A Fault Warning Method, Device, Storage Medium, and Electronic Device

Abstract

This application relates to the technical field of rail transit fault warning, and discloses a fault warning method, which is characterized in that the method includes: obtaining the operation status data of all target sensors, and all the target sensors are of the same type; monitoring and warning all the target sensors according to the operation status data through a fault warning model; wherein, the fault warning model is generated according to the operation status data of multiple sensors of the same type of the target sensors. Based on the operation status data of train sensors of the same type, statistical analysis is carried out on the differences in the status data between sensors of the same type, and then the change characteristics and distribution rules of the differences are extracted. Furthermore, a set of sensor anomaly self-checking methods is constructed to realize the intelligent warning of sudden and gradual faults in the train sensor system of the same type, provide suggestions in advance for the maintenance work of after-sales personnel, and ensure the safe and efficient operation of the train.

Description

Technical Field

[0001] The present application relates to the technical field of rail transit fault warning, and particularly to a fault warning method, device, storage medium, and electronic device. Background Art

[0002] Currently, the research on the fault warning of sensors themselves mainly focuses on the following four categories: warning models based on expert experience, fault warning models based on intelligent hardware analysis, fault warning models based on alarm trip limits, and fault warning models based on data-driven.

[0003] The warning model based on expert experience is mainly based on the accumulated experience and knowledge of relevant experts and operators, and qualitatively or quantitatively describes the fault phenomena and fault modes of each device in the fault process. After abnormal signs of the sensor appear, the reasoning ability of process experts in monitoring is simulated through system reasoning, deduction, etc., so as to automatically complete the fault warning of the sensor. This method relies on empirical knowledge and has a large subjectivity.

[0004] The warning idea of the fault warning model based on intelligent hardware analysis requires an additional high-performance sensor to be installed on the monitored sensor, and early warning is carried out through signal acquisition, demodulation, and analysis. This warning method requires a high additional equipment cost and is not feasible under certain installation conditions.

[0005] The warning mode of the fault warning model based on alarm trip limits mainly uses the manufacturer's threshold or industry threshold as the monitoring standard, and timely alarms or trips after the parameters of the sensor to be monitored exceed the limit to protect the equipment and personal safety. In this warning method, the fault warning time is mostly in the late stage of the fault, and at the same time, the fault hidden dangers within the threshold cannot be detected.

[0006] The idea of the fault warning model based on data-driven is to use the state data during the operation of the sensor itself, horizontally or vertically obtain the characteristics and change rules of the sensor state data, and perform early warning analysis on sudden faults and trend anomalies according to it. This method does not require additional equipment to be installed, and only uses the state data of the sensor itself for warning, with low input cost and strong applicability.

[0007] For the same type of sensors on trains, due to the consistency of the measured objects, it is almost impossible for the measured data of normal working same type of sensors to have a large difference. Summary of the Invention

[0008] In view of the above problems, the present application proposes a fault warning method, device, storage medium and electronic device. Based on the operation status data of the same type of sensors on the train, statistical analysis is performed on the differences in the status data between the same type of sensors, and then the change characteristics and distribution rules of the differences are extracted. Furthermore, a set of sensor anomaly self-check methods is constructed to realize the intelligent warning of sudden and gradual faults in the same type of sensor system of the train, provide suggestions in advance for the maintenance work of after-sales personnel, and ensure the safe and efficient operation of the train.

[0009] In the first aspect of the present application, a fault warning method is provided, and the method includes:

[0010] Obtain the operation status data of all target sensors, and all the target sensors are of the same type;

[0011] Monitor and warn all the target sensors according to the operation status data through a fault warning model;

[0012] Among them, the fault warning model is generated according to the operation status data of multiple same type of sensors of the target sensor.

[0013] In some embodiments, the generation steps of the fault warning model include:

[0014] Obtain the operation status data of the multiple same type of sensors;

[0015] Obtain the original status data difference sequence according to the operation status data;

[0016] Obtain the window feature value difference sequence according to the operation status data;

[0017] Generate the fault warning model according to the original status data difference sequence and the window feature value difference sequence.

[0018] In some embodiments, all the multiple same type of sensors are sensors that have not had faults in history.

[0019] In some embodiments, after obtaining the operation status data of the multiple same type of sensors, it further includes:

[0020] Filter and process the operation status data of the multiple same type of sensors.

[0021] In some embodiments, the generating the fault warning model according to the original status data difference sequence and the window feature value difference sequence includes:

[0022] Obtain the warning threshold interval and the difference value time series according to the original status data difference sequence and the window feature value difference sequence;

[0023] Perform a stationarity analysis on the difference value time series to obtain the analysis result;

[0024] Obtain the historical failure data of the same type of sensors that have failed;

[0025] Generate a sudden failure warning model based on the original state data difference sequence, the window eigenvalue difference sequence, and the warning threshold interval, or generate a gradual failure warning model based on the analysis result and the historical failure data.

[0026] In some embodiments, if the failure warning model is a sudden failure warning model, then the failure warning model monitors and warns all target sensors according to the operating state data, including:

[0027] Obtain the difference of the original state data and the difference of the window eigenvalue in the current time window from the operating state data of all target sensors;

[0028] Perform an anomaly determination on the difference of the original state data in the current time window and the differences of the window eigenvalues in multiple consecutive time windows;

[0029] When it is determined that there is abnormal state difference data in the current time window or abnormal window eigenvalues in multiple consecutive time windows, issue a warning.

[0030] In some embodiments, the anomaly determination of the difference of the original state data in the current time window includes:

[0031] Obtain the number of warning messages that the difference of the original state data in the current time window is not within the warning threshold interval;

[0032] If the number of warning messages is not less than the preset number of warning messages, determine that there is abnormal state difference data in the current time window.

[0033] In some embodiments, the anomaly determination of the differences of the window eigenvalues in multiple consecutive time windows includes:

[0034] If the difference of the window eigenvalue in the current time window is not within the warning threshold interval, then there is an eigenvalue anomaly in the current time window;

[0035] If the number of consecutive windows with eigenvalue anomalies is not less than the preset number of abnormal windows, determine that there are abnormal window eigenvalues in the subsequent multiple consecutive time windows.

[0036] In some embodiments, if the failure warning model is a gradual failure warning model, then the failure warning model monitors and warns all target sensors according to the operating state data, including:

[0037] Obtain the failure probability of the same type of sensors according to the operation status data of all the target sensors;

[0038] When the failure probability is not less than the preset warning probability threshold, give a warning.

[0039] In the second aspect of the present application, a failure warning device is provided, and the device includes:

[0040] An acquisition module, configured to acquire the operation status data of all the target sensors, and all the target sensors are of the same type;

[0041] A warning module, configured to monitor and give a warning to all the target sensors according to the operation status data through a failure warning model;

[0042] A generation module, configured to generate the failure warning model according to the operation status data of multiple same type of sensors of the target sensors.

[0043] In the third aspect of the present application, a storage medium is provided, and a computer program stored in the storage medium can be executed by one or more processors to implement the failure warning method as described above.

[0044] In the fourth aspect of the present application, an electronic device is provided, including a memory and a processor, a computer program is stored on the memory, the memory and the processor are communicatively connected to each other, and when the computer program is executed by the processor, the failure warning method as described above is implemented.

[0045] Compared with the prior art, the technical solution of the present application has the following advantages or beneficial effects:

[0046] 1. Using the operation status data of the same type of sensors itself for failure warning can more effectively and directly discover potential failure possibilities; at the same time, through a large amount of historical normal data and failure data for model construction and verification, the reliability of monitoring and warning results is higher;

[0047] 2. Introducing the method of sliding window to calculate the differential features avoids the loss of data information in the time dimension and ensures the integrity of information;

[0048] 3. Introducing stationarity analysis to judge whether there is a trend change in the differences between the same type of sensors simply and objectively qualitatively determines the trend failure;

[0049] 4. Constructing a trend failure probability prediction model according to the fitted difference value data can timely and accurately predict failures, improve maintenance efficiency, and reduce failure costs. Description of the Drawings

[0050] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained according to the provided accompanying drawings.

[0051] Figure 1 It is a flowchart of a fault warning method provided by an embodiment of the present application;

[0052] Figure 2 It is a schematic flowchart of a sudden fault warning method provided by an embodiment of the present application;

[0053] Figure 3 It is a schematic flowchart of a gradual fault warning method provided by an embodiment of the present application;

[0054] Figure 4 It is a schematic structural diagram of a device provided by an embodiment of the present application;

[0055] Figure 5 It is a connection block diagram of an electronic device provided by an embodiment of the present application. Specific embodiments

[0056] The following will combine the accompanying drawings and embodiments to detail the implementation manner of the present application, so as to fully understand how the present application uses technical means to solve technical problems and the implementation process of achieving corresponding technical effects and implement accordingly. Each feature in the embodiments of the present application and the embodiments can be combined with each other on the premise of not conflicting, and the formed technical solutions are all within the protection scope of the present application.

[0057] Embodiment 1

[0058] This embodiment provides a fault warning method. Figure 1 It is a flowchart of a fault warning method provided by an embodiment of the present application. As Figure 1 shown, the method of this embodiment includes:

[0059] S110. Obtain the operation status data of all target sensors, and all the target sensors are of the same type.

[0060] Optionally, the target sensors include multiple sensors and are all of the same type.

[0061] In some embodiments, after obtaining the operation status data of all target sensors, and all the target sensors are of the same type, it further includes:

[0062] Perform filtering processing on the operation status data.

[0063] Optionally, the filtering process includes removing garbage data from the operation status data, and the garbage data may include: null values, error values, etc. The garbage data may also include data under abnormal working conditions, such as data when just powered on, component isolation or overly equal data.

[0064] It should be noted that when optimizing the operation status data of the target sensor, other data considered as garbage data by the user can also be removed, and the specific garbage data can be set according to the actual needs of the user, and no special limitation is made here.

[0065] S120. Monitor and give early warnings to all the target sensors according to the operation status data through the fault early warning model.

[0066] It should be noted that the fault early warning model is generated according to the operation status data of multiple similar sensors of the target sensor.

[0067] In some embodiments, the generation steps of the fault early warning model include:

[0068] Obtain the operation status data of the multiple similar sensors;

[0069] Obtain the original status data difference sequence according to the operation status data;

[0070] Obtain the window feature value difference sequence according to the operation status data;

[0071] Generate the fault early warning model according to the original status data difference sequence and the window feature value difference sequence.

[0072] In some embodiments, all the multiple similar sensors are sensors that have not had faults in history.

[0073] Optionally, obtaining the operation status data of multiple similar sensors includes: obtaining the operation status data of multiple similar sensors that have not had faults in history.

[0074] It should be noted that all the multiple similar sensors that have not had faults in history do not include any one of the target sensors.

[0075] Optionally, the above sequences all include an array.

[0076] It should be noted that similar sensors include multiple sensors that can detect the same type of data. To a certain extent, as long as the functions of two sensors are the same and they can both detect the same type of data, then these two sensors can be called similar sensors.

[0077] In some embodiments, after obtaining the operation status data of the multiple homogeneous sensors, the method further includes:

[0078] Performing filtering processing on the operation status data of the multiple homogeneous sensors.

[0079] Optionally, after obtaining the operation status data of the homogeneous sensors that have not failed historically, null values or error values are removed, and then the data is filtered to exclude sensor data under abnormal working conditions (for example, when the system is just powered on, components are isolated, or excessive equality).

[0080] Optionally, the garbage data may include: null values, error values, etc. The garbage data may also include data under abnormal working conditions, such as data when just powered on, component isolation, or excessive equality data.

[0081] It should be noted that when optimizing the operation status data of the target sensor, other data considered as garbage data by the user can also be removed. The specific garbage data can be set according to the actual needs of the user, and no special limitation is made here.

[0082] Optionally, the original state difference can be expressed as X_diff = X_a - X_b, where a and b represent two homogeneous sensors, and X_a and X_b are the state sequences of the sensors respectively. The homogeneous sensors can also include more sensors. For example, X_diff = X_a - X_b - X_c - … X_n, where in this formula X_n represents the state sequence of the nth sensor.

[0083] In some embodiments, the window eigenvalue difference sequence includes:

[0084] Mean value, change rate.

[0085] Optionally, determine the time window length T and perform sliding window in the way of overlapping rate P; calculate eigenvalue such as the mean value and change rate of the data within the window for the state sequence of a single window. For example, for the sequence {Xa_t0, Xa_t1, Xa_t2,......, Xa_tn} of sensor a with a time window length from t0 to tn, calculate the mean value of this sequence (average of time series), and the change rate det = (Xa_t1 - Xa_t0) / (t1 - t0). The window eigenvalue difference sequences of sensors a and b can be obtained respectively, and then the difference sequence of the window characteristics of homogeneous sensors can be calculated. For example, Y_diff = Y_a - Y_b, where a and b represent two homogeneous sensors, and Y_a and Y_b are the sequences of window eigenvalues (mean value, change rate, etc.) of sensors a and b respectively.

[0086] It should be noted that in this embodiment, the window feature information is reflected by the mean value and the change rate, and features such as variance and median can also be used to replace them.

[0087] In some embodiments, generating the fault warning model according to the original state data difference sequence and the window feature value difference sequence includes:

[0088] Obtaining an early warning threshold interval and a difference value time series according to the original state data difference sequence and the window feature value difference sequence;

[0089] Performing a stationarity analysis on the difference value time series to obtain an analysis result;

[0090] Obtaining historical fault data of the same type of sensors that have had faults;

[0091] Generating a sudden fault warning model according to the original state data difference sequence, the window feature value difference sequence, and the early warning threshold interval, or generating a gradual fault warning model according to the analysis result and the historical fault data.

[0092] Optionally, other sensors of the same type include sensors of the same type that have had faults.

[0093] In some embodiments, obtaining the early warning threshold interval according to the original state data difference sequence and the window feature value difference sequence includes:

[0094] Obtaining the mean value mean and the standard deviation std according to the original state data difference sequence and the window feature value difference sequence;

[0095] Respectively taking mean - 3*std and mean + 3*std as the upper and lower limits of the early warning threshold interval.

[0096] Optionally, for the above difference sequences (original state data difference sequence, window feature value difference sequence), calculate their mean value mean and standard deviation std respectively, and take [mean - 3*std, mean + 3*std] as the upper and lower limit thresholds. The upper and lower limit thresholds can also be expressed in other ways such as [0.01%, 99.99%] quantiles.

[0097] It should be noted that the upper and lower limit thresholds can be set according to the actual needs of users, and no special limitation is made here specifically.

[0098] Optionally, obtaining the mean value mean and the standard deviation std according to the original state data difference sequence and the window feature value difference sequence, where the mean value mean can be obtained by the following formula:

[0099]

[0100] The standard deviation std can be obtained by the following formula:

[0101]

[0102] Optionally, after obtaining the difference sequences (the difference sequence of the original state data and the difference sequence of the window feature values), a difference value time series is obtained.

[0103] Optionally, based on the difference sequence of the original state data and the difference sequence of the window feature values obtained above, a difference value time series is obtained. For example, if the difference in the difference sequence of the original state data does not meet the preset difference requirement, then this time window is saved to the difference value time series, where the specific preset difference requirement can be set according to expert experience or actual needs.

[0104] Optionally, the obtained difference value time series is analyzed for stationarity by the ADF unit root test method. If there is no unit root in the difference sequence, the sequence is stationary, that is, there is no gradual change trend in the difference value; otherwise, the sequence is considered non-stationary, the difference value time series is fitted, and combined with historical fault data, a gradual fault warning model is constructed.

[0105] It should be noted that although the ADF unit root test method is selected in this embodiment for the stationarity test of the difference value, other methods such as the PP test, KPSS test, and LMC test can also be selected for the stationarity test of the difference value.

[0106] In some embodiments, if the fault warning model is a sudden fault warning model, then the fault warning model monitors and warns all target sensors according to the operation state data, including:

[0107] From the operation state data of all target sensors, the difference of the original state data and the difference of the window feature values in the current time window are obtained;

[0108] Anomaly determination is performed on the difference of the original state data in the current time window and the differences of the window feature values in multiple consecutive time windows;

[0109] When it is determined that there is abnormal state difference data in the current time window or abnormal window feature values in multiple consecutive time windows, a warning is issued.

[0110] Optionally, the current window includes a time series traced back from the current moment for the window length.

[0111] In some embodiments, the anomaly determination of the difference of the original state data in the current time window includes:

[0112] Obtain the number of early warning messages where the difference in the original state data within the current time window is not within the early warning threshold range;

[0113] If the number of early warning messages is not less than the preset number of early warning messages, it is determined that there is abnormal state difference data in the current time window.

[0114] In some embodiments, the abnormal determination of the difference in window feature values for a continuous plurality of time windows includes:

[0115] If the difference in window feature values in the current time window is not within the early warning threshold range, then there is an abnormal feature value in the current time window;

[0116] If the number of windows with continuously abnormal feature values is not less than the preset number of abnormal windows, it is determined that there are abnormal window feature values in the subsequent continuous plurality of time windows.

[0117] Optionally, for the sequence of differences in original values, if the number of state data differences within the current window that do not satisfy the early warning threshold range exceeds p*T, where p is a value between (0, 1), an alarm is issued; for the sequence of differences in window feature values, if the differences in feature values of continuously t windows do not satisfy the early warning threshold range, an alarm is issued, where t is an integer greater than or equal to 1.

[0118] It should be noted that the parameters of the time window length T, overlap rate P, ratio p, and number of windows t can be determined according to the specific performance of historical data and / or expert experience, or can be adjusted according to the actual needs of users. Specifically, no special limitation is made here.

[0119] Optionally, the sudden failure early warning can refer to Figure 2 , Figure 2 which is a schematic diagram of a sudden failure early warning method provided by an embodiment of the present application.

[0120] Optionally, after early warning is performed on the target sensors, the original state data and window feature values of multiple sensors among all the target sensors are respectively obtained, and then, according to the original state data and the change trend of the original state data of the sensors, or according to the window feature values and the change trend of the window feature values of the sensors, the sensors with faults are located.

[0121] In some embodiments, if the fault early warning model is a gradual fault early warning model, then the monitoring and early warning of all the target sensors by the fault early warning model according to the operation state data includes:

[0122] According to the operation state data of all the target sensors, obtain the fault probability of the same type of sensors;

[0123] When the failure probability is not less than the preset warning probability threshold, a warning is issued.

[0124] Optionally, the difference value time series is fitted by an auto-regressive and moving average (ARMA model) to establish a function d between the difference value and the series. t =fun(x t ), where x t is the time window sequence traced back from time t. For gradual faults, when the difference value is 0, the fault probability is defined as 0. When the fault occurs, the fault probability is 1. The difference value d at this moment, for time t, the fault probability at time t can be expressed as Pt = d t / d.

[0125] Optionally, the preset warning probability threshold can be set based on expert experience. For example, the preset warning probability threshold is set to k. When the current failure probability of the same type of sensor is greater than the set threshold, a trend fault warning result is output; otherwise, the sensor is normal and monitoring continues; where k is a value between (0, 1).

[0126] Optional, gradual fault warning can refer to Figure 3 , Figure 3 A schematic diagram of a gradual fault warning method provided in an embodiment of the present application.

[0127] It should be noted that for the real-time operating status data of the sensor, the data with a cumulative time length of T is input as a window, and its difference sequence (original status data difference sequence, window eigenvalue difference sequence) is calculated according to the above steps, and then an early warning judgment is made according to the above early warning strategy; if it is judged that there is a fault, the early warning result is output; otherwise, the sensor status data is continuously monitored.

[0128] The fault warning method provided in this embodiment includes: obtaining operating status data from all target sensors, all of which are of the same type; and monitoring and issuing warnings based on the operating status data using a fault warning model. The fault warning model is generated based on the operating status data of multiple sensors of the same type. Based on the operating status data of sensors of the same type on a train, statistical analysis is performed on the differences in status data between sensors of the same type. The variation characteristics and distribution patterns of these differences are then extracted, and a sensor anomaly self-detection method is constructed to provide intelligent warnings for sudden and gradual faults in the train's sensor systems. This provides proactive maintenance advice to after-sales personnel, ensuring safe and efficient train operation.

[0129] Example 2

[0130] This embodiment provides a fault warning device. The device embodiment can be used to execute the method embodiment of this application. For details not disclosed in the device embodiment, please refer to the method embodiment of this application. Figure 4 It is a schematic structural diagram of a device provided by an embodiment of this application, as Figure 4 shown. The device 400 provided in this embodiment includes:

[0131] An acquisition module 401, configured to acquire the operation status data of all target sensors, and all the target sensors are of the same type;

[0132] A warning module 402, configured to monitor and warn all the target sensors according to the operation status data through a fault warning model;

[0133] A generation module 403, configured to generate the fault warning model according to the operation status data of multiple sensors of the same type of the target sensors.

[0134] In some embodiments, the generation module 403 includes: a first acquisition unit, a second acquisition unit, a third acquisition unit, and a generation unit; wherein,

[0135] The first acquisition unit is configured to acquire the operation status data of the multiple sensors of the same type;

[0136] The second acquisition unit is configured to acquire an original status data difference sequence according to the operation status data;

[0137] The third acquisition unit is configured to acquire a window feature value difference sequence according to the operation status data;

[0138] The generation unit is configured to generate the fault warning model according to the original status data difference sequence and the window feature value difference sequence.

[0139] In some embodiments, all the multiple sensors of the same type are sensors that have not had faults in history.

[0140] In some embodiments, the acquisition module 401 includes a filtering unit, configured to perform filtering processing on the operation status data of the multiple sensors of the same type after acquiring the operation status data of the multiple sensors of the same type.

[0141] In some embodiments, the generation unit includes: a first acquisition subunit, an analysis subunit, a second acquisition subunit, and a generation subunit; wherein,

[0142] The first acquisition subunit is configured to acquire a warning threshold interval and a difference value time series according to the original status data difference sequence and the window feature value difference sequence;

[0143] An analysis subunit, configured to perform a stationarity analysis on the difference value time series to obtain an analysis result;

[0144] A second acquisition subunit, configured to acquire historical fault data of the same type of sensors that have had faults;

[0145] A generation subunit, configured to generate a sudden fault warning model according to the original state data difference sequence, the window eigenvalue difference sequence, and the warning threshold interval, or generate a gradual fault warning model according to the analysis result and the historical fault data.

[0146] In some embodiments, the warning module 402 includes: a first acquisition unit, a first determination unit, and a warning unit; wherein, the first acquisition unit is configured to, if the fault warning model is a sudden fault warning model, acquire the original state data difference and the window eigenvalue difference in the current time window from the operation state data of all the target sensors;

[0147] The first determination unit is configured to, if the fault warning model is a sudden fault warning model, perform an anomaly determination on the original state data difference in the current time window and the window eigenvalue differences in a plurality of consecutive time windows;

[0148] The warning unit is configured to, if the fault warning model is a sudden fault warning model, perform a warning when it is determined that there is abnormal state difference data in the current time window or there are abnormal window eigenvalues in a plurality of consecutive time windows.

[0149] In some embodiments, the determination unit includes: an acquisition subunit and a determination subunit; wherein,

[0150] The acquisition subunit is configured to acquire the number of warning messages that the original state data difference in the current time window is not within the warning threshold interval;

[0151] The determination subunit is configured to, if the number of warning messages is not less than a preset number of warning messages, determine that there is abnormal state difference data in the current time window.

[0152] In some embodiments, the anomaly determination on the window eigenvalue differences in a plurality of consecutive time windows includes:

[0153] If the window eigenvalue difference in the current time window is not within the warning threshold interval, then there is an eigenvalue anomaly in the current time window;

[0154] If the number of consecutive windows with eigenvalue anomalies is not less than a preset number of abnormal windows, determine that there are abnormal window eigenvalues in the subsequent plurality of consecutive time windows.

[0155] In some embodiments, the warning module 402 further includes: a second acquisition unit and a second determination unit. Wherein, if the fault warning model is a gradual fault warning model, the second acquisition unit is configured to obtain the fault probability of the same type of sensors according to the operation status data of all the target sensors; the second determination unit is configured to, when the fault probability is not less than a preset warning probability threshold, perform warning through the warning unit.

[0156] It should be noted that the above-mentioned various modules / units can be functional modules or program modules, and can be implemented either by software or by hardware. For the modules / units implemented by hardware, the above-mentioned various modules / units can be located in the same processor; or the above-mentioned various modules / units can also be located in different processors in any combined form.

[0157] The device of this embodiment includes: an acquisition module 401, configured to acquire the operation status data of all target sensors, and all the target sensors are of the same type; a warning module 402, configured to monitor and warn all the target sensors according to the operation status data through a fault warning model; a generation module 403, configured to generate the fault warning model according to the operation status data of multiple same-type sensors of the target sensors. Based on the operation status data of train same-type sensors, statistically analyze the differences in status data between the same-type sensors, then extract the change characteristics and distribution rules of the differences, and further construct a set of sensor anomaly self-checking methods to realize the intelligent warning of sudden and gradual faults in the train same-type sensor system, provide suggestions in advance for the maintenance work of after-sales personnel, and ensure the safe and efficient operation of the train.

[0158] Embodiment III

[0159] This embodiment also provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, it can implement the method steps as in Embodiment I, and this embodiment will not be repeated here.

[0160] Among them, the storage medium may also separately include computer programs, data files, data structures, etc., or include a combination thereof. The storage medium or computer program can be specifically designed and understood by those skilled in the field of computer software, or the storage medium can be known and available to those skilled in the field of computer software. Examples of storage media include: magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROM discs and DVDs; magneto-optical media such as optical discs; and hardware devices specifically configured to store and execute computer programs, such as read-only memory (ROM), random access memory (RAM), flash memory; or servers, app application stores, etc. Examples of computer programs include machine code (e.g., code generated by a compiler) and files containing high-level code that can be executed by a computer by using an interpreter. The described hardware devices can be configured to act as one or more software modules to perform the operations and methods described above, and vice versa. Additionally, the storage medium can be distributed in a networked computer system and can store and execute program code or computer programs in a decentralized manner.

[0161] Embodiment 4

[0162] Figure 5 The following is a connection block diagram of an electronic device provided by an embodiment of this application, as Figure 5 shown. The electronic device 500 may include: a processor 501, a memory 502, a multimedia component 503, an input / output (I / O) interface 504, and a communication component 505.

[0163] Among them, the processor 501 is used to execute all or part of the steps in the method of Embodiment 1. The memory 502 is used to store various types of data, which may include, for example, instructions of any application program or method in the electronic device, as well as data related to the application program.

[0164] The processor 501 can be implemented by an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic components, and is used to execute the method in Embodiment 1 above.

[0165] The memory 502 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, a magnetic disk, or an optical disc.

[0166] The multimedia component 503 may include a screen and an audio component. The screen can be a touch screen. The audio component is used for outputting and / or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals can be further stored in the memory or sent through the communication component. The audio component also includes at least one speaker for outputting audio signals.

[0167] The I / O interface 504 provides an interface between the processor 501 and other interface modules, and the other interface modules can be a keyboard, a mouse, buttons, etc. These buttons can be virtual buttons or physical buttons.

[0168] The communication component 505 is used for wired or wireless communication between the electronic device 500 and other devices. Wired communication includes communication through a network port, a serial port, etc.; wireless communication includes Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, 4G, 5G, or a combination of one or more of them. Accordingly, the communication component 505 may include a Wi-Fi module, a Bluetooth module, and an NFC module.

[0169] In summary, a fault warning method, device, storage medium, and electronic device provided by the present application. The method includes: obtaining the operating status data of all target sensors, where all the target sensors are of the same type; monitoring and warning all the target sensors according to the operating status data through a fault warning model; wherein, the fault warning model is generated according to the operating status data of multiple sensors of the same type of the target sensor. Based on the operating status data of train sensors of the same type, statistical analysis is performed on the differences in the status data between sensors of the same type, and then the change characteristics and distribution rules of the differences are extracted. Furthermore, a set of sensor anomaly self-checking methods is constructed to achieve intelligent warning of sudden and gradual faults in the train sensor system of the same type, provide suggestions in advance for the maintenance work of after-sales personnel, and ensure the safe and efficient operation of the train.

[0170] In addition, it should be understood that the methods or systems disclosed in the embodiments provided in the present application can also be implemented in other ways. The method or system embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of the methods and devices according to multiple embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a computer program segment, or a part of a computer program. A module, a computer program segment, or a part of a computer program contains one or more computer programs for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings, and may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and a computer program.

[0171] In this application, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, method, apparatus or device comprising the element; if there is a description of "first", "second", etc., it is only for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features or implicitly specifying the sequence of the indicated technical features; in the description of this application, unless otherwise specified, the meaning of the terms "a plurality", "many" is at least two; if there is a description of a server, it should be noted that the server can be an independent physical server or terminal, or a server cluster composed of multiple physical servers, and can be a cloud server capable of providing basic cloud computing services such as cloud servers, cloud databases, cloud storage and CDN; in this application, if there is a description of a smart terminal or mobile device, it should be noted that the smart terminal or mobile device can be a mobile phone, tablet computer, smart watch, netbook, wearable electronic device, personal digital assistant (Personal Digital Assistant, PDA), augmented reality device (Augmented Reality, AR), virtual reality device (Virtual Reality, VR), smart TV, smart speaker, personal computer (Personal Computer, PC), etc., but is not limited thereto, and this application does not make special limitations on the specific form of the smart terminal or mobile device.

[0172] Finally, it should be noted that in the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "one example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0173] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are all exemplary. The described content is only an implementation manner adopted for the convenience of understanding the present application, and is not used to limit the present application. Any person skilled in the art within the technical field to which the present application pertains may make any modifications and changes in the form of implementation and details without departing from the spirit and scope disclosed by the present application. However, the protection scope of the present application shall still be subject to the scope defined by the appended claims.

Claims

1. A fault warning method, characterized in that, The method includes: Obtaining the operation status data of all target sensors, where all the target sensors are of the same type; the operation status data is real-time operation status data; Monitoring and warning all the target sensors according to the operation status data through a fault warning model; Among them, the fault warning model is generated according to the operation status data of multiple sensors of the same type of the target sensor, and all the multiple sensors of the same type have never had a fault in history. The generation steps of the fault warning model include: Obtaining the operation status data of the multiple sensors of the same type; Obtaining the original state data difference sequence according to the operation status data; Obtaining the window feature value difference sequence according to the operation status data; Generating the fault warning model according to the original state data difference sequence and the window feature value difference sequence.

2. The method according to claim 1, characterized in that After obtaining the operation status data of the multiple sensors of the same type, it further includes: Filtering the operation status data of the multiple sensors of the same type.

3. The method according to claim 1, characterized in that The generating the fault warning model according to the original state data difference sequence and the window feature value difference sequence includes: Obtaining the warning threshold interval and the difference value time series according to the original state data difference sequence and the window feature value difference sequence; Performing a stationarity analysis on the difference value time series to obtain an analysis result; Obtaining the historical fault data of the sensors of the same type that have had a fault; Generating a sudden fault warning model according to the original state data difference sequence, the window feature value difference sequence and the warning threshold interval, or generating a gradual fault warning model according to the analysis result and the historical fault data.

4. The method according to claim 1, characterized in that If the fault warning model is a sudden fault warning model, then the monitoring and warning all the target sensors according to the operation status data through the fault warning model includes: Obtaining the original state data difference and the window feature value difference in the current time window from the operation status data of all the target sensors; Performing an anomaly determination on the original state data difference in the current time window and the window feature value differences of a continuous plurality of time windows; When it is determined that there is abnormal state difference data in the current time window or there are abnormal window feature values in a continuous plurality of time windows, a warning is issued.

5. The method according to claim 4, wherein The performing an anomaly determination on the original state data difference in the current time window includes: Obtaining the number of warning items where the original state data difference in the current time window is not within the warning threshold interval; If the number of warning items is not less than a preset number of warning items, it is determined that there is abnormal state difference data in the current time window.

6. The method according to claim 4, wherein The performing an anomaly determination on the window feature value differences of a continuous plurality of time windows includes: If the window feature value difference in the current time window is not within the warning threshold interval, then there is a feature value anomaly in the current time window; If the number of windows with continuous feature value anomalies is not less than a preset number of abnormal windows, it is determined that there are abnormal window feature values in the subsequent continuous plurality of time windows.

7. The method according to claim 1, wherein If the fault warning model is a gradual fault warning model, then the monitoring and warning of all target sensors by the fault warning model according to the operation state data includes: Obtaining the fault probability of the same type of sensors according to the operation state data of all target sensors; When the fault probability is not less than the preset warning probability threshold, a warning is issued.

8. A fault warning device, characterized in that, Including: An acquisition module for acquiring the operation state data of all target sensors, and all target sensors are of the same type; The operation state data is real-time operation state data; A warning module for monitoring and warning all target sensors according to the operation state data through a fault warning model; A generation module for generating the fault warning model according to the operation state data of multiple same-type sensors of the target sensors. The multiple same-type sensors have never had faults in history. The generation module includes: a first acquisition unit, a second acquisition unit, a third acquisition unit, and a generation unit; wherein, The first acquisition unit is used for acquiring the operation state data of the multiple same-type sensors; The second acquisition unit is used for obtaining the original state data difference sequence according to the operation state data; The third acquisition unit is used for obtaining the window feature value difference sequence according to the operation state data; The generation unit is used for generating the fault warning model according to the original state data difference sequence and the window feature value difference sequence.

9. A storage medium, characterized in that, The computer program stored in the storage medium, when executed by one or more processors, implements the fault warning method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, Including a memory and a processor, a computer program is stored on the memory, and the memory and the processor are communicatively connected to each other. When the computer program is executed by the processor, the fault warning method according to any one of claims 1 to 7 is executed.

Images