Alarm methods, devices, equipment, storage media, and computer program products

Abstract

This application relates to an alarm method, apparatus, device, storage medium, and computer program product, relating to the field of information security, and also applicable to the financial sector and other related fields. The method includes: acquiring alarm information sent by a monitoring system; the alarm information including alarm time and alarm content; inputting the alarm information into a pre-constructed fuzzy neural network model to obtain an alarm handling method output by the fuzzy neural network model; and executing the processing operation corresponding to the alarm handling method. This method can determine the alarm handling method corresponding to the alarm information output by the monitoring system based on a fuzzy neural network model, and execute the processing operation corresponding to the alarm handling method, thereby resolving business system failures and ensuring the stable operation of the business system.

Description

Technical Field

[0001] This application relates to the field of information security, and in particular to an alarm method, apparatus, device, storage medium, and computer program product. Background Technology

[0002] Business systems are crucial to a company's business operations, and the realization of various business activities depends on the stable operation of business systems.

[0003] A centralized monitoring system can manage software or hardware devices in a business system, present various events in a standardized manner, monitor the usage of equipment resources and the operating status of the business system, and promptly identify potential risks that may cause business system failures.

[0004] Therefore, how to use a centralized monitoring system to ensure the safe and stable operation of business systems is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] This application provides an alarm method, apparatus, device, storage medium, and computer program product, which can determine the alarm processing method corresponding to the alarm information output by the monitoring system based on a fuzzy neural network model, and execute the processing operation corresponding to the above alarm processing method to solve the business system failure problem and ensure the stable operation of the business system.

[0006] Firstly, this application provides an alarm method. The method includes:

[0007] Obtain alarm information sent by the monitoring system; alarm information includes alarm time and alarm content;

[0008] The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model;

[0009] Perform the corresponding processing operation for the alarm handling method.

[0010] In one embodiment, alarm information is input into a pre-built fuzzy neural network model to obtain an alarm processing method output by the fuzzy neural network model, including: inputting alarm content and alarm time into the fuzzy neural network model; if the alarm content includes a medium-level alarm that is not in maintenance period, then an immediate processing method output by the fuzzy neural network model is obtained; correspondingly, the processing operation corresponding to the alarm processing method is executed, including: immediately executing the processing operation corresponding to the alarm information according to a preset correlation.

[0011] In one embodiment, alarm information is input into a pre-built fuzzy neural network model to obtain an alarm handling method output by the fuzzy neural network model, including: inputting alarm content and alarm time into the fuzzy neural network model; if the alarm content includes one of a high-level alarm during non-maintenance period, a high-level alarm during maintenance period, and a medium-level alarm during maintenance period, or if the alarm content includes a low-level alarm and the alarm time includes non-working periods, then a notification handling method output by the fuzzy neural network model is obtained; correspondingly, the processing operation corresponding to the alarm handling method is executed, including: obtaining a target maintenance terminal from a pre-set set of multiple maintenance terminals; and sending the alarm information to the target maintenance terminal.

[0012] In one embodiment, alarm information is input into a pre-built fuzzy neural network model to obtain an alarm handling method output by the fuzzy neural network model, including: inputting alarm content and alarm time into the fuzzy neural network model; if the alarm content includes a low-level alarm and the alarm time includes a working period, then an ignore handling method output by the fuzzy neural network model is obtained; correspondingly, the processing operation corresponding to the alarm handling method is executed, including: performing an ignore operation on the alarm information.

[0013] In one embodiment, the fuzzy neural network model includes a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer. The alarm information is input into the pre-constructed fuzzy neural network model, and the alarm handling method output by the model is obtained. This includes: inputting the alarm time and alarm content into the fuzzification layer for fuzzification processing to obtain fuzzy values ​​corresponding to the alarm time and alarm content; defuzzifying the fuzzy values ​​corresponding to the alarm time and alarm content to obtain clear values ​​corresponding to the alarm time and alarm content; inputting the clear values ​​corresponding to the alarm time and alarm content into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and alarm content; inputting the multiple membership degrees corresponding to the alarm time and alarm content into the rule layer for second data processing to obtain fuzzy values ​​corresponding to the alarm information; and inputting the fuzzy values ​​corresponding to the alarm information into the defuzzification layer for defuzzification processing to obtain the alarm handling method.

[0014] In one embodiment, the membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. The clear values ​​corresponding to the alarm time and the clear values ​​corresponding to the alarm content are input into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content. This includes: determining the membership function corresponding to the alarm time and the membership function corresponding to the alarm content respectively; inputting the clear values ​​corresponding to the alarm time into the corresponding membership functions respectively to obtain multiple membership degrees corresponding to the alarm time; and inputting the clear values ​​corresponding to the alarm content into the corresponding membership functions respectively to obtain multiple membership degrees corresponding to the alarm content.

[0015] In one embodiment, the rule layer includes multiple rules. Multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content are input into the rule layer for second data processing to obtain the fuzzy value corresponding to the alarm information. This includes: determining the fuzzy value corresponding to the alarm information based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

[0016] In one embodiment, the fuzzy value corresponding to the alarm information is input into the defuzzification layer for defuzzification processing to obtain the alarm processing method, including: determining the fuzzy value range of the fuzzy value corresponding to the alarm information; and determining the alarm processing method corresponding to the alarm information according to a preset mapping relationship.

[0017] In one embodiment, the process of constructing a fuzzy neural network model includes: obtaining the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, a preset membership function, undetermined parameters, and a preset fuzzy control rule; constructing an initial model based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rule; determining the target value of the undetermined parameters using a preset algorithm; and substituting the target value of the undetermined parameters into the initial model to obtain the fuzzy neural network model.

[0018] In one embodiment, a preset algorithm is used to determine the target value of the parameter to be determined, including: establishing an objective function with the goal of minimizing the output error; using a backpropagation algorithm to iteratively calculate the objective function to obtain the output error corresponding to each iteration; if the output error meets the iteration termination condition, the iteration calculation is terminated, and the value of the parameter to be determined at the end of the iteration calculation is determined as the target value of the parameter to be determined.

[0019] Secondly, this application also provides an alarm device. The device includes:

[0020] The acquisition module is used to acquire alarm information sent by the monitoring system; the alarm information includes the alarm time and alarm content.

[0021] The input module is used to input alarm information into a pre-built fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model.

[0022] The execution module is used to perform the processing operations corresponding to the alarm handling method.

[0023] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0024] Obtain alarm information sent by the monitoring system; alarm information includes alarm time and alarm content;

[0025] The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model;

[0026] Perform the corresponding processing operation for the alarm handling method.

[0027] Fourthly, this application also provides a computer-readable storage medium. This computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0028] Obtain alarm information sent by the monitoring system; alarm information includes alarm time and alarm content;

[0029] The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model;

[0030] Perform the corresponding processing operation for the alarm handling method.

[0031] Fifthly, this application also provides a computer program product. This computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0032] Obtain alarm information sent by the monitoring system; alarm information includes alarm time and alarm content;

[0033] The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model;

[0034] Perform the corresponding processing operation for the alarm handling method.

[0035] This application provides an alarm method, apparatus, device, storage medium, and computer program product, relating to the field of information security, and also applicable to the financial sector and other related fields. The method provided in this application can acquire alarm information monitored by a monitoring system, analyze the alarm information based on a pre-established fuzzy neural network model to obtain the corresponding alarm handling method, and then execute the corresponding processing operation. Therefore, this application can promptly analyze and process alarm information corresponding to business system faults monitored by the monitoring system to resolve business system fault problems and ensure the safe and stable operation of the business system. Attached Figure Description

[0036] Figure 1 This is a diagram illustrating the application environment of an alarm method in one embodiment;

[0037] Figure 2 This is a flowchart illustrating an alarm method in one embodiment;

[0038] Figure 3 This is another flowchart illustrating the alarm method in one embodiment;

[0039] Figure 4 This is another flowchart illustrating the alarm method in one embodiment;

[0040] Figure 5 This is another flowchart illustrating the alarm method in one embodiment;

[0041] Figure 6 This is another flowchart illustrating the alarm method in one embodiment;

[0042] Figure 7 This is another flowchart illustrating the alarm method in one embodiment;

[0043] Figure 8 This is another flowchart illustrating the alarm method in one embodiment;

[0044] Figure 9 This is another flowchart illustrating the alarm method in one embodiment;

[0045] Figure 10 This is a schematic diagram of the initial model in one embodiment;

[0046] Figure 11 This is another flowchart illustrating the alarm method in one embodiment;

[0047] Figure 12 This is a schematic diagram of the initial model iteration calculation process in one embodiment;

[0048] Figure 13 This is a structural block diagram of an alarm device in one embodiment;

[0049] Figure 14 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0051] The alarm method provided in this application embodiment can be applied to, for example... Figure 1 The application environment is shown. Server 10 communicates with Server 20 via a network. A data storage system can store the data that Server 20 needs to process. The data storage system can be integrated onto Server 20 or located in the cloud or on another network server. Server 10 can collect alarm information and send it to Server 20. Server 20 can analyze the alarm information to resolve business system failures. The monitoring system runs on Server 10, and the business system can be a banking business system. Both Server 10 and Server 20 can be implemented using independent servers or a server cluster consisting of multiple servers.

[0052] In one embodiment, such as Figure 2 As shown, an alarm method is provided, which can be applied to... Figure 1 Taking server 20 as an example, the explanation includes the following steps:

[0053] Step 101: Obtain alarm information sent by the monitoring system. Alarm information includes alarm time and alarm content.

[0054] The alarm information can be a specific time or a vague term, such as morning, afternoon, or evening. The alarm content can include the identifier of the faulty hardware or software, the status of the faulty hardware or software, the cause of the fault, and the specific details of the fault.

[0055] In this embodiment of the application, the server 20 needs to analyze the faults of the business system based on alarm information. Therefore, when the monitoring system detects a fault in the business system, it can collect the fault information as alarm content, use the fault time as alarm time, and then package the alarm time and alarm content into alarm information and send it to the server 20, so that the server 20 can obtain alarm information related to the faults in the business system.

[0056] Step 102: Input the alarm information into the pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model.

[0057] In one embodiment, server 20 can obtain all historical alarm information of the business system within a preset historical time period, as well as the alarm handling methods corresponding to the historical alarm information, and then perform model training based on the above historical alarm information and the alarm handling methods corresponding to the historical alarm information, so that the trained model can determine the alarm handling method corresponding to the alarm information based on the input alarm information.

[0058] In one possible implementation, a fuzzy control model can be constructed based on preset rules, and the alarm handling method corresponding to the input alarm information can be determined according to the constructed fuzzy control model.

[0059] In one possible implementation, a fuzzy control model can be constructed first based on preset rules, then the fuzzy control model can be optimized based on a neural network to obtain a fuzzy neural network model, and finally the alarm handling method corresponding to the input alarm information can be determined based on the optimized fuzzy neural network model.

[0060] The preset rules are the mapping relationship between alarm handling methods and alarm information, that is, the mapping relationship between alarm information and alarm time and alarm content.

[0061] In this embodiment of the application, after receiving the alarm information sent by the monitoring system, the server 20 can input the alarm information into a pre-built fuzzy neural network model and collect the output of the fuzzy neural network model, which is the alarm processing method corresponding to the alarm information.

[0062] Step 103: Execute the processing operation corresponding to the alarm handling method.

[0063] Alarm handling methods can include immediate handling, notification handling, and ignoring handling. Handling operations can include deleting fault source data, ending the operation of the faulty system or device, notifying maintenance personnel via email or telephone, and storing alarm information.

[0064] In one embodiment, server 20 can store the correspondence between processing operations and alarm information and alarm processing methods. After obtaining the alarm processing method corresponding to the alarm information, server 20 can determine the processing operation based on the above correspondence and execute the processing operation.

[0065] The alarm method provided in this application can acquire alarm information monitored by the monitoring system, analyze the alarm information based on a pre-established fuzzy neural network model, obtain the alarm handling method corresponding to the alarm information, and then execute the processing operation corresponding to the alarm handling method. Therefore, this application can promptly analyze and process alarm information corresponding to business system faults monitored by the monitoring system to resolve business system fault problems and ensure the safe and stable operation of the business system.

[0066] The embodiments described above introduced a scheme for determining the alarm handling method corresponding to alarm information based on a pre-constructed fuzzy neural network model. In another embodiment of this application, the fuzzy neural network model can assist in determining the alarm handling method corresponding to alarm information based on preset fuzzy rules.

[0067] The preset fuzzy rules can be shown in Table 1 below:

[0068] Table 1

[0069]

[0070] Wherein, L is the fuzzy term set corresponding to the alarm content, specifically {Level 5 alarm during non-maintenance period, Level 5 alarm during maintenance period, Level 6 alarm during non-maintenance period, Level 6 alarm during maintenance period, Level 7 alarm}; F is the fuzzy term set corresponding to the alarm time, specifically {early morning, morning, noon, evening, night}; {NB, NS, ZO, PS, PB} is the fuzzy term set corresponding to the alarm handling method; where NB indicates notifying maintenance personnel by phone and email, NS indicates notifying maintenance personnel by phone, ZO indicates immediate handling, PS indicates notifying maintenance personnel by email, and PB indicates no handling.

[0071] Specifically, morning and noon can be designated as working hours; early morning, evening, and night can be designated as non-working hours; level 5 alarms can be designated as high-level alarms; level 6 alarms as medium-level alarms; level 7 alarms as low-level alarms; notification to maintenance personnel via telephone and email, or telephone and email notification to maintenance personnel, can be designated as notification processing methods; immediate processing can be designated as immediate processing methods; and no processing can be designated as ignoring processing methods.

[0072] The maintenance period refers to the time during which the business system is updated and upgraded; the non-maintenance period refers to the time outside of the maintenance period.

[0073] In one embodiment, the fuzzy neural network model assists in determining the alarm handling method corresponding to the alarm information based on preset fuzzy rules, and the server 20 may perform the processing operation corresponding to the alarm handling method, which may include, for example: Figure 3 The steps shown are as follows:

[0074] Step 201: Input the alarm content and alarm time into the fuzzy neural network model.

[0075] Step 202: If the alarm content includes a medium-level alarm that is not in the maintenance period, then the real-time processing method output by the fuzzy neural network model is obtained.

[0076] Step 203: Execute the corresponding processing operation for the alarm information in real time according to the preset association relationship.

[0077] In this embodiment, after receiving alarm information, the server 20 can parse the alarm information to obtain the alarm content and alarm time, and input the alarm content and alarm time into the fuzzy neural network model. The fuzzy neural network model analyzes the alarm content and alarm time based on preset fuzzy rules. If it is determined that the alarm content of the alarm information is a medium-level alarm during the non-maintenance period, then regardless of the alarm time, the alarm handling method for the alarm information is determined to be the immediate handling method.

[0078] In this embodiment, the server 20 can store pre-set associations between alarm information and processing operations. After determining the alarm processing operation corresponding to the alarm information, the server 20 can determine the corresponding processing operation based on the aforementioned pre-set associations and execute the determined processing operation. For example, when the alarm processing method is real-time processing, if the fault corresponding to the alarm information is a data error, the server 20 determines the corresponding processing operation to delete the faulty data based on the pre-set associations and executes it; if the fault corresponding to the alarm information is a business subsystem malfunction, the server 20 determines the corresponding processing operation to terminate the operation of the faulty system or device based on the pre-set associations and executes it.

[0079] The method provided in this application embodiment can determine and execute the corresponding immediate processing operation based on a preset association relationship when the fault corresponding to the alarm information is relatively serious. In this method, the server can promptly process relatively serious alarm information, avoiding security risks caused by maintenance personnel being unable to reach the fault site in a timely manner, and ensuring the safe and stable operation of the business system.

[0080] In one embodiment, the fuzzy neural network model assists in determining the alarm handling method corresponding to the alarm information based on preset fuzzy rules, and the server 20 may perform the processing operation corresponding to the alarm handling method, which may include, for example: Figure 4 The steps shown are as follows:

[0081] Step 301: Input the alarm content and alarm time into the fuzzy neural network model.

[0082] Step 302: If the alarm content includes one of the following: high-level alarms during non-maintenance periods, high-level alarms during maintenance periods, and medium-level alarms during maintenance periods, or if the alarm content includes low-level alarms and the alarm time includes non-working periods, then the notification processing method output by the fuzzy neural network model is obtained.

[0083] Step 303: Obtain the target maintenance terminal from the multiple pre-set maintenance terminals.

[0084] Step 304: Send the alarm information to the target maintenance terminal.

[0085] In this embodiment, after receiving alarm information, the server 20 can parse the alarm information to obtain the alarm content and alarm time, and input the alarm content and alarm time into the fuzzy neural network model. The fuzzy neural network model analyzes the alarm content and alarm time based on preset fuzzy rules. If it is determined that the alarm content of the alarm information is one of the following: a high-level alarm during non-maintenance period, a high-level alarm during maintenance period, or a medium-level alarm during maintenance period, then regardless of the alarm time, the alarm handling method for the alarm information is determined to be the notification handling method; or, if the alarm content of the alarm information is a low-level alarm and the alarm time is during a non-working period, then the alarm handling method for the alarm information is determined to be the notification handling method.

[0086] In this embodiment, server 20 can store the correspondence between software or hardware devices and maintenance personnel's maintenance terminals. After determining the faulty software or hardware device based on the alarm information, server 20 can find the maintenance terminal of the maintenance personnel who should be notified from multiple pre-set maintenance terminal terminals according to the above correspondence, i.e., the target maintenance terminal, and then send the alarm information to the target maintenance terminal.

[0087] In this embodiment, when the processing operation is to notify the maintenance personnel by phone, the server can call the maintenance personnel and notify them of the alarm information via voice. When the processing operation is to notify the maintenance personnel by email, the server can send the alarm information to the maintenance personnel's maintenance terminal via email to remind them. When the processing operation is to notify the maintenance personnel by both phone and email, the server can send the alarm information to the maintenance personnel's maintenance terminal via email and call the maintenance personnel, notifying them of the alarm information via both email and voice.

[0088] The method provided in this application embodiment can send alarm information to the target maintenance terminal of maintenance personnel when the fault problem corresponding to the alarm information can be handled later, reminding maintenance personnel that there is a fault in the business system and that it needs to be handled, thus ensuring the safe and stable operation of the business system.

[0089] In one embodiment, the fuzzy neural network model assists in determining the alarm handling method corresponding to the alarm information based on preset fuzzy rules, and the server 20 may perform the processing operation corresponding to the alarm handling method, which may include, for example: Figure 5 The steps shown are as follows:

[0090] Step 401: Input the alarm content and alarm time into the fuzzy neural network model.

[0091] Step 402: If the alarm content includes low-level alarms and the alarm time includes working hours, then the fuzzy neural network model output is ignored.

[0092] Step 403: Ignore the alarm information.

[0093] In this embodiment, after receiving alarm information, the server 20 can parse the alarm information to obtain the alarm content and alarm time, and input the alarm content and alarm time into the fuzzy neural network model. The fuzzy neural network model analyzes the alarm content and alarm time based on preset fuzzy rules. If it is determined that the alarm content of the alarm information is a low-level alarm and the alarm time is during working hours, then the alarm handling method for the alarm information is determined to be the ignore handling method.

[0094] In this embodiment of the application, after determining that the alarm handling method for the alarm information is the ignore handling method, the server 20 can store the alarm information locally, only record the alarm information, and does not need to solve the fault problem corresponding to the alarm information.

[0095] The method provided in this application embodiment can reduce the workload of operation and maintenance personnel by not processing alarm information when the alarm information is a normal system alarm, thus ensuring the stable operation of the business system.

[0096] The embodiments described above introduce a scheme for analyzing and processing alarm information using a fuzzy neural network model to obtain an alarm handling method. In another embodiment of this application, the fuzzy neural network model may include a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer. The fuzzy neural network processes alarm information based on the above four-layer structure, thereby converting the alarm information into a corresponding alarm handling method. Specifically, it includes, for example... Figure 6 The steps shown are as follows:

[0097] Step 501: Input the alarm time and alarm content into the fuzzing layer for fuzzing processing to obtain the fuzzy value corresponding to the alarm time and the fuzzy value corresponding to the alarm content. Then, perform defuzzing processing on the fuzzy value corresponding to the alarm time and the fuzzy value corresponding to the alarm content to obtain the clear value corresponding to the alarm time and the clear value corresponding to the alarm content.

[0098] In this embodiment, the alarm time and alarm content are first input to the fuzzification layer of the fuzzy neural network model. The fuzzification layer performs fuzzification processing on the alarm time and alarm content, converting the alarm time to the value of the corresponding domain of discourse at an appropriate ratio to obtain the sharp value corresponding to the alarm time; the alarm content is converted to the value of the corresponding domain of discourse at an appropriate ratio to obtain the sharp value corresponding to the alarm content, and the sharp values ​​corresponding to the alarm time and alarm content are input to the membership function layer.

[0099] Step 502: Input the clear value corresponding to the alarm time and the clear value corresponding to the alarm content into the membership function layer for the first data processing to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content.

[0100] In this embodiment, the membership function layer calculates the clarity value corresponding to the alarm time and the clarity value corresponding to the alarm content based on multiple preset formulas to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content, and inputs the above multiple membership degrees to the rule layer.

[0101] Among them, the number of membership degrees corresponding to the alarm time is the same as the number of fuzzy subsets in the fuzzy set corresponding to the alarm time; the number of membership degrees corresponding to the alarm content is the same as the number of fuzzy subsets in the fuzzy set corresponding to the alarm content; the number of membership functions in the membership function layer is the sum of the number of fuzzy subsets in the fuzzy set corresponding to the alarm time and the number of fuzzy subsets in the fuzzy set corresponding to the alarm content.

[0102] For example, if the fuzzy set corresponding to the alarm time includes five fuzzy subsets, then the membership function layer calculates the clear value corresponding to the alarm time based on the five membership functions to obtain five membership degrees; if the fuzzy set corresponding to the alarm content includes five fuzzy subsets, then the membership function layer calculates the clear value corresponding to the alarm time based on the five membership functions to obtain five membership degrees; the membership function layer includes a total of 10 membership functions.

[0103] In one embodiment, the membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. The first data processing procedure may then include, for example: Figure 7 The steps shown are as follows:

[0104] Step 601: Determine the membership function corresponding to the alarm time and the membership function corresponding to the alarm content respectively.

[0105] Step 602: Input the clear values ​​corresponding to the alarm time into the corresponding membership functions to obtain multiple membership degrees corresponding to the alarm time.

[0106] Step 603: Input the clear values ​​corresponding to the alarm content into the corresponding membership functions to obtain multiple membership degrees corresponding to the alarm content.

[0107] In this embodiment, the membership function layer first determines the membership function corresponding to the alarm time and the membership function corresponding to the alarm content from multiple membership functions. Then, based on the multiple membership functions corresponding to the alarm time, it calculates the clarity value corresponding to the alarm time to obtain multiple membership degrees corresponding to the alarm time. Based on the multiple membership functions corresponding to the alarm content, it calculates the clarity value corresponding to the alarm content to obtain multiple membership degrees corresponding to the alarm content, and inputs the above multiple membership degrees to the rule layer.

[0108] The membership function can be a Gaussian function as shown in equation (1):

[0109] (1)

[0110] Where i represents the number of input parameters of the fuzzy neural network model (i.e., the two input parameters of alarm time and alarm content); j represents the number of membership functions corresponding to the alarm time or alarm content; x i Let a be the i-th input parameter of the fuzzy neural network; ij b is the center value of the j-th membership function corresponding to the i-th input parameter of the fuzzy neural network; ij f is the width of the j-th membership function corresponding to the i-th input parameter of the fuzzy neural network; ij The membership degree is calculated by using the j-th membership function corresponding to the i-th input parameter of the fuzzy neural network.

[0111] Step 503: Input the multiple membership degrees corresponding to the alarm time and the multiple membership degrees corresponding to the alarm content into the rule layer for the second data processing to obtain the fuzzy value corresponding to the alarm information.

[0112] In this embodiment, the rule layer includes multiple rules, each corresponding to a weight. The number of rules is the product of the number of fuzzy subsets corresponding to the alarm time and the number of fuzzy subsets corresponding to the alarm content, i.e., the rule layer includes 25 rules. The rule layer determines the fuzzy value corresponding to the alarm information based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

[0113] In this embodiment, the rule layer first iterates through the product of the membership degree corresponding to any alarm time and the membership degree corresponding to any alarm content based on the 5 membership degrees corresponding to the alarm time and the 5 membership degrees corresponding to the alarm content, thereby obtaining 25 product values. Then, the 25 product values ​​are weighted and summed according to the following formula (2) to obtain the fuzzy value corresponding to the alarm information.

[0114] (2)

[0115] Where o is the fuzzy value corresponding to the alarm information; cm The weight of the rule; u m It is the product of the membership degree corresponding to any alarm time and the membership degree corresponding to any alarm content.

[0116] Step 504: Input the fuzzy value corresponding to the alarm information into the defuzzification layer for defuzzification processing to obtain the alarm handling method.

[0117] In this embodiment, the fuzzification layer defuzzifies the fuzzy value corresponding to the alarm information, converting the fuzzy value into the corresponding text form to obtain the alarm handling method. Specifically, this may include, for example... Figure 8 The steps shown are as follows:

[0118] Step 701: Determine the range of fuzzy values ​​corresponding to the alarm information.

[0119] Step 702: Determine the alarm handling method corresponding to the alarm information according to the preset mapping relationship.

[0120] In this embodiment, the deblurring layer may include a mapping relationship between a range of fuzzy values ​​and an alarm handling method. The fuzzing layer may first determine the range of fuzzy values ​​corresponding to the alarm information, then determine the alarm handling method corresponding to the alarm information according to the aforementioned preset mapping relationship, and output the alarm handling method corresponding to the alarm information.

[0121] The method provided in this application can analyze alarm information based on a pre-established fuzzy neural network model to obtain the corresponding alarm handling method. Therefore, this application can promptly analyze and process alarm information corresponding to business system faults monitored by the monitoring system, automatically determine the corresponding alarm handling method, resolve business system faults, and ensure the safe and stable operation of the business system.

[0122] In one embodiment, analyzing and processing alarm information based on a fuzzy neural network model requires pre-constructing the fuzzy neural network model. The process of constructing the fuzzy neural network model may include, for example... Figure 9 The steps shown are as follows:

[0123] Step 801: Obtain the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rules.

[0124] In this embodiment of the application, before construction, it is necessary to obtain the various components of the fuzzy neural network model, including the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rules.

[0125] The fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, and the preset fuzzy control rules are pre-set by experts or staff and then input into server 20. The preset membership function is pre-selected by experts or staff, and the undetermined parameters include the weights corresponding to each rule and the undetermined parameters in the membership function. The undetermined parameters in the membership function are determined based on the membership function; for example, if the membership function is a Gaussian function, then the undetermined parameters in the membership function are the center value and width value of the Gaussian function.

[0126] In one embodiment, the parameters to be determined may also be predetermined by experts or staff based on experience.

[0127] Step 802: Construct an initial model based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rules.

[0128] Server 20 constructs, based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rules obtained above, as follows: Figure 10 The initial model shown is a four-layer structure consisting of a fuzzing layer, a membership function layer, a rule layer, and a defuzzing layer. The fuzzing layer has two nodes; the input is the alarm time and alarm content, and the output is the sharpness value corresponding to the alarm time and the sharpness value corresponding to the alarm content. The membership function layer has 10 nodes; the input is the sharpness value corresponding to the alarm time and the sharpness value corresponding to the alarm content, and the output is 5 membership degrees corresponding to the alarm time and 5 membership degrees corresponding to the alarm content. The rule layer has 25 nodes; the input is the 5 membership degrees corresponding to the alarm time and the 5 membership degrees corresponding to the alarm content, and the output is the sum of the products of the 25 membership degrees corresponding to any alarm time and any membership degree corresponding to any alarm content. The defuzzing layer has one node; the input is the sum of the products of the 25 membership degrees corresponding to any alarm time and any membership degree corresponding to any alarm content, and the output is the alarm handling method.

[0129] Step 803: Determine the target value of the parameter to be determined using a preset algorithm.

[0130] In this embodiment of the application, the values ​​of the undetermined parameters in the initial model are unknown. A preset algorithm, such as a neural network algorithm, can be used to train the initial model, and the value of the undetermined parameter when the initial model performs best is determined as the target value of the undetermined parameter.

[0131] In one embodiment, determining the target value of the parameter to be determined using a preset algorithm may include, for example: Figure 11 The steps shown are as follows:

[0132] Step 901: Establish an objective function with the goal of minimizing the output error.

[0133] Step 902: Use the backpropagation algorithm to iteratively calculate the objective function and obtain the output error corresponding to each iteration.

[0134] Step 903: If the output error meets the iteration termination condition, the iteration operation ends, and the value of the undetermined parameter at the end of the iteration operation is determined as the target value of the undetermined parameter.

[0135] In this embodiment, server 20 first establishes the objective function shown in equation (3) with the goal of minimizing the output error:

[0136] E=(dy) 2 / twenty three)

[0137] Where d is the expected output of the initial model; y is the actual output of the initial model; and E is the output error of the initial model.

[0138] Then, the backpropagation (BP) algorithm is used to iteratively calculate the objective function, obtaining the output error for each iteration. When the output error meets the termination condition, the iteration ends, and the value of the undetermined parameter at the end of the iteration is determined as the target value of the undetermined parameter. For example, the iteration ends when the maximum number of iterations is reached, or when the output error is less than the preset error, and the value of the undetermined parameter at the end of the iteration is determined as the target value of the undetermined parameter.

[0139] like Figure 12 As shown, Figure 12 This diagram illustrates the iterative computation process of the initial model. The horizontal axis represents the number of iterations, and the vertical axis represents the output error of the initial model.

[0140] Step 804: Substitute the target values ​​of the undetermined parameters into the initial model to obtain the fuzzy neural network model.

[0141] After determining the target value of the parameters to be determined, server 20 substitutes the target value into the initial model to obtain a complete fuzzy neural network model.

[0142] The method provided in this application embodiment can construct an initial model based on the various components of a fuzzy control model. Then, a preset algorithm is used to iteratively calculate the target values ​​of the undetermined parameters in the initial model. These target values ​​are then substituted into the initial model to obtain a fuzzy neural network model. Optimizing the undetermined parameters in the initial model using a preset algorithm yields target values ​​that make the fuzzy neural network model with the target values ​​more accurate and effective, improving the accuracy of the output alarm processing method and ensuring the safe and stable operation of the business system.

[0143] In one embodiment, the MATLAB toolbox can be used to experiment on the parameter-optimized fuzzy neural network model. The width and center values ​​of the five membership functions corresponding to the alarm time in the parameter-optimized fuzzy neural network model can be obtained: (0.01466-0.07764; 0.2606-0.3184; 0.07561-0.5915; 0.1178-0.7875; 0.04958-1.055). The width and center values ​​of the five membership functions corresponding to the alarm content are: (0.05839-0.001352; 0.0746-0.1536; 0.09875-0.4517; 0.3522-0.7028; 0.02241-1.079). The weights of the 25 rules are: (0.1947; 0.1745; 0.2669; 0.2847; 0.2454; 0.7558; 0.355; 1.002; 0.526; 0.1213; 0.03985; 0.2029; 0.2741; 0.153; -0.387; 0.2307; 0.08005; 0.7366; 0.2197; -0.01286; 0.006163; -0.08543; 0.3815; 0.1582).

[0144] The fuzzy values ​​corresponding to the alarm information are obtained by comparing the standard fuzzy control model and the optimized fuzzy neural network model as shown in Table 2 below:

[0145] Table 2

[0146]

[0147] As shown in Table 2 above, the fuzzy values ​​corresponding to alarm information obtained by the optimized fuzzy neural network are more accurate.

[0148] In practice, the probability of exactly the same situation as in the rule layer occurring is extremely low; any situation can happen. If a situation arises where the rules in the rule layer are not fully met, it will severely impact the ability to handle emergency monitoring. The optimized fuzzy neural network model, however, possesses powerful adaptive self-learning capabilities and is highly practical. Furthermore, the optimized fuzzy neural network model contains only four layers, making it concise and ensuring that the physical meaning of each layer is more clearly defined.

[0149] The optimized fuzzy neural network model in this embodiment does not determine the alarm handling method through individual rules, but rather implicitly distributes the rules throughout the entire neural network, providing a learning function for fuzzy control. In actual application of fuzzy control, there is no need for complex reasoning and rule search; the alarm handling method can be obtained simply through the calculation of the improved fuzzy neural network. It has high-speed parallelism and is superior in processing time compared to normal large-scale rule reasoning and search in actual work.

[0150] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0151] Based on the same inventive concept, this application also provides an alarm device for implementing the alarm method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more alarm device embodiments provided below can be found in the limitations of the alarm method described above, and will not be repeated here.

[0152] In one embodiment, such as Figure 13 As shown, an alarm device is provided, including: an acquisition module 1001, an input module 1002, and an execution module 1003, wherein:

[0153] The acquisition module 1001 is used to acquire alarm information sent by the monitoring system; the alarm information includes the alarm time and alarm content.

[0154] The input module 1002 is used to input alarm information into a pre-built fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model.

[0155] The execution module 1003 is used to perform the processing operations corresponding to the alarm handling method.

[0156] In one embodiment, the input module 1002 is further configured to input the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes a medium-level alarm that is not in maintenance period, the notification processing method output by the fuzzy neural network model is obtained; correspondingly, the execution module 1003 is specifically configured to execute the processing operation corresponding to the alarm information in real time according to the preset association relationship.

[0157] In one embodiment, the input module 1002 is further configured to input the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes one of the following: a high-level alarm during non-maintenance period, a high-level alarm during maintenance period, and a medium-level alarm during maintenance period, or if the alarm content includes a low-level alarm and the alarm time includes non-working periods, then the real-time processing method output by the fuzzy neural network model is obtained; correspondingly, the execution module 1003 is further configured to obtain the target maintenance terminal from a plurality of pre-set maintenance terminals; and send the alarm information to the target maintenance terminal.

[0158] In one embodiment, the input module 1002 is further configured to input the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes low-level alarms and the alarm time includes working hours, then the fuzzy neural network model outputs an ignoring processing method; correspondingly, the execution module 1003 is specifically configured to perform an ignoring operation on the alarm information.

[0159] In one embodiment, the fuzzy neural network model includes a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer. The input module 1002 is further configured to input the alarm time and alarm content into the fuzzification layer for fuzzification processing to obtain fuzzy values ​​corresponding to the alarm time and alarm content, and to defuzzify the fuzzy values ​​corresponding to the alarm time and alarm content to obtain clear values ​​corresponding to the alarm time and alarm content. The clear values ​​corresponding to the alarm time and alarm content are then input into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and alarm content. The multiple membership degrees corresponding to the alarm time and alarm content are then input into the rule layer for second data processing to obtain fuzzy values ​​corresponding to the alarm information. Finally, the fuzzy values ​​corresponding to the alarm information are input into the defuzzification layer for defuzzification processing to obtain the alarm handling method.

[0160] In one embodiment, the membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. The input module 1002 is further used to determine the membership function corresponding to the alarm time and the membership function corresponding to the alarm content respectively; input the clear value corresponding to the alarm time into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm time; input the clear value corresponding to the alarm content into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm content.

[0161] In one embodiment, the rule layer includes multiple rules, and the input module 1002 is further configured to determine the fuzzy value corresponding to the alarm information based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

[0162] In one embodiment, the input module 1002 is further configured to determine the fuzzy value range of the fuzzy value corresponding to the alarm information; and to determine the alarm processing method corresponding to the alarm information according to a preset mapping relationship.

[0163] In one embodiment, the process of constructing a fuzzy neural network model includes: obtaining the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, a preset membership function, undetermined parameters, and a preset fuzzy control rule; constructing an initial model based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rule; determining the target value of the undetermined parameters using a preset algorithm; and substituting the target value of the undetermined parameters into the initial model to obtain the fuzzy neural network model.

[0164] In one embodiment, the construction process of the fuzzy neural network model further includes: establishing an objective function with the goal of minimizing the output error; performing iterative calculations on the objective function using the backpropagation algorithm to obtain the output error corresponding to each iteration; and ending the iteration if the output error meets the termination condition, and determining the value of the undetermined parameter at the end of the iteration as the target value of the undetermined parameter.

[0165] Each module in the aforementioned alarm device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0166] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 14As shown. The computer device includes a processor, memory, and network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores data related to the alarm method described in this application embodiment, such as the alarm information, fuzzy neural network model, and alarm processing methods mentioned above. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements an alarm method.

[0167] Those skilled in the art will understand that Figure 14 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0168] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0169] Obtain alarm information sent by the monitoring system; alarm information includes alarm time and alarm content;

[0170] The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model;

[0171] Perform the corresponding processing operation for the alarm handling method.

[0172] In one embodiment, when the processor executes the computer program, it further implements the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes a medium-level alarm that is not in the maintenance period, obtaining the immediate processing method output by the fuzzy neural network model; and immediately executing the processing operation corresponding to the alarm information according to the preset correlation.

[0173] In one embodiment, when the processor executes the computer program, it further implements the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes one of the following: a high-level alarm during non-maintenance period, a high-level alarm during maintenance period, and a medium-level alarm during maintenance period, or if the alarm content includes a low-level alarm and the alarm time includes non-working periods, then the notification processing method output by the fuzzy neural network model is obtained; the target maintenance terminal is obtained from a plurality of pre-set maintenance terminals; and the alarm information is sent to the target maintenance terminal.

[0174] In one embodiment, when the processor executes the computer program, it further implements the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes low-level alarms and the alarm time includes working hours, then obtaining the ignoring processing method output by the fuzzy neural network model; and performing an ignoring operation on the alarm information.

[0175] In one embodiment, the fuzzy neural network model includes a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer. When the processor executes the computer program, it further implements the following steps: inputting the alarm time and alarm content into the fuzzification layer for fuzzification processing to obtain fuzzy values ​​corresponding to the alarm time and alarm content; defuzzifying the fuzzy values ​​corresponding to the alarm time and alarm content to obtain clear values ​​corresponding to the alarm time and alarm content; inputting the clear values ​​corresponding to the alarm time and alarm content into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and alarm content; inputting the multiple membership degrees corresponding to the alarm time and alarm content into the rule layer for second data processing to obtain fuzzy values ​​corresponding to the alarm information; and inputting the fuzzy values ​​corresponding to the alarm information into the defuzzification layer for defuzzification processing to obtain the alarm handling method.

[0176] In one embodiment, the membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. When the processor executes the computer program, it further implements the following steps: determining the membership function corresponding to the alarm time and the membership function corresponding to the alarm content respectively; inputting the clear value corresponding to the alarm time into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm time; and inputting the clear value corresponding to the alarm content into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm content.

[0177] In one embodiment, the rule layer includes multiple rules, and when the processor executes the computer program, it also implements the following steps: determining the fuzzy value corresponding to the alarm information based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

[0178] In one embodiment, when the processor executes the computer program, it further performs the following steps: determining the fuzzy value range of the fuzzy value corresponding to the alarm information; and determining the alarm handling method corresponding to the alarm information according to a preset mapping relationship.

[0179] In one embodiment, the process of constructing a fuzzy neural network model includes: obtaining the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, a preset membership function, undetermined parameters, and a preset fuzzy control rule; constructing an initial model based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rule; determining the target value of the undetermined parameters using a preset algorithm; and substituting the target value of the undetermined parameters into the initial model to obtain the fuzzy neural network model.

[0180] In one embodiment, the construction process of the fuzzy neural network model further includes: establishing an objective function with the goal of minimizing the output error; performing iterative calculations on the objective function using the backpropagation algorithm to obtain the output error corresponding to each iteration; and ending the iteration if the output error meets the termination condition, and determining the value of the undetermined parameter at the end of the iteration as the target value of the undetermined parameter.

[0181] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0182] Obtain alarm information sent by the monitoring system; alarm information includes alarm time and alarm content;

[0183] The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model;

[0184] Perform the corresponding processing operation for the alarm handling method.

[0185] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes a medium-level alarm that is not in maintenance period, obtaining the immediate processing method output by the fuzzy neural network model; and immediately executing the processing operation corresponding to the alarm information according to the preset correlation.

[0186] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes one of the following: a high-level alarm during non-maintenance period, a high-level alarm during maintenance period, and a medium-level alarm during maintenance period, or if the alarm content includes a low-level alarm and the alarm time includes non-working periods, then obtaining the notification processing method output by the fuzzy neural network model; obtaining the target maintenance terminal from a plurality of pre-set maintenance terminals; and sending the alarm information to the target maintenance terminal.

[0187] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes low-level alarms and the alarm time includes working hours, then obtaining the ignoring processing method output by the fuzzy neural network model; and performing an ignoring operation on the alarm information.

[0188] In one embodiment, the fuzzy neural network model includes a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer. When the computer program is executed by a processor, it further implements the following steps: inputting the alarm time and alarm content into the fuzzification layer for fuzzification processing to obtain fuzzy values ​​corresponding to the alarm time and alarm content, and defuzzifying the fuzzy values ​​corresponding to the alarm time and alarm content to obtain clear values ​​corresponding to the alarm time and alarm content; inputting the clear values ​​corresponding to the alarm time and alarm content into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content; inputting the multiple membership degrees corresponding to the alarm time and alarm content into the rule layer for second data processing to obtain fuzzy values ​​corresponding to the alarm information; and inputting the fuzzy values ​​corresponding to the alarm information into the defuzzification layer for defuzzification processing to obtain the alarm handling method.

[0189] In one embodiment, the membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. When the computer program is executed by the processor, it further implements the following steps: determining the membership function corresponding to the alarm time and the membership function corresponding to the alarm content respectively; inputting the clear value corresponding to the alarm time into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm time; and inputting the clear value corresponding to the alarm content into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm content.

[0190] In one embodiment, the rule layer includes multiple rules, and when the computer program is executed by the processor, it further implements the following steps: determining the fuzzy value corresponding to the alarm information based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

[0191] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the fuzzy value range of the fuzzy value corresponding to the alarm information; and determining the alarm handling method corresponding to the alarm information according to a preset mapping relationship.

[0192] In one embodiment, the process of constructing a fuzzy neural network model includes: obtaining the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, a preset membership function, undetermined parameters, and a preset fuzzy control rule; constructing an initial model based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rule; determining the target value of the undetermined parameters using a preset algorithm; and substituting the target value of the undetermined parameters into the initial model to obtain the fuzzy neural network model.

[0193] In one embodiment, the construction process of the fuzzy neural network model further includes: establishing an objective function with the goal of minimizing the output error; performing iterative calculations on the objective function using the backpropagation algorithm to obtain the output error corresponding to each iteration; and ending the iteration if the output error meets the termination condition, and determining the value of the undetermined parameter at the end of the iteration as the target value of the undetermined parameter.

[0194] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0195] Obtain alarm information sent by the monitoring system; alarm information includes alarm time and alarm content;

[0196] The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm handling method output by the fuzzy neural network model;

[0197] Perform the corresponding processing operation for the alarm handling method.

[0198] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes a medium-level alarm that is not in maintenance period, obtaining the immediate processing method output by the fuzzy neural network model; and immediately executing the processing operation corresponding to the alarm information according to the preset correlation.

[0199] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes one of the following: a high-level alarm during non-maintenance period, a high-level alarm during maintenance period, and a medium-level alarm during maintenance period, or if the alarm content includes a low-level alarm and the alarm time includes non-working periods, then obtaining the notification processing method output by the fuzzy neural network model; obtaining the target maintenance terminal from a plurality of pre-set maintenance terminals; and sending the alarm information to the target maintenance terminal.

[0200] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: inputting the alarm content and alarm time into the fuzzy neural network model; if the alarm content includes low-level alarms and the alarm time includes working hours, then obtaining the ignoring processing method output by the fuzzy neural network model; and performing an ignoring operation on the alarm information.

[0201] In one embodiment, the fuzzy neural network model includes a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer. When the computer program is executed by a processor, it further implements the following steps: inputting the alarm time and alarm content into the fuzzification layer for fuzzification processing to obtain fuzzy values ​​corresponding to the alarm time and alarm content, and defuzzifying the fuzzy values ​​corresponding to the alarm time and alarm content to obtain clear values ​​corresponding to the alarm time and alarm content; inputting the clear values ​​corresponding to the alarm time and alarm content into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content; inputting the multiple membership degrees corresponding to the alarm time and alarm content into the rule layer for second data processing to obtain fuzzy values ​​corresponding to the alarm information; and inputting the fuzzy values ​​corresponding to the alarm information into the defuzzification layer for defuzzification processing to obtain the alarm handling method.

[0202] In one embodiment, the membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. When the computer program is executed by the processor, it further implements the following steps: determining the membership function corresponding to the alarm time and the membership function corresponding to the alarm content respectively; inputting the clear value corresponding to the alarm time into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm time; and inputting the clear value corresponding to the alarm content into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm content.

[0203] In one embodiment, the rule layer includes multiple rules, and when the computer program is executed by the processor, it further implements the following steps: determining the fuzzy value corresponding to the alarm information based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

[0204] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: determining the fuzzy value range of the fuzzy value corresponding to the alarm information; and determining the alarm handling method corresponding to the alarm information according to a preset mapping relationship.

[0205] In one embodiment, the process of constructing a fuzzy neural network model includes: obtaining the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, a preset membership function, undetermined parameters, and a preset fuzzy control rule; constructing an initial model based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rule; determining the target value of the undetermined parameters using a preset algorithm; and substituting the target value of the undetermined parameters into the initial model to obtain the fuzzy neural network model.

[0206] In one embodiment, the construction process of the fuzzy neural network model further includes: establishing an objective function with the goal of minimizing the output error; performing iterative calculations on the objective function using the backpropagation algorithm to obtain the output error corresponding to each iteration; and ending the iteration if the output error meets the termination condition, and determining the value of the undetermined parameter at the end of the iteration as the target value of the undetermined parameter.

[0207] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0208] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0209] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0210] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An alarm method, characterized in that, The method includes: Obtain alarm information sent by the monitoring system; the alarm information includes the alarm time and alarm content; The alarm information is input into a pre-built fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model; Perform the processing operation corresponding to the alarm processing method described above; The fuzzy neural network model includes a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer; the step of inputting the alarm information into the pre-constructed fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model includes: The alarm time and alarm content are input into the blurring layer for blurring processing to obtain the blur value corresponding to the alarm time and the blur value corresponding to the alarm content. Then, the blur value corresponding to the alarm time and the blur value corresponding to the alarm content are deblurred to obtain the clear value corresponding to the alarm time and the clear value corresponding to the alarm content. The clarity value corresponding to the alarm time and the clarity value corresponding to the alarm content are input into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content. The multiple membership degrees corresponding to the alarm time and the multiple membership degrees corresponding to the alarm content are input into the rule layer for second data processing to obtain the fuzzy value corresponding to the alarm information. The fuzzy value corresponding to the alarm information is input into the defuzzification layer for defuzzification processing to obtain the alarm processing method; The membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. The step of inputting the sharp value corresponding to the alarm time and the sharp value corresponding to the alarm content into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content includes: Determine the membership function corresponding to the alarm time and the membership function corresponding to the alarm content respectively; The clarity value corresponding to the alarm time is input into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm time; The clarity values ​​corresponding to the alarm content are input into the corresponding membership functions to obtain multiple membership degrees corresponding to the alarm content; The rule layer includes multiple rules. The second data processing involves inputting multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content into the rule layer to obtain a fuzzy value corresponding to the alarm information, including: The fuzzy value corresponding to the alarm information is determined based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

2. The method according to claim 1, characterized in that, The step of inputting the alarm information into a pre-constructed fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model includes: The alarm content and the alarm time are input into the fuzzy neural network model; If the alarm content includes a medium-level alarm that is not in the maintenance period, then the immediate processing method output by the fuzzy neural network model is obtained; Correspondingly, the processing operation corresponding to the alarm processing method includes: The processing operation corresponding to the alarm information is executed immediately according to the preset association relationship.

3. The method according to claim 1, characterized in that, The step of inputting the alarm information into a pre-constructed fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model includes: The alarm content and the alarm time are input into the fuzzy neural network model; If the alarm content includes one of the following: a high-level alarm during non-maintenance period, a high-level alarm during maintenance period, and a medium-level alarm during maintenance period, or if the alarm content includes a low-level alarm and the alarm time includes non-working periods, then the notification processing method output by the fuzzy neural network model is obtained. Correspondingly, the processing operation corresponding to the alarm processing method includes: Obtain the target maintenance terminal from multiple pre-set maintenance terminals; The alarm information is sent to the target maintenance terminal.

4. The method according to claim 1, characterized in that, The step of inputting the alarm information into a pre-constructed fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model includes: The alarm content and the alarm time are input into the fuzzy neural network model; If the alarm content includes low-level alarms and the alarm time includes working hours, then the ignoring processing method of the output of the fuzzy neural network model is obtained; Correspondingly, the processing operation corresponding to the alarm processing method includes: The alarm information is ignored.

5. The method according to claim 1, characterized in that, The step of inputting the fuzzy value corresponding to the alarm information into the defuzzification layer for defuzzification processing to obtain the alarm processing method includes: Determine the fuzzy value range corresponding to the fuzzy value of the alarm information; The alarm handling method corresponding to the alarm information is determined according to the preset mapping relationship.

6. The method according to claim 1, characterized in that, The construction process of the fuzzy neural network model includes: Obtain the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rules; An initial model is constructed based on the fuzzy set corresponding to the alarm time, the fuzzy set corresponding to the alarm content, the preset membership function, the undetermined parameters, and the preset fuzzy control rules. The target value of the parameter to be determined is determined using a preset algorithm; Substituting the target value of the undetermined parameter into the initial model yields the fuzzy neural network model.

7. The method according to claim 6, characterized in that, The step of determining the target value of the parameter to be determined using a preset algorithm includes: Establish an objective function with the goal of minimizing the output error; The objective function is iteratively calculated using the backpropagation algorithm to obtain the output error corresponding to each iteration. If the output error meets the iteration termination condition, the iteration operation ends, and the value of the undetermined parameter at the end of the iteration operation is determined as the target value of the undetermined parameter.

8. An alarm device, characterized in that, The device includes: The acquisition module is used to acquire alarm information sent by the monitoring system; the alarm information includes alarm time and alarm content. The input module is used to input the alarm information into a pre-built fuzzy neural network model to obtain the alarm processing method output by the fuzzy neural network model; The execution module is used to perform the processing operations corresponding to the alarm processing method; The fuzzy neural network model includes a fuzzification layer, a membership function layer, a rule layer, and a defuzzification layer. The input module is specifically used to input the alarm time and the alarm content into the fuzzification layer for fuzzification processing to obtain the fuzzy value corresponding to the alarm time and the fuzzy value corresponding to the alarm content, and to defuzzify the fuzzy value corresponding to the alarm time and the fuzzy value corresponding to the alarm content to obtain the clear value corresponding to the alarm time and the clear value corresponding to the alarm content. The clarity value corresponding to the alarm time and the clarity value corresponding to the alarm content are input into the membership function layer for first data processing to obtain multiple membership degrees corresponding to the alarm time and multiple membership degrees corresponding to the alarm content. The multiple membership degrees corresponding to the alarm time and the multiple membership degrees corresponding to the alarm content are input into the rule layer for second data processing to obtain the fuzzy value corresponding to the alarm information. The fuzzy value corresponding to the alarm information is input into the defuzzification layer for defuzzification processing to obtain the alarm processing method; The membership function layer includes a fuzzy set corresponding to the alarm time and a fuzzy set corresponding to the alarm content. Each fuzzy set includes multiple fuzzy subsets, and each fuzzy subset corresponds to a membership function. The input module is specifically used to determine the membership function corresponding to the alarm time and the membership function corresponding to the alarm content, respectively. The clarity value corresponding to the alarm time is input into the corresponding membership function to obtain multiple membership degrees corresponding to the alarm time; The clarity values ​​corresponding to the alarm content are input into the corresponding membership functions to obtain multiple membership degrees corresponding to the alarm content; The rule layer includes multiple rules. The input module is specifically used to determine the fuzzy value corresponding to the alarm information based on the multiple membership degrees corresponding to the alarm time, the multiple membership degrees corresponding to the alarm content, and the weights corresponding to each rule.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images