Subway signal fault diagnosis and prediction method and system based on artificial intelligence

Abstract

The invention discloses a subway signal fault diagnosis and prediction method and system based on artificial intelligence, and particularly relates to the technical field of signal faults, and the method comprises the steps: simulating the activities of passengers in a subway in a carriage through a wireless network tester, determining different flow modes in a subway operation process, and carrying out the diagnosis and prediction of the subway signal faults; the signal switching process in the running process of the subway is monitored, the signal performance index information generated by the wireless network tester and the actual switching information of the actual wireless network tester are collected, and the signal quality in different flow modes is quantified through a signal evaluation coefficient; signal evaluation coefficients under different flow modes and duration proportions of different flow modes in the operation process of a subway are used as input of a regression model, and the possibility of subway signal faults is quantified through the regression model, so that potential signal faults can be found, advanced prevention measures can be taken, and the probability that the subway signal faults occur is reduced. Therefore, the stability and reliability of subway signals are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal failure, and more specifically, to a subway signal failure diagnosis and prediction method and system based on artificial intelligence. Background Art

[0002] The signal switching of the current subway communication system mainly relies on the traditional location-based method, that is, when the train passes through different base station coverage areas, the user's mobile phone signal automatically switches from one base station to another. With the development of modern urban subway systems, passengers have higher and higher requirements for seamless communication experience, especially when the subway car is running fast, the signal switching is frequent and the environment is complex, resulting in large fluctuations in signal quality;

[0003] In the prior art, when the subway signal system switches base stations or signal frequency bands, problems such as signal interruption, decreased transmission rate, and increased packet loss rate often occur, which affects the communication quality and user experience of passengers. At the same time, the signal problems in the subway system are complex and hidden. Many signal failures can only be discovered after they occur, and there is a lack of effective advance prediction and prevention mechanisms. Therefore, the existing signal switching monitoring method cannot provide effective prediction and real-time warning of signal failures, and it is difficult to ensure the communication continuity and stability during subway operation. In addition, simulating passenger flow patterns in a dynamic environment, monitoring the signal switching process in real time, and predicting and preventing signal failures according to different flow patterns have become a technical problem in the current subway communication system.

[0004] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a subway signal fault diagnosis and prediction method and system based on artificial intelligence to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The subway signal fault diagnosis and prediction method based on artificial intelligence specifically includes the following steps:

[0008] S1: Use the wireless network tester to simulate the activities of passengers in the subway carriage, determine the different flow patterns during the operation of the subway, and monitor the signal switching process during the operation of the subway;

[0009] S2: collecting signal performance index information generated by the wireless network tester and actual switching information of the actual wireless network tester;

[0010] S3: Comprehensively analyze the signal performance index information and the actual switching information, and quantify the signal quality under different traffic modes through the signal evaluation coefficient;

[0011] S4: The signal evaluation coefficients under different traffic patterns and the proportion of time when different traffic patterns exist during subway operation are used as inputs of the regression model, and the possibility of subway signal failure is quantified through the regression model.

[0012] In a preferred embodiment, the signal performance indicator information includes:

[0013] The actual switching information is represented by the switching difference density coefficient and the switching delay fluctuation coefficient;

[0014] The acquisition logic of the switching difference density coefficient is as follows: set a unit monitoring interval, determine the number of virtual users who successfully switch signals during the signal switching process under a specific traffic mode through a wireless network tester, and mark the number of virtual users who successfully switch signals in different unit monitoring intervals as: CG n , where n = 1, 2, 3, ..., N, N is a positive integer, and n is the number of the unit monitoring interval during the signal switching process;

[0015] The probability density of the number of virtual users who successfully switched signals during the signal switching process in different unit monitoring intervals is estimated by using the Gaussian kernel. The expression of the probability density is: in, is the probability density of the number of virtual users who successfully switch signals in the xth unit monitoring interval, h is the bandwidth parameter, K is the Gaussian kernel function,

[0016] The probability density of the number of virtual users who successfully switched signals during the signal switching process of different unit monitoring intervals is obtained, and the probability density of the number of virtual users who successfully switched signals during the signal switching process of different unit monitoring intervals is marked as:

[0017] Set a probability density threshold, compare the probability density threshold with the probability density of the number of virtual users who successfully switch signals during signal switching in different unit monitoring intervals, obtain the probability density in the unit monitoring interval that is less than the probability density threshold, and obtain the number of virtual users who successfully switch signals in the probability density that is less than the probability density threshold;

[0018] By setting a time threshold, the unit monitoring interval greater than the time threshold during the signal switching process will be obtained, and the number of virtual users that successfully switch signals within the unit monitoring interval greater than the time threshold will be obtained;

[0019] Calculate the switching difference density coefficient, the calculation formula is: Among them, CYmd is the switching difference density coefficient, SJ sl is the number of virtual users that successfully switch signals within a unit monitoring interval greater than the time threshold, MD sl is the number of virtual users who successfully switch signals in the probability density less than the probability density threshold, SB sl It is the number of virtual users that failed to switch successfully during the signal switching process.

[0020] In a preferred implementation, the logic for obtaining the network speed deviation coefficient is:

[0021] The average network transmission rate of the signal switching process in different unit monitoring intervals under a specific traffic mode is determined by a wireless network tester, and the average network transmission rate of the signal switching process in different unit monitoring intervals under a specific traffic mode is marked as: V n ;

[0022] Set the network transmission rate threshold and mark the network transmission rate threshold as: V yz ,Compare the network transmission rate threshold with the average network transmission rate of different unit monitoring intervals, obtain the average network transmission rate less than the network transmission rate threshold in different unit monitoring intervals, and re-mark the average network transmission rate less than the network transmission rate threshold in different unit monitoring intervals as: V i , where i=1, 2, 3, ..., I, and I is a positive integer;

[0023] Calculate the network speed deviation coefficient, the calculation formula is: Among them, PC ws is the network speed deviation coefficient.

[0024] In a preferred embodiment, the actual switching information of the actual wireless network tester includes:

[0025] The signal performance index information is expressed by the network speed deviation coefficient;

[0026] The logic for obtaining the switching delay fluctuation coefficient is as follows: obtaining the time required for the signal switching process under a specific flow mode, and marking the time required for the signal switching process under a specific flow mode as: YC m , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the virtual user in the traffic mode;

[0027] Calculate the average value and standard deviation of the time required for signal switching in a specific flow mode, and mark the average value and standard deviation of the time required for signal switching in a specific flow mode as: YC avg and YC std ,in,

[0028] The coefficient of variation of the time required for signal switching in a specific flow mode is calculated, and the coefficient of variation of the time required for signal switching in a specific flow mode is marked as:

[0029] Calculate the switching delay fluctuation coefficient, the calculation formula is: Among them, QH bd is the switching delay fluctuation coefficient.

[0030] In a preferred embodiment, the signal quality under different traffic modes is quantified by a signal evaluation coefficient, including:

[0031] The signal performance index information generated by the wireless network tester and the actual switching information of the actual wireless network tester are comprehensively analyzed. By weighted calculation of the switching difference density coefficient, network speed deviation coefficient and switching delay fluctuation coefficient, a signal evaluation model is constructed to generate a signal evaluation coefficient. The expression is: Among them, PG is the signal evaluation coefficient, α1, α2, and α3 are the proportional coefficients of the switching difference density coefficient, the network speed deviation coefficient, and the switching delay fluctuation coefficient, respectively, and α1, α2, and α3 are all greater than 0.

[0032] In a preferred embodiment, the possibility of subway signal failure is quantified by a regression model, including:

[0033] Determine the signal evaluation coefficients under different flow modes and mark the signal evaluation coefficients under different flow modes as: PG r , where r = 1, 2, 3, ..., R, R is a positive integer, r is the number of the traffic mode, obtain the signal evaluation coefficients under different traffic modes, and obtain the proportion of the duration of different traffic modes in the operation of the subway, and obtain the possible signal impact values ​​under different signal evaluation coefficients. The signal impact value represents the impact of the signal evaluation coefficient on the signal failure, which is set by professional staff;

[0034] The signal evaluation coefficients under different traffic patterns and the proportion of time when different traffic patterns exist in the subway during operation are used as the input of the regression model to generate the fault evaluation coefficient. The expression of the fault evaluation coefficient is: Among them, PG gz are the fault assessment coefficients, β1, β2, β3, ..., β R are the signal impact values ​​of different signal evaluation coefficients, ZB1, ZB2, ZB3, ..., ZB Ris the proportion of time when different traffic patterns exist during the operation of the subway, and ZB1+ZB2+ZB3+…+ZB R =1;

[0035] A fault assessment coefficient threshold is set, and the fault assessment coefficient of the subway during operation is compared with the fault assessment coefficient threshold. If the fault assessment coefficient is greater than the fault assessment coefficient threshold, a fault warning signal is generated; if the fault assessment coefficient is less than the fault assessment coefficient threshold, no signal is generated.

[0036] In a preferred embodiment, the subway signal fault diagnosis and prediction system based on artificial intelligence includes a wireless network test module, a data acquisition module, a signal quality assessment module and a fault assessment module, and the modules are signal-connected;

[0037] The signal quality quantification module is used to simulate the passenger traffic pattern through the wireless network tester, generate virtual users and transmit data, monitor the signal performance of passengers during the subway signal switching process, and ensure that the system can collect real signal switching data under different traffic patterns;

[0038] A data acquisition module, used to collect signal performance index information generated by the wireless network tester and actual switching information of the actual wireless network tester;

[0039] The signal quality assessment module is used to quantify the signal quality under different traffic modes, generate signal quality assessment coefficients based on the collected data, and digitize the complex signal performance;

[0040] The fault assessment module is used to analyze the signal quality quantification results using a regression model, discover potential signal faults based on historical subway operation data, calculate the probability of fault occurrence, and generate early warning signals.

[0041] Technical effects and advantages of the present invention:

[0042] The present invention simulates the flow pattern of passengers through a wireless network tester, monitors the performance of signals during the signal switching process, collects data during the signal switching process when the subway is running, quantifies the quality of signals under different flow patterns, and uses the quantification results as the input of a regression model to predict the possibility of failure of the subway during operation. The present invention helps to discover potential signal failures and take preventive measures in advance, thereby improving the stability and reliability of subway signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0044] Figure 1It is a flow chart of the subway signal fault diagnosis and prediction method based on artificial intelligence of the present invention;

[0045] Figure 2 It is a structural schematic diagram of the subway signal fault diagnosis and prediction system based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Example 1

[0048] Figure 1 A schematic flow chart of the subway signal fault diagnosis and prediction method based on artificial intelligence of the present invention is given, which specifically includes the following steps:

[0049] S1: Use the wireless network tester to simulate the activities of passengers in the subway carriage, determine the different flow patterns during the operation of the subway, and monitor the signal switching process during the operation of the subway;

[0050] S2: collecting signal performance index information generated by the wireless network tester and actual switching information of the actual wireless network tester;

[0051] S3: Comprehensively analyze the signal performance index information and the actual switching information, and quantify the signal quality under different traffic modes through the signal evaluation coefficient;

[0052] S4: The signal evaluation coefficients under different traffic patterns and the proportion of time when different traffic patterns exist during subway operation are used as inputs of the regression model, and the possibility of subway signal failure is quantified through the regression model.

[0053] In subway stations and carriages, multiple base stations or wireless access points (APs) are usually deployed to ensure full coverage in stations and carriages. In the carriage, passengers' mobile phone signals are usually provided by wireless access points (APs) in the carriage. The access points are connected to the station's main base station or communication system to ensure that passengers remain connected in the carriage. In the carriage, the frequency of signal switching is low. The signal may switch from the previous station to the next station in the subway, that is, the passenger's mobile phone may switch from the base station of one station to the base station of the next station. The switching process usually starts before the train enters the station and continues after leaving the station.

[0054] When a train enters a new station, the mobile phone will automatically switch from the old base station to the new base station. The passenger's mobile phone will dynamically select the strongest signal source for switching. The switching is automatic and passengers usually do not feel obvious interruption. The subway system will ensure the continuity of signal coverage as much as possible. Therefore, the subway system usually conducts real-time monitoring to ensure the coverage and signal quality of the wireless network, especially when the train enters a new station or moves to a different area.

[0055] The wireless network tester is used to simulate the activities of passengers in the subway car, for example, simulating users to surf the Internet, make video calls, send data, etc., and test the signal switching of multiple virtual users from one station to another.

[0056] Set user load and define the load type of virtual users, such as data transmission rate, communication protocol, application usage, etc., to simulate different traffic patterns and load conditions;

[0057] Simulate different traffic patterns. The following are the specific traffic patterns:

[0058] Constant load, simulating a constant traffic pattern, where users continuously transmit data at a fixed rate. This mode is suitable for testing the performance of the network under a stable load;

[0059] Burst load: simulates bursty data transmission, such as video viewing or file downloading during peak hours, sets the load during peak hours, and evaluates the network's processing capacity and stability;

[0060] Simulate intermittent traffic patterns, such as users occasionally downloading large files, to test the network's responsiveness under load changes;

[0061] Dynamic load, which simulates users switching between different applications and services, and the load changes dynamically, for example, users switch from browsing the web to watching a video stream;

[0062] Simulate fluctuations in network traffic, such as in a subway system where changes in the number of users and activity affect traffic loads.

[0063] It should be noted that there are many types of traffic modes, and professional staff are required to set various traffic modes according to actual application scenarios. By simulating different traffic modes and conducting detailed signal evaluations, professional staff can identify and resolve signal failures in the subway system. For example, a wireless network tester can be used to simulate parameters such as data transmission rate, number of users, connection protocol, and test time to switch different traffic modes. The details are not elaborated here.

[0064] The signal switching process of the subway during its operation is monitored by a wireless network tester, and the signal performance index information generated by the wireless network tester and the actual switching information of the actual wireless network tester are collected. After collection, the actual switching information is represented by the switching difference density coefficient and the switching delay fluctuation coefficient, and the signal performance index information is represented by the network speed deviation coefficient.

[0065] The acquisition logic of the switching difference density coefficient is as follows: set a unit monitoring interval, determine the number of virtual users who successfully switch signals during the signal switching process under a specific traffic mode through a wireless network tester, and mark the number of virtual users who successfully switch signals in different unit monitoring intervals as: CG n , where n = 1, 2, 3, ..., N, N is a positive integer, and n is the number of the unit monitoring interval during the signal switching process;

[0066] It should be noted that the unit monitoring interval is a specific time period. The signal switching process is equally divided into multiple unit detection intervals, and the unit monitoring intervals are sorted according to the duration of the signal switching process. The unit monitoring intervals are set by professional staff. The wireless network tester determines the number of simulated users in the traffic pattern and records the number of simulated users who successfully switch signals in different unit monitoring intervals during the signal switching process.

[0067] The probability density of the number of virtual users who successfully switched signals during the signal switching process in different unit monitoring intervals is estimated by using the Gaussian kernel. The expression of the probability density is: in, is the probability density of the number of virtual users who successfully switch signals in the xth unit monitoring interval, h is the bandwidth parameter, K is the Gaussian kernel function,

[0068] It should be noted that the bandwidth h is a key parameter of kernel density estimation, which determines the degree of smoothness. A larger bandwidth will lead to a smoother density estimate but may ignore the details of the data. A smaller bandwidth will fit the data points more closely but may cause noise and discontinuity. The bandwidth parameter is set by professional staff.

[0069] The probability density of the number of virtual users who successfully switched signals during the signal switching process of different unit monitoring intervals is obtained, and the probability density of the number of virtual users who successfully switched signals during the signal switching process of different unit monitoring intervals is marked as:

[0070] Set a probability density threshold, compare the probability density threshold with the probability density of the number of virtual users who successfully switch signals during signal switching in different unit monitoring intervals, obtain the probability density in the unit monitoring interval that is less than the probability density threshold, and obtain the number of virtual users who successfully switch signals in the probability density that is less than the probability density threshold;

[0071] By setting a time threshold, the unit monitoring interval greater than the time threshold during the signal switching process will be obtained, and the number of virtual users that successfully switch signals within the unit monitoring interval greater than the time threshold will be obtained;

[0072] It should be noted that the unit monitoring interval with a larger probability density represents the range of normal signal switching and the time range in which most virtual user signals are successfully switched. The unit monitoring interval with a smaller probability density can be used to identify weak links and potential failure points in signal switching, and to help evaluate the efficiency and timeliness of signal switching by setting time thresholds. The probability density threshold and time threshold are set by professional staff.

[0073] Calculate the switching difference density coefficient, the calculation formula is: Among them, CY md is the switching difference density coefficient, SJ sl is the number of virtual users that successfully switch signals within a unit monitoring interval greater than the time threshold, MD sl is the number of virtual users who successfully switch signals in the probability density less than the probability density threshold, SB sl It is the number of virtual users that failed to switch successfully during the signal switching process.

[0074] It can be seen from the formula that the larger the switching difference density coefficient is, the greater the difference between different unit monitoring intervals during the signal switching process is, the poorer the stability and uniformity of the signal switching is, and there may be a delay or untimely switching.

[0075] The logic for obtaining the network speed deviation coefficient is as follows: using a wireless network tester, determine the average network transmission rate of different unit monitoring intervals during signal switching under a specific flow mode, and mark the average network transmission rate of different unit monitoring intervals during signal switching under a specific flow mode as: V n ;

[0076] Set the network transmission rate threshold and mark the network transmission rate threshold as: V yz ,Compare the network transmission rate threshold with the average network transmission rate of different unit monitoring intervals, obtain the average network transmission rate less than the network transmission rate threshold in different unit monitoring intervals, and re-mark the average network transmission rate less than the network transmission rate threshold in different unit monitoring intervals as: V i, where i=1, 2, 3, ..., I, and I is a positive integer;

[0077] Calculate the network speed deviation coefficient, the calculation formula is: Among them, PC ws is the network speed deviation coefficient.

[0078] From the formula, we can see that the larger the network speed deviation coefficient is, the more unstable the network performance is and the larger the fluctuation range of the network speed is. This indicates that it is necessary to optimize the allocation of network resources, improve network coverage and performance stability to ensure a more consistent user experience. Possible reasons for this phenomenon include:

[0079] When the signal is switched, if a large number of users are connected to the new base station or frequency band, the base station resources (such as bandwidth and spectrum) are over-occupied, resulting in a significant fluctuation or drop in network speed. Users may feel a significant drop in network speed immediately after the switch, especially during peak hours or in crowded areas.

[0080] During the signal switching process, the signal coverage of the two base stations may be uneven. If the signal of the target base station is weak or has a blind spot, the signal strength of the device will decrease, affecting network performance. When the signal strength is weak, the network speed will slow down, and the signal may be intermittent, resulting in significant fluctuations in network speed.

[0081] The performance difference between two base stations or frequency bands may be large. For example, one base station has higher bandwidth resources, while the other base station has lower bandwidth. When the signal is switched, if it switches to the base station with lower performance, the network speed will drop significantly.

[0082] The logic for obtaining the switching delay fluctuation coefficient is as follows: obtaining the time required for the signal switching process under a specific flow mode, and marking the time required for the signal switching process under a specific flow mode as: YC m , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the virtual user in the traffic mode;

[0083] It should be noted that different virtual users may spend different amounts of time in the signal switching process, but they should be close to the same baseline. The switching time of different virtual users should be compared with the baseline to check whether the time is within the expected range. Under normal circumstances, the switching time of all users should be close to the baseline.

[0084] Calculate the average value and standard deviation of the time required for signal switching in a specific flow mode, and mark the average value and standard deviation of the time required for signal switching in a specific flow mode as: YC avg and YC std ,in,

[0085] The coefficient of variation of the time required for signal switching in a specific flow mode is calculated, and the coefficient of variation of the time required for signal switching in a specific flow mode is marked as:

[0086] It should be noted that, the larger the coefficient of variation of the time required for signal switching in a specific traffic mode, the higher the dispersion of the switching time of the virtual user relative to the average value during the signal switching process, that is, the greater the volatility of time, which may indicate the instability of the signal switching process or the large difference in network conditions between different virtual users.

[0087] Calculate the switching delay fluctuation coefficient, the calculation formula is: Among them, QH bd is the switching delay fluctuation coefficient.

[0088] It can be seen from the formula that the larger the switching delay fluctuation coefficient is, the higher the instability of the signal switching process is.

[0089] The signal performance index information generated by the wireless network tester and the actual switching information of the actual wireless network tester are comprehensively analyzed. By weighted calculation of the switching difference density coefficient, network speed deviation coefficient and switching delay fluctuation coefficient, a signal evaluation model is constructed to generate a signal evaluation coefficient. The expression is: Among them, PG is the signal evaluation coefficient, α1, α2, and α3 are the proportional coefficients of the switching difference density coefficient, the network speed deviation coefficient, and the switching delay fluctuation coefficient, respectively, and α1, α2, and α3 are all greater than 0.

[0090] It can be seen from the formula that the larger the signal evaluation coefficient is, the larger the switching difference density coefficient, the network speed deviation coefficient, and the switching delay fluctuation coefficient are, which means that the possibility of a fault in the signal switching during the subway operation is higher. Conversely, the smaller the signal evaluation coefficient is, the smaller the switching difference density coefficient, the network speed deviation coefficient, and the switching delay fluctuation coefficient are, which means that the possibility of a fault in the signal switching during the subway operation is lower.

[0091] Determine the signal evaluation coefficients under different flow modes and mark the signal evaluation coefficients under different flow modes as: PG r , where r=1, 2, 3, ..., R, R is a positive integer, r is the number of the flow pattern, the signal evaluation coefficients under different flow patterns are obtained, and the proportion of the time when different flow patterns exist in the subway during operation is obtained, and the possible signal impact values ​​under different signal evaluation coefficients are obtained. The signal impact value represents the impact of the signal evaluation coefficient on signal failure and is set by professional staff.

[0092] It should be noted that when the subway completes its entire journey, it will go through multiple signal switching processes, and each signal switching process may be completed in different traffic modes. Based on the historical data of subway operation, the proportion of the length of time of each traffic mode in the signal switching process per unit time can be predicted. The unit time is usually one day, which is set by professional staff.

[0093] For example: During the operation of the subway, there are flow patterns a, flow pattern b, and flow pattern c. Through the historical data of the subway operation, we know that flow pattern a is at noon, flow pattern b is in the afternoon, and flow pattern c is in the morning. That is, in the process of obtaining the switching signal per unit time, the time length of each flow pattern per unit time accounts for one third.

[0094] The signal evaluation coefficients under different traffic patterns and the proportion of time when different traffic patterns exist in the subway during operation are used as the input of the regression model to generate the fault evaluation coefficient. The expression of the fault evaluation coefficient is: Among them, PG gz are the fault assessment coefficients, β1, β2, β3, ..., β R are the signal impact values ​​of different signal evaluation coefficients, ZB1, ZB2, ZB3, ..., ZB R is the proportion of time when different traffic patterns exist during the operation of the subway, and ZB1+ZB2+ZB3+…+ZB R =1.

[0095] A fault assessment coefficient threshold is set, and the fault assessment coefficient of the subway during operation is compared with the fault assessment coefficient threshold. If the fault assessment coefficient is greater than the fault assessment coefficient threshold, a fault warning signal is generated, and relevant personnel are immediately notified to monitor and repair each base station of the subway. If the fault assessment coefficient is less than the fault assessment coefficient threshold, no signal is generated.

[0096] The present invention simulates the flow pattern of passengers through a wireless network tester, monitors the performance of signals during the signal switching process, collects data during the signal switching process when the subway is running, quantifies the quality of signals under different flow patterns, and uses the quantification results as the input of a regression model to predict the possibility of failure of the subway during operation. The present invention helps to discover potential signal failures and take preventive measures in advance, thereby improving the stability and reliability of subway signals.

[0097] Example 2

[0098] Figure 2 The present invention provides a schematic diagram of the structure of the subway signal fault diagnosis and prediction system based on artificial intelligence, including a wireless network test module, a data acquisition module, a signal quality assessment module and a fault assessment module, and the modules are connected by signals;

[0099] The signal quality quantification module is used to simulate the passenger traffic pattern through the wireless network tester, generate virtual users and transmit data, monitor the signal performance of passengers during the subway signal switching process, and ensure that the system can collect real signal switching data under different traffic patterns;

[0100] A data acquisition module, used to collect signal performance index information generated by the wireless network tester and actual switching information of the actual wireless network tester;

[0101] The signal quality assessment module is used to quantify the signal quality under different traffic modes, generate signal quality assessment coefficients based on the collected data, and digitize the complex signal performance;

[0102] The fault assessment module is used to analyze the signal quality quantification results using a regression model, discover potential signal faults based on historical subway operation data, calculate the probability of fault occurrence, and generate early warning signals.

[0103] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0104] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0105] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0106] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0107] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0108] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0109] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0110] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A subway signal fault diagnosis and prediction method based on artificial intelligence, characterized in that: The specific steps include: S1: Use the wireless network tester to simulate the activities of passengers in the subway carriage, determine the different flow patterns during the operation of the subway, and monitor the signal switching process during the operation of the subway; S2: collecting signal performance index information generated by the wireless network tester and actual switching information of the actual wireless network tester; S3: Comprehensively analyze the signal performance index information and the actual switching information, and quantify the signal quality under different traffic modes through the signal evaluation coefficient; S4: The signal evaluation coefficients under different traffic patterns and the proportion of time when different traffic patterns exist during subway operation are used as inputs of the regression model, and the possibility of subway signal failure is quantified through the regression model.

2. The subway signal fault diagnosis and prediction method based on artificial intelligence according to claim 1 is characterized in that: Signal performance indicator information, including: The actual switching information is represented by the switching difference density coefficient and the switching delay fluctuation coefficient; The acquisition logic of the switching difference density coefficient is as follows: set a unit monitoring interval, determine the number of virtual users who successfully switch signals during the signal switching process under a specific traffic mode through a wireless network tester, and mark the number of virtual users who successfully switch signals in different unit monitoring intervals as: CG n , where n = 1, 2, 3, ..., N, N is a positive integer, and n is the number of the unit monitoring interval during the signal switching process; The probability density of the number of virtual users who successfully switched signals during the signal switching process in different unit monitoring intervals is estimated by using the Gaussian kernel. The expression of the probability density is: in, is the probability density of the number of virtual users who successfully switch signals in the xth unit monitoring interval, h is the bandwidth parameter, K is the Gaussian kernel function, The probability density of the number of virtual users who successfully switched signals during the signal switching process of different unit monitoring intervals is obtained, and the probability density of the number of virtual users who successfully switched signals during the signal switching process of different unit monitoring intervals is marked as: Set a probability density threshold, compare the probability density threshold with the probability density of the number of virtual users who successfully switch signals during signal switching in different unit monitoring intervals, obtain the probability density in the unit monitoring interval that is less than the probability density threshold, and obtain the number of virtual users who successfully switch signals in the probability density that is less than the probability density threshold; By setting a time threshold, the unit monitoring interval greater than the time threshold during the signal switching process will be obtained, and the number of virtual users that successfully switch signals within the unit monitoring interval greater than the time threshold will be obtained; Calculate the switching difference density coefficient, the calculation formula is: Among them, CY md is the switching difference density coefficient, SJ sl is the number of virtual users that successfully switch signals within a unit monitoring interval greater than the time threshold, MD sl is the number of virtual users who successfully switch signals in the probability density less than the probability density threshold, SB sl It is the number of virtual users that failed to switch successfully during the signal switching process.

3. The subway signal fault diagnosis and prediction method based on artificial intelligence according to claim 2 is characterized in that: The logic for obtaining the network speed deviation coefficient is: The average network transmission rate of the signal switching process in different unit monitoring intervals under a specific traffic mode is determined by a wireless network tester, and the average network transmission rate of the signal switching process in different unit monitoring intervals under a specific traffic mode is marked as: V n ; Set the network transmission rate threshold and mark the network transmission rate threshold as: V yz , compare the network transmission rate threshold with the average network transmission rate of different unit monitoring intervals, obtain the average network transmission rate less than the network transmission rate threshold in different unit monitoring intervals, and re-mark the average network transmission rate less than the network transmission rate threshold in different unit monitoring intervals as: V i , where i=1, 2, 3, ..., I, and I is a positive integer; Calculate the network speed deviation coefficient, the calculation formula is: Among them, PC ws is the network speed deviation coefficient.

4. The subway signal fault diagnosis and prediction method based on artificial intelligence according to claim 3 is characterized in that: Actual switching information from an actual wireless network tester, including: The signal performance index information is expressed by the network speed deviation coefficient; The logic for obtaining the switching delay fluctuation coefficient is as follows: obtaining the time required for the signal switching process under a specific flow mode, and marking the time required for the signal switching process under a specific flow mode as: YC m , where m = 1, 2, 3, ..., M, M is a positive integer, and m is the number of the virtual user in the traffic mode; Calculate the average value and standard deviation of the time required for signal switching in a specific flow mode, and mark the average value and standard deviation of the time required for signal switching in a specific flow mode as: YC avg and YC std ,in, The coefficient of variation of the time required for signal switching in a specific flow mode is calculated, and the coefficient of variation of the time required for signal switching in a specific flow mode is marked as: Calculate the switching delay fluctuation coefficient, the calculation formula is: Among them, QH bd is the switching delay fluctuation coefficient.

5. The subway signal fault diagnosis and prediction method based on artificial intelligence according to claim 4 is characterized in that: Signal quality in different traffic patterns is quantified using signal evaluation coefficients, including: The signal performance index information generated by the wireless network tester and the actual switching information of the actual wireless network tester are comprehensively analyzed. By weighted calculation of the switching difference density coefficient, network speed deviation coefficient and switching delay fluctuation coefficient, a signal evaluation model is constructed to generate a signal evaluation coefficient. The expression is: Among them, PG is the signal evaluation coefficient, α1, α2, and α3 are the proportional coefficients of the switching difference density coefficient, the network speed deviation coefficient, and the switching delay fluctuation coefficient, respectively, and α1, α2, and α3 are all greater than 0.

6. The subway signal fault diagnosis and prediction method based on artificial intelligence according to claim 5 is characterized in that: The probability of subway signal failure is quantified through regression model, including: Determine the signal evaluation coefficients under different flow modes and mark the signal evaluation coefficients under different flow modes as: PG r , where r = 1, 2, 3, ..., R, R is a positive integer, r is the number of the traffic mode, obtain the signal evaluation coefficients under different traffic modes, and obtain the proportion of the duration of different traffic modes in the operation of the subway, and obtain the possible signal impact values ​​under different signal evaluation coefficients. The signal impact value represents the impact of the signal evaluation coefficient on the signal failure, which is set by professional staff; The signal evaluation coefficients under different traffic patterns and the proportion of time when different traffic patterns exist in the subway during operation are used as the input of the regression model to generate the fault evaluation coefficient. The expression of the fault evaluation coefficient is: Among them, PG gz are the fault assessment coefficients, β1, β2, β3, ..., β R are the signal impact values ​​of different signal evaluation coefficients, ZB1, ZB2, ZB3, ..., ZB R is the proportion of time when different traffic patterns exist during the operation of the subway, and ZB1+ZB2+ZB3+…+ZB R =1; A fault assessment coefficient threshold is set, and the fault assessment coefficient of the subway during operation is compared with the fault assessment coefficient threshold. If the fault assessment coefficient is greater than the fault assessment coefficient threshold, a fault warning signal is generated; if the fault assessment coefficient is less than the fault assessment coefficient threshold, no signal is generated.

7. A subway signal fault diagnosis and prediction system based on artificial intelligence, used to implement the subway signal fault diagnosis and prediction method based on artificial intelligence according to any one of claims 1 to 6, characterized in that: It includes a wireless network test module, a data acquisition module, a signal quality assessment module and a fault assessment module, and the signal connections between the modules; The signal quality quantification module is used to simulate the passenger traffic pattern through the wireless network tester, generate virtual users and transmit data, monitor the signal performance of passengers during the subway signal switching process, and ensure that the system can collect real signal switching data under different traffic patterns; A data acquisition module, used to collect signal performance index information generated by the wireless network tester and actual switching information of the actual wireless network tester; The signal quality assessment module is used to quantify the signal quality under different traffic modes, generate signal quality assessment coefficients based on the collected data, and digitize the complex signal performance; The fault assessment module is used to analyze the signal quality quantification results using a regression model, discover potential signal faults based on historical subway operation data, calculate the probability of fault occurrence, and generate early warning signals.

Images