancing heat input, output and loss, a dynamic integral form is used to quantify the temperature change trend over time, and a dynamic temperature prediction model is constructed.

[0024] As a preferred solution of the method for remote intelligent monitoring of the health status of rail transit vehicle air-conditioning systems of the present invention, the following specific steps are used to analyze the temperature deviation trend by combining real-time temperature data collection, determine the operating status by changes in refrigerant pressure, classify and locate faults, and generate a fault deduction report:

[0025] Based on the dynamic temperature prediction model, the real-time collected data is compared with the predicted value to deduce the fault type and display the abnormal points through the time series graph;

[0026] Use the fault type results to evaluate the air conditioning system status, classify and summarize the fault types and causes, combine multi-parameter trend charts to display the fault evolution process and abnormal points, and generate a fault deduction report.

[0027] As a preferred solution of the remote intelligent monitoring method for the health status of the rail transit vehicle air-conditioning system described in the present invention, the fault deduction report is input into the machine learning algorithm for analysis, and the fault type and fault source are determined based on historical fault data and the operating mode of the air-conditioning system, including: extracting fault parameters in the fault deduction report and normalizing them to generate multidimensional feature vectors, and using a dynamic physical model to determine the fault type and fault source classification.

[0028] As a preferred solution of the method for remote intelligent monitoring of the health status of the rail transit vehicle air-conditioning system of the present invention, the method includes: analyzing the current working status of the air-conditioning system by a machine learning algorithm and combining remote intelligent adaptive control, including:

[0029] The fault type and source are analyzed through machine learning models, and the control strategy is generated and adjustment instructions are output based on the current working status parameters. The expected value evaluation results are monitored and compared in real time to iterate the control strategy.

[0030] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the remote intelligent monitoring method for the health status of a rail transit vehicle air-conditioning system as described in the first aspect of the present invention is implemented.

[0031] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the remote intelligent monitoring method for the health status of a rail transit vehicle air-conditioning system as described in the first aspect of the present invention is implemented.

[0032] The beneficial effects of the present invention are as follows: By combining dynamic physical modeling with machine learning, the present invention achieves real-time analysis and optimized control. Distributed sensors and edge computing are used to improve the accuracy of temperature, humidity, and refrigerant data acquisition and the real-time transmission. Dynamic temperature prediction models and heat load-refrigerant flow coupling analysis are used to accurately deduce fault types and sources. Machine learning deeply mines historical fault data, combined with real-time monitoring, to achieve efficient fault classification and health assessment. Ultimately, the integration of real-time acquisition, modeling deduction, and machine learning improves the accuracy and response speed of fault diagnosis, enhancing the reliability and intelligence of rail transit air-conditioning systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 This is a flow chart of the method for remote intelligent monitoring of the health status of the rail transit vehicle air-conditioning system in Example 1. DETAILED DESCRIPTION

[0035] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0036] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0037] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0038] Example 1, reference Figure 1 , which is the first embodiment of the present invention, provides a method for remote intelligent monitoring of the health status of a rail transit vehicle air-conditioning system, comprising the following steps:

[0039] S1. Initialize the rail transit air conditioning system and collect data in real time, and send it to edge computing using wireless communication.

[0040] Specifically, the air-conditioning system is first initialized using an embedded control platform to automatically load preset operating parameters and dynamically adjust them based on environmental conditions to reduce dependence on the central controller and improve the adaptability and stability of the air-conditioning system.

[0041] Subsequently, real-time data collection is performed based on a distributed sensor architecture. Temperature, humidity, and refrigerant parameters are collected through multi-point sensor deployment and stored in the control unit cache. Redundant design enhances the fault tolerance of the air conditioning system, preventing single-point failures from impacting overall monitoring. The collected data is transmitted to the edge computing unit via Narrowband Internet of Things (NB-IoT). Compared to traditional Wi-Fi or 4G, this communication technology has stronger anti-interference capabilities and low power consumption in rail transit environments. It can still transmit stably even at high speeds or in weak signal areas, and supports large-scale device access, improving the expansion of the monitoring air conditioning system.

[0042] S2. Pre-process the received data, compare it with the preset working parameters, and calculate the current working status of the air conditioning system.

[0043] Specifically, normalization processing is used to eliminate the dimensional differences in temperature, humidity and refrigerant data, so that all types of data are at the same scale. The air-conditioning system compares the preset working parameter library (temperature weight, humidity weight and refrigerant pressure weight) with the real-time collected data to adapt to the differences in different vehicle models, environmental conditions and operating conditions.

[0044] Deviation calculation is used to output deviation results and mark anomalies, and the health status and abnormal status of the air-conditioning system are evaluated in combination with parameter weights.

[0045] Specifically, by introducing deviation calculation, the system calculates the deviation between real-time data and preset parameters and analyzes it in conjunction with operational trends, thereby reducing misjudgments caused by short-term fluctuations and improving the accuracy of anomaly identification. Finally, the deviation calculation results are marked as abnormal. When the deviation exceeds the normal range, the anomaly is promptly marked for subsequent analysis and health assessment.

[0046] Among them, the health status and abnormal status of the air conditioning system are evaluated by combining parameter weights as follows:

[0047]

[0048] Where w i ′ is the weight after evaluation, w i is the initial parameter weight, σ i Real-time data for air conditioning system, To normalize the dynamic weights so that they sum to 1, w j is the historical fault parameter weight, σ j Dynamically adjust values ​​based on real-time data of the air conditioning system.

[0049] S=w′ T ·Te+w P ′·Pe+w′ H He;

[0050] Where S is the health score, w′ T is the dynamic weight of the temperature parameter, w P ′ is the dynamic weight of the refrigerant pressure parameter, w′ H is the dynamic weight of the humidity parameter, Te is the temperature deviation value, Pe is the refrigerant pressure deviation value, and He is the humidity deviation value.

[0051] The air conditioning system status judgment threshold is set as S based on the health score d :

[0052] If S>S d When , it means the air conditioning system is in abnormal state;

[0053] If S is less than S d When , it means the air conditioning system is in a healthy state.

[0054] S3. Build a physical model to simulate the fault of the current working status of the air-conditioning system. Simulate the operation of the air-conditioning system through heat load prediction and pressure change analysis, compare it with real-time data, and generate a fault simulation report.

[0055] Specifically, by identifying the heat load, refrigerant pressure and refrigerant flow rate under the current working state, combined with the air specific heat capacity and the cabin volume, the core thermodynamic relationship of the cabin temperature change is constructed;

[0056] Based on the core relationship of thermodynamics, by dynamically balancing heat input, output and loss, a dynamic integral form is used to quantify the temperature change trend over time, and a dynamic temperature prediction model is constructed.

[0057] The heat load is expressed as:

[0058] Q=m·s p ·(T in -T out );

[0059] Where Q is the heat load of the air conditioning system, m is the refrigerant flow rate, c p is the specific heat capacity of the refrigerant, T in is the refrigerant inlet temperature, T out is the refrigerant outlet temperature.

[0060] Among them, the heat load is combined with the real-time data, and the correction is expressed as:

[0061] Q corr (t) = Q(t)·f e (Tenv ,H env );

[0062] Where Q corr (t) is the corrected heat load of the air conditioning system, Q(t) is the heat load during operation, t is time, T env is the real-time external temperature, H env is the real-time external humidity, f e is the environmental correction factor.

[0063] The change of refrigerant flow rate with temperature and volume is expressed as:

[0064]

[0065] Where P is the air conditioning system pressure, n is the amount of refrigerant, R is the gas constant of the refrigerant, T is the refrigerant temperature, and V is the refrigerant volume.

[0066] Among them, the refrigerant pressure in the pipeline is expressed as:

[0067]

[0068] Where ΔP is the pressure drop of the refrigerant, L is the length of the pipe, D is the diameter of the pipe, ρ is the density of the refrigerant, v is the flow rate of the refrigerant, and f f is the pipeline friction loss coefficient.

[0069] Among them, verifying the status of the air conditioning system to diagnose potential heat loss problems is expressed as:

[0070] Q in -Q out =ΔU+W;

[0071] Where Q in is the heat entering the air conditioning system, Q out To discharge the heat of the air conditioning system, ΔU is the change in heat of the air conditioning system, and W is the work done by the air conditioning system.

[0072] The dynamic temperature model is constructed by combining heat load, refrigerant pressure and refrigerant flow, air specific heat capacity and compartment volume parameters, and is specifically expressed as follows:

[0073]

[0074] Where Δt is the change in internal air temperature during vehicle operation, c p ·ρ·V is the difficulty of temperature change when the cabin air absorbs or releases heat. The larger the heat capacity, the slower the temperature change; the smaller the heat capacity, the faster the temperature change. Q in is the heat entering the air conditioning system, Q out To remove heat from the air conditioning system.

[0075] The proposed method optimally describes the dynamic process of air temperature changes within a vehicle cabin, comprehensively considering the effects of heat input, heat output, and the heat capacity of the cabin air. It not only has clear physical meaning but also provides a basis for dynamic temperature prediction and air conditioning system optimization.

[0076] S4. Input the fault deduction report into the machine learning algorithm for analysis, and determine the fault type and source based on historical fault data and the air conditioning system working mode.

[0077] Specifically, the fault parameters in the fault simulation report are extracted and normalized to generate a multi-dimensional feature vector. The dynamic physical model is used to determine the fault type and fault source classification.

[0078] Among them, the normalization processing of the fault parameters in the fault simulation report is expressed as:

[0079]

[0080] Where X′ is the normalized fault parameter data, X is the fault simulation report, and X max is the maximum value in the fault deduction report, X min This is the minimum value in the fault deduction report.

[0081] Among them, the generated multidimensional feature vector is expressed as:

[0082] F={X1′,X2′,...,X′ n};

[0083] Where F is the fault feature vector, X1′,X2′,...,X′ n is the normalized parameter data of each fault.

[0084] Among them, the simulation using the dynamic physical model is expressed as:

[0085]

[0086] Where S is the physical simulation model, W is the working state of the air conditioning system, is the temperature change rate, Refrigerant pressure change rate, is the refrigerant flow rate change rate, and F is the fault feature vector.

[0087] The adaptive physical modeling is expressed as:

[0088] M(F,S)=argmaxP(y i |F,S;θ);

[0089] Where M(F,S) is the fault diagnosis model, S is the physical simulation data, argmaxP is the fault type selection, argmaxP(y i |F,S; θ) is the prediction of a specific fault type based on the current operating status of the air-conditioning system and the collected fault data, y i is the fault type diagnosis confidence, and θ is the parameter set in the air conditioning system fault diagnosis model.

[0090] The dynamic physical model is used to determine the fault type and fault source classification as follows:

[0091] y=N(F,S);

[0092] Where y is the fault type or fault source, M is the adaptive physical modeling algorithm, and S is the physical simulation data.

[0093] S5. Analyze the fault type and source through machine learning models, generate control strategies based on current working state parameters and output adjustment instructions, monitor and compare expected value evaluation results in real time, and iterate the control strategy.

[0094] Among them, the fault type and fault source analyzed by the machine learning model are expressed as:

[0095] R=f r (y,X′);

[0096] Where, f r is a deep learning network, R is the fault deduction report, y is the fault type or fault source, and X′ is the normalized fault parameter data.

[0097] (F t ,F s )=f m (R,H);

[0098] Where, (F t ,F s ) is the output fault type and fault source, H is the historical data of the air conditioning system, f m For failure analysis report.

[0099] Combined with the current working state parameters, the control strategy is generated and the adjustment instructions are output as follows:

[0100] C=f c (F t ,F s ,X′);

[0101] Where, f c is a genetic algorithm, and adjusts the air conditioning system, C is the calculation adjustment strategy, F t ,F sis the output fault type and fault source, and X′ is the normalized fault parameter data.

[0102] The output adjustment instruction is expressed as:

[0103] U=f u (C);

[0104] Where U is the output adjustment instruction, f u For PID control.

[0105] This embodiment also provides a computer device, which is suitable for the remote intelligent monitoring method of the health status of the rail transit vehicle air-conditioning system, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the remote intelligent monitoring method of the health status of the rail transit vehicle air-conditioning system as proposed in the above embodiment.

[0106] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through Wi-Fi, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0107] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the remote intelligent monitoring method for the health status of a rail transit vehicle air-conditioning system as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0108] In summary, the present invention achieves real-time analysis and optimized control by combining dynamic physical modeling with machine learning. Distributed sensors and edge computing are used to improve the accuracy of temperature, humidity, and refrigerant data collection and the real-time transmission. Through dynamic temperature prediction models and heat load-refrigerant flow coupling analysis, the fault type and source are accurately deduced. Machine learning deeply mines historical fault data and combines it with real-time monitoring to achieve efficient fault classification and health assessment. Finally, by collecting the operating parameters of the air-conditioning system in real time and combining it with the physical model to build a dynamic temperature prediction model, fault deduction is achieved, improving diagnostic accuracy and response speed.

[0109] Example 2, referring to Table 1, is the second embodiment of the present invention. In order to further verify the technical solution of the present invention, experimental simulation data of the file encryption method is provided.

[0110] Three air conditioning systems (numbered A, B, and C) on a certain city's subway lines were selected as test subjects. Groups A and B used existing technology (based on Wi-Fi communication and threshold alarms), while Group C adopted the solution proposed in this invention. The test lasted 30 days and covered complex operating conditions such as high temperature, high humidity, and morning and evening rush hours.

[0111] First, Group A / B deployed traditional temperature and humidity sensors (±1°C accuracy), refrigerant pressure sensors (range 0-5MPa, ±2%FS), and Wi-Fi communication modules (speed 54Mbps, packet loss rate 8%-15%).

[0112] Next, Group C uses a distributed sensor architecture (temperature ±0.5°C, humidity ±3%RH, refrigerant pressure ±0.5%FS), NB-IoT communication module (bandwidth 180kHz, transmission power 23dBm), and edge computing unit (dual-core ARM Cortex-A72, 1.5GHz). Group C dynamically calculates the heat load and combines the vehicle compartment volume (120m 3 ) and specific heat capacity of air to predict the temperature change trend.

[0113] Then, when the refrigerant flow deviation is greater than 15%, the physical model is triggered to verify the pressure drop and locate the source of the fault in combination with the machine learning model (confidence level greater than 90%).

[0114] Finally, group C generated PID control parameters using a genetic algorithm to adjust the compressor frequency (20-60 Hz) and fan speed, with a response time of <3 seconds (manual intervention in groups A / B required more than 30 seconds).

[0115] The details are shown in Table 1 below:

[0116] Table 1 Experimental data comparison table

[0117]

[0118] Through analysis of the data in the above table, the advantages of the solution of the present invention are as follows:

[0119] Existing technology A / B uses Wi-Fi communication, but the Doppler effect causes signal attenuation at high train speeds (80 km / h), resulting in a peak packet loss rate of 15%. This invention utilizes NB-IoT narrowband technology (anti-interference coding + retransmission mechanism) to improve communication stability to 99.1% under the same operating conditions, ensuring complete data transmission.

[0120] Traditional solutions rely on threshold alarms (e.g., an alarm is triggered when the refrigerant pressure exceeds 4MPa), and are unable to distinguish short-term fluctuations (such as frequent door openings and closings) from actual faults. This invention uses a dynamic model to quantify the trend of heat load changes and combines it with an LSTM network to analyze historical fault characteristics, reducing the false positive rate from 28% to 5.3%. For example, during the experiment, the refrigerant flow rate experienced a brief anomaly (deviation of 18%) due to sensor noise. The physical model verified the relationship between pressure drop and flow rate to confirm that there was no actual leakage, thus avoiding false shutdowns.

[0121] While existing technologies require manual parameter adjustment after fault confirmation, this invention uses edge computing to generate real-time control commands (using a genetic algorithm to optimize PID parameters), reducing compressor frequency adjustment latency by 90%. For example, when the cabin temperature deviates from the set point (25°C to 28°C) due to a surge in passengers, Group C increases the refrigerant flow rate from 0.8kg / s to 1.2kg / s within 3 seconds, achieving a temperature recovery rate of 0.5°C / min (compared to Groups A and B, which required over 5 minutes).

[0122] The traditional solution causes frequent starts and stops of the compressor (an average of 45 starts and stops per day) due to communication delays. However, the present invention predicts the heat load through a dynamic temperature prediction model and adopts a smooth control strategy to reduce the number of starts and stops to 12 times per day, thereby extending the equipment life and reducing energy consumption.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A remote intelligent monitoring method for the health status of a rail transit vehicle air conditioning system, characterized by: include, Initialize the rail transit air conditioning system and collect data in real time, and send it to edge computing using wireless communication. This includes: using an embedded control platform to initialize the air conditioning system, using distributed sensors to collect temperature, humidity, and refrigerant data in real time, storing it in the control unit cache, and transmitting it to edge computing using narrowband Internet of Things; The received data is preprocessed and compared with preset operating parameters to calculate the current operating status of the air conditioning system. This includes normalizing temperature, humidity, and refrigerant data and extracting preset operating parameters built into edge computing. Deviation calculation is used to output deviation results and mark anomalies. The health status and abnormal status of the air conditioning system are evaluated based on parameter weights. Build a physical model to simulate the current working status of the air conditioning system. Use heat load prediction and pressure change analysis to simulate the operation of the air conditioning system and compare it with real-time data to generate a fault simulation report. The specific steps are as follows: A dynamic physical model is constructed to couple the changes in heat load and refrigerant flow to generate a dynamic temperature prediction model. This includes identifying the heat load, refrigerant pressure, and refrigerant flow under the current operating state, combined with the specific heat capacity of air and the cabin volume, to construct the core thermodynamic relationship of cabin temperature changes. Based on the core relationship of thermodynamics, by dynamically balancing heat input, output and loss, a dynamic integral form is used to quantify the temperature change trend over time and a dynamic temperature prediction model is constructed; Update dynamic temperature prediction parameters through real-time data, correct changes in heat load and refrigerant pressure, and optimize the dynamic temperature prediction model in real time; Combined with real-time collected temperature data, the system analyzes temperature deviation trends, determines operating status based on refrigerant pressure changes, classifies and locates faults, and generates a fault deduction report. This includes: Based on a dynamic temperature prediction model, it compares real-time collected data with predicted values, deduces fault types, and displays anomalies through time series graphs; uses fault type results to evaluate the air conditioning system status, classifies and summarizes fault types and causes, and displays the fault evolution process and anomalies through multi-parameter trend graphs to generate a fault deduction report; The fault simulation report is fed into a machine learning algorithm for analysis, and the fault type and source are determined based on historical fault data and the air conditioning system operating mode. This includes: extracting temperature, pressure, and refrigerant fault parameters from the fault simulation report, normalizing them, and generating a multidimensional feature vector; using a physical simulation model to build a model to simulate the fault state, and using a machine learning algorithm to classify the fault type and infer the fault source; Through machine learning algorithm analysis and combined with remote intelligent adaptive control, the current working status of the air-conditioning system is updated. This includes: combining the output fault type and fault source with the current working status parameters to generate a control strategy and output adjustment instructions; real-time remote monitoring of the working status of the air-conditioning system and comparing it with the expected value, evaluating the results through a real-time feedback mechanism, and updating the strategy to adjust the air-conditioning system status.

2. The method for remote intelligent monitoring of the health status of a rail transit vehicle air conditioning system according to claim 1, characterized in that: The following steps are involved in evaluating the health and abnormal status of the air conditioning system using parameter weights: The dynamic threshold is calculated based on the mean and standard deviation of the health score and is dynamically adjusted through the adjustment coefficient to adapt to the health status assessment under different working conditions.

3. The method for remote intelligent monitoring of the health status of a rail transit vehicle air conditioning system according to claim 2, characterized in that: Analysis using machine learning algorithms involves the following steps: Long short-term memory network is used as the machine learning algorithm to judge the fault type and source by analyzing the multi-dimensional feature vectors in the fault deduction report.

Images