Alarm Prediction Method, Device and Computer Readable Storage Medium

Abstract

The present application discloses an alarm prediction method, apparatus, and computer-readable storage medium, relating to the technical field of device alarms. The alarm prediction method includes: obtaining relevant information of a target monitoring device, where the relevant information includes at least one of a data capture rate, a manual intervention rate, and a license plate recognition rate; inputting the relevant information into a preset neural network model to obtain a prediction result for the target monitoring device, where the prediction result is used to predict whether the target monitoring device will issue an alarm. Based on the above method, it is beneficial to improve the reliability of the monitoring device.

Description

Technical Field

[0001] This application relates to the technical field of device alarm, in particular to an alarm prediction method, device and computer-readable storage medium. Background Art

[0002] In the prior art, a corresponding alarm module is usually equipped for a monitoring device. When the monitoring device fails, the alarm module is triggered to give an alarm to prompt the corresponding maintenance personnel to maintain the device.

[0003] The defect of the prior art is that when the monitoring device alarms, the fault usually has occurred. At this time, even if the corresponding maintenance personnel are prompted to maintain, a large amount of data lost due to the fault cannot be retrieved, resulting in poor reliability of the monitoring device. Summary of the Invention

[0004] The main technical problem to be solved by this application is how to improve the reliability of the monitoring device.

[0005] To solve the above technical problem, the first technical solution adopted by this application is: to provide an alarm prediction method, including: obtaining relevant information of a target monitoring device, where the relevant information includes at least one of a data capture rate, an artificial intervention rate, and a license plate recognition rate. The data capture rate is the proportion of parking events recognized by the monitoring device in the total parking events. The artificial intervention rate is the proportion of parking events processed manually in the total parking events. The license plate recognition rate is the proportion of the number of license plates correctly recognized by the monitoring device in the total number of license plates; inputting the relevant information into a preset neural network model to obtain a prediction result for the target monitoring device, where the prediction result is used to predict whether the target monitoring device will alarm.

[0006] To solve the above technical problem, the second technical solution adopted by this application is: to provide an alarm prediction device, including: a memory and a processor; the memory is used to store program instructions, and the processor is used to execute the program instructions to implement the above alarm prediction method.

[0007] To solve the above technical problem, the third technical solution adopted by this application is: to provide a computer-readable storage medium, where the computer-readable storage medium stores program instructions, and when the program instructions are executed by a processor, the above alarm prediction method is implemented.

[0008] Compared with the prior art, the beneficial effects of the present application are as follows: By obtaining at least one of the data capture rate, manual intervention rate, and license plate recognition rate of the target monitoring device, and inputting at least one of the obtained data capture rate, manual intervention rate, and license plate recognition rate information into a preset neural network model to obtain a prediction result for the target monitoring device, and then predicting whether the target monitoring device will alarm. Based on the above method, it is possible to predict whether the target monitoring device may be about to alarm when it has not alarmed, so that corresponding preparations for alarm can be made in advance according to the prediction result, avoiding data loss of the target monitoring device and improving the reliability of the target monitoring device. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0010] Figure 1 is a schematic flowchart of an embodiment of the alarm prediction method of the present application;

[0011] Figure 2 is a schematic flowchart of an embodiment of some steps of the alarm prediction method of the present application;

[0012] Figure 3 is a schematic structural diagram of an embodiment of the AlexNet network model of the present application;

[0013] Figure 4 is a schematic flowchart of another embodiment of the alarm prediction method of the present application;

[0014] Figure 5 is a schematic structural diagram of an embodiment of the alarm prediction device of the present application;

[0015] Figure 6 is a schematic structural diagram of an embodiment of the computer-readable storage medium of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0017] The terms "first" and "second" in this application are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. In the description of this application, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include unlisted steps or units, or may optionally further include other steps or units inherent to these processes, methods, products, or devices.

[0018] As Figure 1 shown, in the first embodiment, this application proposes an alarm prediction method, including:

[0019] Step S11: Obtain relevant information of the target monitoring device.

[0020] Among them, the relevant information includes at least one of the data capture rate, the manual intervention rate, and the license plate recognition rate. The data capture rate is the proportion of parking events recognized by the monitoring device in the total parking events. The manual intervention rate is the proportion of parking events processed manually in the total parking events. The license plate recognition rate is the proportion of the number of license plates correctly recognized by the monitoring device in the total number of license plates.

[0021] A parking event may be an event in which a vehicle is blocked and stops moving due to a recognition problem of the monitoring device. For example, when the monitoring device misidentifies a vehicle as an indebted vehicle and does not allow the vehicle to leave the parking lot, but the vehicle is not an indebted vehicle, this will result in a parking event. The parking events processed manually may specifically include parking events not recognized by the monitoring device and / or parking events misrecognized by the monitoring device. Such events usually require human intervention for processing.

[0022] Optionally, step S11 may specifically include:

[0023] Obtain the relevant information of the target monitoring device every first preset time interval.

[0024] Specifically, the relevant information of the target monitoring device can be updated once every first preset time interval, and the subsequent step S12 is triggered to perform a prediction on the target monitoring device based on the updated relevant information to ensure the real-time nature of the target monitoring device and improve the accuracy of the prediction result. The first preset time interval can be 1 minute or 5 minutes or any time length, which is not limited here.

[0025] Optionally, as Figure 2 shown, before step S11, the alarm prediction method may specifically further include:

[0026] Step S21: Obtain training data.

[0027] Among them, the training data includes abnormal data and normal data. The abnormal data includes the relevant information of a monitoring device within a second preset duration before an alarm, and the normal data includes the relevant information of a monitoring device during normal operation. The second preset duration can be 10 minutes, 30 minutes, or any duration, which is not limited here.

[0028] It should be noted that since the operation to be performed by this application based on the neural network model is alarm prediction, therefore, the abnormal data in the training data needs to include at least the relevant information of the corresponding monitoring device within a period of time before the alarm. Based on this, alarm prediction can be performed based on the neural network model trained using this training data.

[0029] Step S22: Preprocess the training data.

[0030] Among them, the preprocessing at least includes performing corresponding annotations on the abnormal data and normal data. Before training based on the training data, it is first necessary to label the different training data as abnormal or normal, and then corresponding training can be performed based on this label, so that the neural network model can be trained to have the ability of alarm prediction.

[0031] Step S23: Input the preprocessed training data into the neural network model to train the neural network model.

[0032] Among them, the training data can be input into the neural network model in batches for training until the alarm prediction accuracy rate of the neural network model is not less than the preset accuracy rate. The preset accuracy rate can be 95% or other preset probabilities, which is not limited here.

[0033] Specifically, the preprocessed training data can be divided into a training set, a test set, and a validation set. The training set is used to train the neural network model, and both the test set and the validation set are used to test or verify the alarm prediction accuracy rate of the trained neural network model.

[0034] If after all the training data is input into the neural network model for training, the alarm prediction accuracy rate of the neural network model still cannot reach the preset accuracy rate, then the neural network model can be changed from the current model framework to other model frameworks and retrained based on the training data. For example, the neural network model can be an AlexNet (Alex Network) model. If a model with an alarm prediction accuracy rate not less than the preset accuracy rate cannot be trained based on the AlexNet model and the training data, then the AlexNet model can be replaced with other types of neural network models and retrained based on the training data until a model with an alarm prediction accuracy rate not less than the preset accuracy rate is obtained. What specific model framework the neural network model is, is not limited here.

[0035] As Figure 3 shown, the AlexNet network model 30 may specifically include: a first convolutional layer 31, a second convolutional layer 32, a third convolutional layer 33, a fourth convolutional layer 34, a fifth convolutional layer 35, a first fully connected layer 36, a second fully connected layer 37, and a third fully connected layer 38.

[0036] In addition, further, if the alarm prediction accuracy rate of the neural network model still cannot reach the preset accuracy rate after all the training data is input into the neural network model for training, a check prompt may be output first to notify relevant personnel to check whether there are any problems during the training process. If there are no problems, the neural network model can be replaced from the current model framework to other model frameworks and retrained based on the training data. The possible problems during the training process include: whether the selected parameters as training data are reasonable, whether the operation-related parameters of the parking lot system need to be added to the training data, and whether the data cleaning method in the preprocessing needs to be adjusted.

[0037] Further, step S22 may specifically include:

[0038] Performing data cleaning on the training data.

[0039] Specifically, the steps of performing data cleaning on the training data may specifically include:

[0040] Normalizing the data format of the training data, and / or

[0041] Removing duplicates from the training data, and / or

[0042] Replacing the error values or abnormal values in the training data with corresponding average values, and / or

[0043] Normalizing the training data.

[0044] Specifically, after performing data cleaning on the training data, the data consistency of the training data can be effectively improved, and further, the training efficiency when training the neural network model based on the training data and the alarm prediction accuracy rate of the trained model can be improved.

[0045] Optionally, the relevant information may further include at least one of working hours, alarm times, alarm duration, alarm delay, and occupied storage space.

[0046] Specifically, the alarm times may include at least one of the alarm times when there is no memory card, the alarm times for memory card errors, the alarm times for insufficient memory card storage space, the alarm times for network interruption, the alarm times for Internet protocol conflicts, the alarm times for illegal access, the alarm times for security exceptions, and the alarm times for other types, which are not limited here.

[0047] Step S12: Input relevant information into a preset neural network model to obtain a prediction result for the target monitoring device.

[0048] Among them, the prediction result is used to predict whether the target monitoring device will issue an alarm. After the preset neural network model processes the input relevant information, it can output the prediction result of whether the target monitoring device will alarm. Based on this prediction result, it can be judged whether the target monitoring device may be about to alarm. If so, the data of the target monitoring device can be saved in advance and replacement parts can be prepared to ensure that data loss will not be caused by factors such as device alarms and failures. Compared with traditional methods, the alarm prediction based on the neural network model can have better generalization performance, enabling the platform device that executes this prediction to run continuously and stably.

[0049] Optionally, after the step of inputting relevant information into a preset neural network model to obtain a prediction result for the target monitoring device, the alarm prediction method further includes:

[0050] Judge whether there is an unpredicted monitoring device currently, where the unpredicted monitoring device is a monitoring device for which alarm prediction has not been performed yet.

[0051] If there is an unpredicted monitoring device, use the unpredicted monitoring device as the target monitoring device, execute the steps of obtaining the relevant information of the target monitoring device and inputting the relevant information into a preset neural network model to obtain a prediction result for the target monitoring device, and return to execute the step of judging whether there is an unpredicted monitoring device currently.

[0052] Specifically, based on the above method, each monitoring device in a batch of monitoring devices can be predicted for alarm one by one until all monitoring devices have been predicted, which can avoid missing the alarm prediction of a certain monitoring device and improve the reliability of the alarm prediction method.

[0053] As Figure 4 shown, in an application scenario, the alarm prediction method may include:

[0054] Step S31: Obtain training data.

[0055] Step S32: Preprocess the training data.

[0056] Step S33: Input the preprocessed training data into the neural network model to train the neural network model.

[0057] Step S34: Obtain the relevant information of the target monitoring device.

[0058] Step S35: Input the relevant information into the trained neural network model to obtain a prediction result for the target monitoring device.

[0059] Step S36: Determine whether there is an unforecast monitoring device currently.

[0060] If the determination result of step S36 is yes, that is, there is an unforecast monitoring device, then execute step S37. If the determination result of step S36 is no, that is, there is no unforecast monitoring device, then execute step S38.

[0061] Step S37: Use the unforecast monitoring device as the target monitoring device, and then return to execute step S34 and subsequent steps.

[0062] Step S38: End to complete all forecasts of the monitoring devices involved in this round of forecasting.

[0063] Different from the prior art, the present application obtains at least one of the data capture rate, manual intervention rate, and license plate recognition rate of the target monitoring device, and inputs at least one of the obtained data capture rate, manual intervention rate, and license plate recognition rate information into a preset neural network model to obtain a prediction result for the target monitoring device, thereby predicting whether the target monitoring device will issue an alarm. Based on the above method, it is possible to predict whether the target monitoring device may be about to issue an alarm when it has not issued an alarm, so that corresponding preparations for the alarm can be made in advance according to the prediction result, avoiding data loss of the target monitoring device, and improving the reliability of the target monitoring device.

[0064] As Figure 5 shown, in the second embodiment, the present application proposes an alarm prediction device. The alarm prediction device 50 of this embodiment includes: a processor 51, a memory 52, and a bus 53.

[0065] The processor 51 and the memory 52 are respectively connected to the bus 53. The memory 52 stores program instructions, and the processor 51 is used to execute the program instructions to implement the alarm prediction method in the above embodiment.

[0066] In this embodiment, the processor 51 can also be referred to as a CPU (Central Processing Unit, central processing unit). The processor 51 may be an integrated circuit chip with signal processing capabilities. The processor 51 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor 51 may also be any conventional processor, etc.

[0067] Different from the prior art, the present application obtains at least one of the data capture rate, manual intervention rate, and license plate recognition rate of the target monitoring device, and inputs at least one of the obtained data capture rate, manual intervention rate, and license plate recognition rate information into a preset neural network model to obtain a prediction result for the target monitoring device, and further predicts whether the target monitoring device will issue an alarm. Based on the above method, it is possible to predict whether the target monitoring device may be about to issue an alarm when the target monitoring device has not issued an alarm, so that the corresponding preparation for the alarm can be made in advance according to the prediction result, avoiding the loss of data of the target monitoring device and improving the reliability of the target monitoring device.

[0068] As Figure 6 shown, in the third embodiment, the present application proposes a computer-readable storage medium. On the computer-readable storage medium 60, there is stored a program instruction 61. When the program instruction 61 is executed by a processor (not shown in the figure), the alarm prediction method in the above embodiment is implemented.

[0069] The computer-readable storage medium 60 in this embodiment may be, but is not limited to, a USB flash drive, an SD card, a PD optical drive, a mobile hard disk, a large-capacity floppy drive, a flash memory, a multimedia memory card, a server, etc.

[0070] Different from the prior art, the present application obtains at least one of the data capture rate, manual intervention rate, and license plate recognition rate of the target monitoring device, and inputs at least one of the obtained data capture rate, manual intervention rate, and license plate recognition rate information into a preset neural network model to obtain a prediction result for the target monitoring device, and further predicts whether the target monitoring device will issue an alarm. Based on the above method, it is possible to predict whether the target monitoring device may be about to issue an alarm when the target monitoring device has not issued an alarm, so that the corresponding preparation for the alarm can be made in advance according to the prediction result, avoiding the loss of data of the target monitoring device and improving the reliability of the target monitoring device.

[0071] In the description of the present application, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean 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 the present 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 may be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0072] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In the description of this application, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0073] Any process or method description represented in a flowchart or described otherwise herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a specific logical function or process. The scope of the preferred embodiments of this application includes additional implementations, where functions may be executed in a manner not shown or discussed, including substantially concurrently according to the involved functions or in a reverse order, which should be understood by those skilled in the technical field to which the embodiments of this application pertain.

[0074] The logic and / or steps represented in a flowchart or described otherwise herein, for example, can be considered as an ordered list of executable instructions for implementing a logical function and can be specifically implemented in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device (which can be a personal computer, server, network device, or other system that can fetch and execute instructions from the instruction execution system, apparatus, or device). For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, a computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0075] The above description is only the embodiment of this application and does not limit the patent scope of this application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of this application, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of this application.

Claims

1. An alarm prediction method, characterized in that, Including: Obtain training data, where the training data includes abnormal data and normal data, the abnormal data includes relevant information of a monitoring device within a second preset duration before an alarm, and the normal data includes relevant information of a monitoring device during normal operation; Preprocess the training data, where the preprocessing at least includes performing corresponding annotations on the abnormal data and the normal data; Input the training data that has undergone the preprocessing into a preset neural network model to train the neural network model; Every first preset duration, obtain relevant information of a target monitoring device, where the relevant information includes at least one of a data capture rate, a manual intervention rate, and a license plate recognition rate. The data capture rate is the proportion of parking events recognized by the monitoring device in the total parking events, the manual intervention rate is the proportion of parking events processed manually in the total parking events, and the license plate recognition rate is the proportion of the number of license plates correctly recognized by the monitoring device in the total number of license plates; Input the relevant information into the neural network model to obtain a prediction result for the target monitoring device, where the prediction result is used to predict whether the target monitoring device will alarm; Determine whether there is an un-predicted monitoring device currently, where the un-predicted monitoring device is a monitoring device for which alarm prediction has not been performed yet; If there is the un-predicted monitoring device, use the un-predicted monitoring device as the target monitoring device, execute the steps of obtaining the relevant information of the target monitoring device and inputting the relevant information into the preset neural network model to obtain a prediction result for the target monitoring device, and return to execute the step of determining whether there is an un-predicted monitoring device currently.

2. The alarm prediction method according to claim 1, wherein The step of preprocessing the training data includes: Perform data cleaning on the training data.

3. The alarm prediction method according to claim 2, characterized in that The step of performing data cleaning on the training data includes: Normalize the data format of the training data, and / or Deduplicate the training data, and / or Replace error values or abnormal values in the training data with corresponding average values, and / or Normalize the training data.

4. The alarm prediction method according to claim 1 or 2, characterized in that The relevant information further includes at least one of a working duration, an alarm frequency, an alarm duration, an alarm delay, and an occupied storage space.

5. The alarm prediction method according to claim 1 or 2, characterized in that, The neural network model is an AlexNet network model.

6. An alarm prediction device, characterized in that, Including: A memory and a processor; The memory is used to store program instructions, and the processor is used to execute the program instructions to implement the method according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program instructions, and when the program instructions are executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images