Communication equipment inspection method and device, equipment and medium

Abstract

The embodiment of the invention provides an inspection method and device for communication equipment, equipment and a medium, and the method specifically comprises the steps: executing an inspection task instruction through a large language model and a logic processing model, so as to obtain result data corresponding to the inspection task instruction; generating an inspection result report according to the result data; the large language model reads a current instruction in the inspection task instructions; if the current instruction is the first instruction, obtaining a first instruction result corresponding to the first instruction from a log of a database; if the current instruction is a second instruction, the second instruction is sent to the PLC, so that the PLC sends the second instruction to a logic processing model, the logic processing model executes the second instruction to obtain a corresponding second instruction result, and the second instruction result transferred by the PLC is received; and fusing the first instruction result and the second instruction result to obtain result data corresponding to the inspection task instruction. According to the embodiment of the invention, the accuracy of the inspection result report can be improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of operation and maintenance of communication equipment, and in particular, to a method, apparatus, device and medium for inspecting communication equipment. Background Art

[0002] Regular inspection of communication equipment is an important measure to ensure stable network operation and timely discover potential problems. Regular inspection usually includes checking the physical status, performance indicators and configuration parameters of various communication equipment (such as routers, switches, servers, wireless base stations, etc.). With the rapid development of the communication industry, the scale of computer room construction and communication equipment has continued to expand, which has put forward higher requirements for the inspection of communication equipment.

[0003] In related technologies, robots or automated tools can be used to collect indicator data of communication equipment in real time and upload the indicator data to a database. It is also possible to identify abnormal situations based on the analysis of indicator data in the database and send alerts to relevant personnel.

[0004] However, the relevant technology only provides a means of collecting and recording indicator data, but does not provide corresponding inspection results based on specific inspection needs. Summary of the invention

[0005] The embodiment of the present application provides a communication equipment inspection method, which can improve the accuracy of the inspection result report. The accurate inspection result report can provide reliable data support for operation and maintenance personnel and help quickly locate problems or abnormal conditions in the communication equipment.

[0006] Correspondingly, an embodiment of the present application also provides an inspection device for a communication device, an electronic device and a machine-readable medium to ensure the implementation and application of the above method.

[0007] In order to solve the above problems, an embodiment of the present application discloses a method for inspecting a communication device, the method comprising:

[0008] In order to solve the above problems, an embodiment of the present application discloses a patrol inspection device for communication equipment, the device comprising:

[0009] A receiving module, used for receiving an inspection task instruction; the inspection task instruction includes: a first instruction and a second instruction;

[0010] An instruction execution module, used to execute the inspection task instruction by using a large language model and a logic processing model to obtain result data corresponding to the inspection task instruction;

[0011] A report generation module, used to generate an inspection result report based on the result data;

[0012] Wherein, the instruction execution module includes:

[0013] A reading module, used for reading the current instruction in the inspection task instruction by a large language model;

[0014] A first instruction execution module, configured to obtain, by the large language model, a first instruction result corresponding to the first instruction from a log of a database if the current instruction is a first instruction; the log of the database records a collection result of device information of the communication device;

[0015] A second instruction execution module, configured to, if the current instruction is a second instruction, send the second instruction to a programmable logic controller (PLC) by the large language model, so that the PLC sends the second instruction to the logic processing model, and the logic processing model executes the second instruction to obtain a corresponding second instruction result, and receive the second instruction result transferred by the PLC;

[0016] The fusion module is used to fuse the first instruction result and the second instruction result by using a large language model to obtain result data corresponding to the inspection task instruction.

[0017] Optionally, the first instruction execution module includes:

[0018] An indicator instruction execution module, used for determining the number of indicators according to the task category of the inspection task when the first instruction is an inspection indicator instruction, and obtaining the target inspection indicator corresponding to the number of indicators from the inspection indicator database according to the weight of the inspection indicator; different task categories correspond to different numbers of indicators;

[0019] an object instruction execution module, for determining the target object category of the inspection object according to the task category of the inspection task when the first instruction is an inspection object instruction, and obtaining the target inspection object corresponding to the target object category from the inspection object database according to the object category of the inspection object; different task categories correspond to different target object categories;

[0020] The indicator value instruction execution module is used to obtain the indicator value of the target inspection indicator corresponding to the target inspection object from the indicator value database when the first instruction is an inspection indicator value instruction.

[0021] Optionally, the first instruction execution module further includes:

[0022] An associated object acquisition module, used to determine, in the case where the first instruction is an associated relationship filtering instruction, a first associated target inspection object having an associated relationship among the target inspection objects according to an associated relationship database;

[0023] The associated index value acquisition module is used to acquire the index value of the target inspection index corresponding to the first associated target inspection object from the index value of the target inspection index corresponding to the target inspection object.

[0024] Optionally, the first instruction execution module further includes:

[0025] An extended instruction execution module, used to determine a second associated target inspection object corresponding to the target inspection object according to an associated relationship database when the first instruction is an associated relationship extended instruction;

[0026] The extended indicator value acquisition module is used to obtain the indicator value of the target inspection indicator corresponding to the second associated target inspection object from the indicator value database.

[0027] Optionally, the first instruction result includes: the last first instruction result; the last first instruction result includes: the indicator value of the target inspection indicator corresponding to the target inspection object;

[0028] The fusion module is specifically used to add a second instruction result to the indicator value of the target inspection indicator corresponding to the target inspection object by the large language model.

[0029] Optionally, the second instruction is a logic judgment instruction;

[0030] The logic judgment instruction includes:

[0031] The first logic judgment instruction is used to judge whether the indicator result is abnormal or normal according to the indicator value and indicator threshold of the target inspection indicator corresponding to the target inspection object;

[0032] The second logic judgment instruction judges the logical relationship between the indicator results of any two target inspection indicators;

[0033] The third logic judgment instruction is used to determine the abnormal probability corresponding to the target inspection object when the logical relationship between the indicator results of any two target inspection indicators of the target inspection object is inconsistent.

[0034] Optionally, the logic processing model executes the second instruction to obtain a corresponding second instruction result, including:

[0035] The logic processing model obtains the most recent N indicator values ​​for any two target inspection indicators whose indicator results corresponding to the target inspection objects are inconsistent; N is a positive integer greater than 1;

[0036] According to the index values ​​of the last N times, determine the dynamic thresholds corresponding to any two target inspection indicators;

[0037] Compare the most recent N indicator values ​​with the dynamic threshold and collect statistics on the comparison results;

[0038] According to the statistical results of the comparison results, the abnormal probability corresponding to the target inspection object is determined.

[0039] Optionally, the report generating module includes:

[0040] A table generation module is used to generate a corresponding inspection result table according to the result data; the inspection result table includes: a first field corresponding to the first instruction result, and a second field corresponding to the second instruction result;

[0041] The first field includes: a target inspection object field, a target inspection index field, and an index value field; the second field includes: a logical relationship field between any two target inspection indexes corresponding to the target inspection object, and an abnormal probability field corresponding to the target inspection object.

[0042] Optionally, the report generating module further includes:

[0043] A template acquisition module is used to acquire a report template; the fields in the report template include: a task category field of the inspection task, an object quantity field, a quantity field of target inspection objects of the object category, an indicator quantity field, a normal target inspection indicator quantity field, an abnormal target inspection indicator quantity field, and a distribution field of abnormal target inspection indicators in the target inspection object;

[0044] The filling module is used to fill in the fields in the report template according to the inspection result table to obtain the inspection result text.

[0045] Optionally, the report generating module further includes:

[0046] The prompt text generation module is used to generate corresponding prompt text for any target inspection object whose logical relationship between any two target inspection indicators is inconsistent with the indicator results.

[0047] The embodiment of the present application further discloses an electronic device, including: a processor; and a memory, on which executable code is stored. When the executable code is executed, the processor executes the method described in the embodiment of the present application.

[0048] The embodiment of the present application further discloses a machine-readable medium on which executable code is stored. When the executable code is executed, a processor executes the method described in the embodiment of the present application.

[0049] The embodiments of the present application include the following advantages:

[0050] In the technical solution of the embodiment of the present application, a large language model and a logic processing model are used to execute inspection task instructions to obtain corresponding result data, and an inspection result report is generated based on the above result data.

[0051] Since the inspection task instruction can reflect the inspection requirements, the execution process of the instructions in the inspection task instruction includes: a means of analyzing the collection results of the equipment information according to the inspection requirements. On this basis, the embodiment of the present application can obtain result data and inspection result reports that meet the inspection requirements based on the analysis of the collection results; therefore, the embodiment of the present application can improve the accuracy of the inspection result report. Accurate inspection result reports can provide reliable data support for operation and maintenance personnel and help quickly locate problems or abnormal conditions in communication equipment.

[0052] Furthermore, the embodiment of the present application utilizes a large language model and a logic processing model to automatically execute inspection task instructions, thereby reducing the time and cost of manual operations and improving inspection efficiency. This automated and intelligent processing method makes the inspection process more efficient and accurate.

[0053] In addition, the embodiment of the present application uses PLC to realize the collaborative work of the large language model and the logical processing model. The above collaborative work can give play to the respective advantages of the large language model and the logical processing model. Among them, the large language model is good at processing and understanding natural language, and can accurately obtain log data from the database for the first instruction, and perform text analysis and understanding. The logical processing model can perform precise logical and arithmetic operations for the second instruction. Therefore, the collaborative work of the embodiment of the present application can improve the accuracy of the result data and the inspection result report. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a schematic diagram of the steps of a method for inspecting a communication device according to an embodiment of the present application;

[0055] Figure 2 is a schematic diagram of a report template of an embodiment of the present application;

[0056] Figure 3 is a schematic diagram of an inspection result report of an embodiment of the present application;

[0057] Figure 4 It is a structural schematic diagram of a patrol inspection device for communication equipment according to an embodiment of the present application;

[0058] Figure 5 It is a schematic diagram of the structure of a device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0060] The embodiments of the present application can be applied to the field of operation and maintenance technology of communication equipment, and can be used to inspect various communication equipment (such as routers, switches, servers, wireless base stations, etc.) in a computer room to determine whether there is an abnormality in the communication equipment at present, or to determine whether an abnormality will occur in the communication equipment next.

[0061] In related technologies, robots or automated tools can be used to collect indicator data of communication equipment in real time and upload the indicator data to a database. It is also possible to identify abnormal situations based on the analysis of indicator data in the database and send alerts to relevant personnel.

[0062] However, the relevant technology only provides a means of collecting and recording indicator data, but does not provide corresponding inspection results based on specific inspection needs.

[0063] In view of the technical problem that the related technology does not provide corresponding inspection results according to specific inspection requirements, the embodiment of the present application provides an inspection method for a communication device, which may specifically include the following steps:

[0064] Receive an inspection task instruction; the inspection task instruction includes: a first instruction and a second instruction;

[0065] Utilize the large language model and the logic processing model to execute the above inspection task instructions to obtain result data corresponding to the inspection task instructions;

[0066] Generate an inspection result report based on the above result data;

[0067] The above-mentioned use of the large language model and the logic processing model to execute the above-mentioned inspection task instructions includes:

[0068] The large language model reads the current instruction in the above inspection task instructions;

[0069] If the current instruction is the first instruction, the large language model obtains the first instruction result corresponding to the first instruction from the log of the database; the log of the database records the collection result of the device information of the communication device;

[0070] If the current instruction is the second instruction, the large language model sends the second instruction to a PLC (Programmable Logic Controller), so that the PLC sends the second instruction to the logic processing model, and the logic processing model executes the second instruction to obtain a corresponding second instruction result, and receives the second instruction result transferred by the PLC;

[0071] The large language model fuses the first instruction result and the second instruction result to obtain result data corresponding to the above inspection task instruction.

[0072] The embodiment of the present application utilizes a large language model and a logic processing model to execute inspection task instructions to obtain corresponding result data, and generates an inspection result report based on the above result data.

[0073] Since the inspection task instruction can reflect the inspection requirements, the execution process of the instructions in the inspection task instruction includes: a means of analyzing the collection results of the equipment information according to the inspection requirements. On this basis, the embodiment of the present application can obtain result data and inspection result reports that meet the inspection requirements based on the analysis of the collection results; therefore, the embodiment of the present application can improve the accuracy of the inspection result report. Accurate inspection result reports can provide reliable data support for operation and maintenance personnel and help quickly locate problems or abnormal conditions in communication equipment.

[0074] Furthermore, the embodiment of the present application utilizes a large language model and a logic processing model to automatically execute inspection task instructions, thereby reducing the time and cost of manual operations and improving inspection efficiency. This automated and intelligent processing method makes the inspection process more efficient and accurate.

[0075] In addition, the embodiment of the present application uses PLC to realize the collaborative work of the large language model and the logical processing model. The above collaborative work can give play to the respective advantages of the large language model and the logical processing model. Among them, the large language model is good at processing and understanding natural language, and can accurately obtain log data from the database for the first instruction, and perform text analysis and understanding. The logical processing model can perform precise logical and arithmetic operations for the second instruction. Therefore, the collaborative work of the embodiment of the present application can improve the accuracy of the result data.

[0076] It should be noted that the inspection task instruction of the embodiment of the present application specifically includes multiple instructions such as the first instruction and the second instruction. Since the large language model does not have the execution capability of the second instruction, in the related art, when the large language model encounters the second instruction, it will interrupt the execution of the inspection task instruction; in this way, the large language model will spend multiple operations to complete the execution of the inspection task instruction. The interrupted execution and multiple operations of the above-mentioned inspection task instructions will not only affect the execution efficiency of the inspection task instructions, but also increase the resource consumption of the large language model. The embodiment of the present application utilizes PLC to realize the collaborative work of the large language model and the logic processing model, which enables the large language model to complete the execution of the inspection task instruction based on one operation without interrupting the execution of the inspection task instruction. Therefore, the embodiment of the present application can improve the execution efficiency of the inspection task instruction and reduce the resource consumption of the large language model.

[0077] Method Example 1

[0078] refer to Figure 1 , shows a schematic flow chart of the steps of a method for inspecting a communication device according to an embodiment of the present application, the method may specifically include the following steps:

[0079] Step 101, receiving an inspection task instruction; the inspection task instruction specifically includes: a first instruction and a second instruction;

[0080] Step 102: Utilize the large language model and the logic processing model to execute the inspection task instruction to obtain result data corresponding to the inspection task instruction;

[0081] Step 103: Generate an inspection result report based on the above result data;

[0082] The above step 102 uses a large language model and a logic processing model to execute the above inspection task instructions, specifically including:

[0083] Step 121: The large language model reads the current instruction in the inspection task instruction;

[0084] Step 122: If the current instruction is the first instruction, the large language model obtains the first instruction result corresponding to the first instruction from the log of the database; the log of the database records the collection result of the device information of the communication device;

[0085] Step 123: If the current instruction is the second instruction, the large language model sends the second instruction to the PLC, so that the PLC sends the second instruction to the logic processing model, and the logic processing model executes the second instruction to obtain a corresponding second instruction result, and receives the second instruction result transferred by the PLC;

[0086] Step 124: The large language model fuses the first instruction result and the second instruction result to obtain result data corresponding to the inspection task instruction.

[0087] Figure 1 At least one step included in the method shown can be executed by the server. The server can inspect various communication devices (such as routers, switches, servers, wireless base stations, etc.) in the computer room according to the inspection method of communication devices in the embodiment of the present application to determine whether the communication device is currently abnormal, or determine whether the communication device will be abnormal next.

[0088] In step 101, an inspection task instruction input by a user may be received. In actual applications, a user may use a text editing tool to obtain an inspection task file containing an inspection task instruction. Alternatively, a user may input information contained in the inspection task instruction according to an input box provided by the server. Alternatively, a user may upload a picture corresponding to the inspection task sheet, wherein the picture may contain information about the inspection task instruction. It is understandable that a user may input an inspection task instruction in a variety of ways, and the embodiment of the present application does not limit the specific input method of the inspection task instruction.

[0089] The inspection task instructions may reflect inspection requirements, and the inspection task instructions may include instructions related to the inspection requirements.

[0090] An example of an inspection task instruction is provided here, and the inspection task instruction may specifically include the following information:

[0091] The task category of this inspection task is Category A, which is to inspect the communication equipment in room A.

[0092] Inspection indicator instructions: Which indicators need to be inspected?

[0093] Inspection object instructions: Which communication equipment needs to be inspected?

[0094] Inspection index value instructions: What are the index values?

[0095] The first logical relationship judgment instruction: whether there is a conflict between the indicator values ​​of any two indicators

[0096] The second logic judgment instruction: if there is a conflict between the indicator values ​​of any two indicators, whether there is an abnormality in the corresponding communication device.

[0097] The task category of the inspection task can be determined by those skilled in the art according to actual application requirements. For example, the task category can include: important category, ordinary category, etc. Category A can represent any task category. Computer room A can represent any computer room.

[0098] In step 102, a large language model is an artificial intelligence model designed to understand and generate human language. The large language model can perform a wide range of tasks, including the first instruction execution task of the embodiment of the present application, etc. by training on a large amount of training data. The large language model is usually based on a deep learning architecture, such as a transformer (Transformer) structure. The embodiment of the present application does not limit the specific type of the large language model. For example, the type of the large language model may include: GPT (Generative Pre-Trained Transformer) model and GLM (Generative Language Model, General Language Model), etc. Among them, the GPT model is a large-scale natural language generation model based on the Transformer structure, which is trained using a large amount of training data (the training data may be text data) and can generate high-quality natural language texts, including articles, dialogues, summaries, etc. In the embodiment of the present application, the first instruction result and result data obtained by the large language model are both lower concepts of high-quality natural language texts.

[0099] The logic processing model is a computational model used to process and analyze logic problems. It usually involves the logical reasoning, judgment, and decision-making process of input data. In the inspection method of communication equipment, the role of the logic processing model is to execute specific logic judgment instructions, such as judging whether there is a conflict between indicator values, and judging whether there is an abnormality in the equipment based on these conflicts.

[0100] The logical processing model has the following characteristics:

[0101] Logical reasoning ability: Ability to reason and draw conclusions based on given logical rules and input data.

[0102] Decision support: Ability to make decisions or provide decision support based on reasoning results.

[0103] Programmability: Different logical rules and processing flows can be implemented through programming.

[0104] Efficiency: Ability to process large amounts of data quickly and produce results.

[0105] In the embodiment of the present application, the logic processing model may be located on the inspection robot side. The inspection robot uses an automated robot system to collect and monitor device information such as the operating status, performance indicators and other relevant data of the communication device.

[0106] Optionally, the inspection robot may be equipped with a network tester, such as a network performance analyzer (Network Performance Analyzer), for measuring performance information such as delay, packet loss rate, and bandwidth utilization of communication equipment.

[0107] Optionally, the inspection robot may use an optical power meter or an optical time domain reflectometer to collect the optical attenuation of the communication device.

[0108] The inspection robot can record the collection results in the log of the database. In addition to the fields corresponding to the collection results, the log can also include information such as the collection time, the communication device identification, and the identification of the room where the communication device is located.

[0109] The embodiment of the present application uses PLC to realize the collaborative work of the large language model and the logic processing model. PLC can be a software program running on the server, which has the ability to interact with both the large language model and the logic processing model.

[0110] Specifically, the PLC software can receive the second instruction and the first instruction result from the large language model, and forward the second instruction and the first instruction result to the logic processing model. After receiving the second instruction and the first instruction result, the logic processing model will execute the second instruction according to the first instruction result, and send the second instruction result to the PLC. Therefore, the PLC can also send the second instruction result to the large language model.

[0111] In step 121, the large language model can read the current instruction in the inspection task instruction in a front-to-back order. The front-to-back order can represent the order in which the instructions are executed. Assuming that the inspection task instruction is recorded in the inspection task file, the inspection task file can record the instructions that are executed in a front-to-back order.

[0112] In step 122, if the current instruction is the first instruction, the large language model can obtain the first instruction result corresponding to the first instruction from the log of the database; the log of the database records the collection result of the device information of the communication device. Among them, the access address corresponding to the log of the database can be provided to the large language model. The access address can be carried in the inspection task instruction, and can also be sent to the large language model in advance.

[0113] In a specific implementation, the process of the large language model obtaining the first instruction result corresponding to the first instruction from the log of the database specifically includes:

[0114] Step A1: when the first instruction is an inspection indicator instruction, determine the number of indicators according to the task category of the inspection task, and obtain the target inspection indicator corresponding to the number of indicators from the inspection indicator database according to the weight of the inspection indicator; different task categories correspond to different numbers of indicators;

[0115] Step A2: when the first instruction is an inspection object instruction, determine the target object category of the inspection object according to the task category of the inspection task, and obtain the target inspection object corresponding to the target object category from the inspection object database according to the object category of the inspection object; different task categories correspond to different target object categories;

[0116] Step A3: When the first instruction is an inspection indicator value instruction, obtain the indicator value of the target inspection indicator corresponding to the target inspection object from the indicator value database.

[0117] In step A1, if the task category is an important category, the indicator quantity may be a first indicator quantity. If the task category is a common category, the indicator quantity may be a second indicator quantity. Usually, the first indicator quantity is greater than the second indicator quantity. For example, the first indicator quantity is 10 and the second indicator quantity is 5.

[0118] The weight of the inspection index may also be stored as a field of the database. For example, the index weight database may record the name of the inspection index and the weight of the inspection index.

[0119] The weight of the inspection indicator may indicate the importance of the inspection indicator. The weight of the inspection indicator may be presented as a weight value or as a weight category. For example, the weight category of the inspection indicator may include: common indicators and important indicators. Examples of common indicators may include: packet loss, delay, CPU utilization, etc. Important indicators may include: optical attenuation, etc. The weight value of an important indicator may be greater than the weight value of a common indicator.

[0120] The large language model can obtain the target inspection indicators corresponding to the number of indicators from the inspection indicator database according to the order of the weight values ​​of the inspection indicators from large to small. For example, the inspection indicators can be sorted according to the order of the weight values ​​from large to small, and the target inspection indicators ranked in front can be selected from the sorting results.

[0121] In step A2, the object categories of the inspection object may include, from low to high, ordinary categories, important categories, and core categories. Different task categories may correspond to different target object categories. For example, if the task category is an important category, the target object category may include: core category, important category, and ordinary category. If the task category is an ordinary category, the target object category may include: core category and important category.

[0122] In step A3, when the first instruction is an inspection indicator value instruction, the indicator value of the target inspection indicator corresponding to the target inspection object can be obtained from the indicator value database.

[0123] The index value of the target inspection index corresponding to the target inspection object can be used as the first instruction result obtained by the large language model. Referring to Table 1, an example of the index value of the target inspection index corresponding to a target inspection object of the present application is shown. Table 1 may include fields such as target inspection object name, target inspection index name and index value. Of course, in addition to the fields corresponding to the target inspection object name, target inspection index name and index value, it may also include fields such as inspection task ID (identification, Identity), task category, etc.

[0124] Table 1

[0125]

[0126] In an optional embodiment of the present application, the process of the large language model obtaining the first instruction result corresponding to the above-mentioned first instruction from the log of the database may also include: when the above-mentioned first instruction is an association relationship filtering instruction, determining the first associated target inspection object with an association relationship in the target inspection object according to the association relationship database; obtaining the indicator value of the target inspection indicator corresponding to the first associated target inspection object from the indicator value of the target inspection indicator corresponding to the target inspection object.

[0127] For example, the target inspection object is: 1000 devices in computer room A. Assuming that the 1000 devices are: device A1, device A2, device A3... device A1000, etc., then the association relationship filtering instruction is used to obtain the first associated target inspection object with an association relationship from the 1000 devices. The first associated target inspection object can correspond to multiple associated object sets. Among them, each associated object set corresponds to multiple target inspection objects. For example, one associated object set includes: device A1, device A3 and device A5, etc. Another associated object set includes: device A10, device A13 and device A14, etc.

[0128] The large language model can extract the index value of the target inspection index corresponding to the first associated target inspection object from the index value of the target inspection index corresponding to the target inspection object. The index value of the target inspection index corresponding to the first associated target inspection object can be a subset of the index value of the target inspection index corresponding to the target inspection object.

[0129] In another optional embodiment of the present application, the process of the large language model obtaining the first instruction result corresponding to the above-mentioned first instruction from the log of the database may also include: when the above-mentioned first instruction is an association relationship extension instruction, determining the second associated target inspection object corresponding to the target inspection object according to the association relationship database; obtaining the indicator value of the target inspection indicator corresponding to the second associated target inspection object from the indicator value database.

[0130] Assuming that the target inspection object is: 1000 devices in computer room A, the second associated target inspection object may be: devices in other computer rooms except computer room A that have an associated relationship with the devices in computer room A. The associated relationship extension instruction can realize the extension of the target inspection object.

[0131] After determining the second associated target inspection object, the large language model may also obtain the index value of the target inspection index corresponding to the second associated target inspection object from the index value database.

[0132] In a specific implementation, the second instruction may be a logic judgment instruction;

[0133] The above logic judgment instructions specifically include:

[0134] The first logic judgment instruction is used to judge whether the indicator result is abnormal or normal according to the indicator value and indicator threshold of the target inspection indicator corresponding to the target inspection object;

[0135] The second logic judgment instruction judges the logical relationship between the indicator results of any two target inspection indicators;

[0136] The third logic judgment instruction is used to determine the abnormal probability corresponding to the target inspection object when the logical relationship between the indicator results of any two target inspection indicators of the target inspection object is inconsistent.

[0137] For example, the indicator threshold corresponding to packet loss specifically includes: a packet loss number threshold or a packet loss rate threshold, etc. If the indicator value of packet loss is greater than the packet loss number threshold or the packet loss rate threshold, it can be considered that the indicator result corresponding to packet loss is abnormal.

[0138] The logical relationship between any two target inspection indicators may be: a conflict relationship between indicator results of any two target inspection indicators.

[0139] In practical applications, the indicator values ​​of any two target inspection indicators can be compared with the corresponding indicator ranges, and the indicator result can be determined based on the comparison results. For example, if the indicator value is within the indicator range, the indicator result is: the indicator is normal. For another example, if the indicator value is not within the indicator range, the indicator result is abnormal.

[0140] For any two target inspection indicators, if the indicator results of the two are inconsistent. For example, if any two target inspection indicators are inspection indicator 1 and inspection indicator 2, if the indicator result of inspection indicator 1 is normal, and the indicator result of inspection indicator 2 is abnormal, then the indicator results of the two are inconsistent. Or, if the indicator result of inspection indicator 1 is abnormal, and the indicator result of inspection indicator 2 is normal, then the indicator results of the two are inconsistent.

[0141] The second logic judgment instruction is used to determine the abnormal probability corresponding to the target inspection object when the logical relationship between any two target inspection indicators of the target inspection object is that the indicator results are inconsistent.

[0142] Accordingly, the process in which the logic processing model executes the second instruction to obtain the corresponding second instruction result specifically includes:

[0143] Step B1, the logic processing model obtains the most recent N indicator values ​​for any two target inspection indicators whose indicator results corresponding to the target inspection objects are inconsistent; N is a positive integer greater than 1;

[0144] Step B2: Determine the dynamic thresholds corresponding to any two target inspection indicators according to the most recent N indicator values;

[0145] Step B3: compare the most recent N index values ​​with the dynamic threshold, and collect statistics on the comparison results;

[0146] Step B4: Determine the abnormal probability corresponding to the target inspection object according to the statistical result of the comparison result.

[0147] The value of N can be determined by those skilled in the art according to actual application requirements. For example, N is 10.

[0148] The indicator values ​​of the most recent N times may be the indicator values ​​included in the most recent N collection results. Examples of any two target inspection indicators include: packet loss and optical attenuation; both packet loss and optical attenuation are important indicators of the device status of communication equipment. Packet loss refers to the situation where a data packet fails to reach the destination during data transmission. Packet loss rate is an important indicator for measuring network performance. A high packet loss rate can lead to problems such as a decline in user experience, incomplete data transmission, and decline in network performance. Optical attenuation refers to the weakening of signal strength due to absorption, scattering, and other reasons when an optical signal is transmitted in an optical fiber. The optical attenuation value is an important indicator for evaluating the performance of an optical fiber network. A high optical attenuation value can lead to problems such as a decline in signal quality.

[0149] The embodiment of the present application determines the process of dynamic thresholds corresponding to any two target inspection indicators based on the indicator values ​​of the most recent N times, specifically including: using a simple moving average method or a weighted moving average method to determine the average value of the indicator values ​​of the most recent N times in the corresponding time window as the dynamic threshold.

[0150] Both simple moving average and weighted moving average are used to calculate the average value of time series data within a certain time window. However, in the calculation process of simple moving average, each data point (that is, the indicator value at each time) has a different weight. Weighted moving average gives higher weight to the most recent data point, while the weight of earlier data points gradually decreases.

[0151] In step B3, taking packet loss as an example, the most recent N indicator values ​​may be compared with the dynamic threshold, and the comparison results may include: greater than or not greater than, etc. The comparison results may be counted by recording the number of occurrences of each comparison result.

[0152] In step B4, the abnormal probability corresponding to the target inspection object can be determined according to the ratio of the number of occurrences of the target comparison results corresponding to any two target inspection indicators to N.

[0153] Among them, the target comparison result can be a comparison result corresponding to the impact on network quality or abnormality. Taking packet loss as an example, the target comparison result can be: greater than. The ratio of the number of occurrences of "greater than" corresponding to packet loss to N can be a first ratio. Similarly, the ratio of the number of occurrences of "greater than" corresponding to optical attenuation to N can be a second ratio. By taking a weighted average of the first ratio and the second ratio, the abnormal probability corresponding to the target inspection object can be obtained. For example, the ratio of the number of occurrences of "greater than" corresponding to packet loss, 6, to N(10) is 0.6, and the ratio of the number of occurrences of "greater than" corresponding to optical attenuation, 7, to N(10) is 0.7. Assuming that the weight of packet loss is w1 and the weight of optical attenuation is w2, the abnormal probability corresponding to the target inspection object can be: w1*0.6+w2*0.7.

[0154] The abnormal probability corresponding to the target inspection object may include: the probability that the target inspection object is currently abnormal, or the probability that the target inspection object is abnormal in the future, etc. The embodiment of the present application does not limit the specific meaning of the abnormal probability corresponding to the target inspection object.

[0155] It can be understood that the process of determining the abnormal probability corresponding to the target inspection object from step B1 to step B2 is only an example. In fact, those skilled in the art can adopt other processes for determining the abnormal probability corresponding to the target inspection object according to actual application requirements. For example, a machine learning classification method can be used to determine the abnormal probability corresponding to the target inspection object.

[0156] It should be noted that the logic processing model provides a corresponding abnormal probability for a target inspection object whose indicator results of any two target inspection indicators are consistent, and the abnormal probability can be 0.

[0157] In a specific implementation, the first instruction result includes: the last first instruction result; the last first instruction result may be the first instruction result corresponding to the last first instruction. Assume that the last first instruction result specifically includes: the indicator value of the target inspection indicator corresponding to the target inspection object;

[0158] The process of the large language model fusing the first instruction result and the second instruction result specifically includes: the large language model adds the second instruction result to the indicator value of the target inspection indicator corresponding to the target inspection object. After adding the second instruction result to the indicator value of the target inspection indicator corresponding to the target inspection object, the corresponding fusion result can be used as the result data.

[0159] Referring to Table 2, an example of result data of the present application is shown. In addition to the fields such as the target inspection object name, target inspection indicator name and indicator value of Table 1, Table 2 may also include: an indicator result of at least one target inspection indicator name, and an abnormal probability corresponding to the target inspection object, which may be provided by a logical processing model.

[0160] Table 2

[0161]

[0162] In step 103, the embodiment of the present application generates an inspection result report based on the result data. Since the result data can meet the inspection requirements, the inspection result report generated based on the result data can also meet the inspection requirements, and the inspection result report can provide data support for decision-making and help maintenance personnel to discover and solve problems in a timely manner.

[0163] In one implementation, the process of generating the inspection result report according to the result data may specifically include:

[0164] Generate a corresponding inspection result table according to the result data; the inspection result table includes: a first field corresponding to the first instruction result, and a second field corresponding to the second instruction result;

[0165] Among them, the first field includes: target inspection object field, target inspection indicator field, indicator value field; the second field includes: logical relationship field between any two target inspection indicators corresponding to the target inspection object, and abnormal probability field corresponding to the target inspection object.

[0166] Referring to Table 3, an example of an inspection result table of the present application is shown. In addition to the target inspection object name, target inspection indicator name and indicator value, and abnormal probability corresponding to the target inspection object in Table 2, Table 3 may also include: a logical relationship field between any two target inspection indicators. Any two target inspection indicators specifically include: two target inspection indicators that the user is concerned about, such as packet loss and optical attenuation.

[0167] Table 3

[0168]

[0169] It should be noted that, in addition to generating the inspection result table, the embodiment of the present application can also generate an inspection result report in a chart or graphic format. The inspection result report in a graphic format specifically includes: a line chart, etc. It can be understood that the embodiment of the present application does not limit the specific format of the inspection result report.

[0170] It should be noted that before generating the inspection result report based on the result data, the two-step spelling correction model (DE-CO, Detector-Corrector) can also be used to perform spelling correction on the result data. In this way, the inspection result report can be generated based on the spelling-corrected result data.

[0171] The two-step spelling correction model specifically includes: a misspelled word recognizer and a misspelled word corrector. The function of the misspelled word recognizer is to recognize misspelled words in the adversarial sample (original text) and provide these misspelled words to the misspelled word corrector for correction, while ensuring that the correctly spelled words in the adversarial sample are not modified. The function of the misspelled word corrector is to correct the misspelled words based on the spelling information and context information of the misspelled words.

[0172] In another optional embodiment, the above process of generating an inspection result report based on the result data may also include: obtaining a report template; the fields in the report template specifically include: a task category field of the inspection task, an object quantity field, a target inspection object quantity field of the object category, an indicator quantity field, a normal target inspection indicator quantity field, an abnormal target inspection indicator quantity field, and a distribution field of abnormal target inspection indicators in the target inspection object; according to the inspection result table, the fields in the report template are filled in to obtain the inspection result text.

[0173] Reference Figure 2 , shows an example of a report template of an embodiment of the present application. The report template specifically includes:

[0174] The text corresponding to the task category field of the inspection task is 201: "The task category of this inspection is ，” , "" indicates the content to be filled.

[0175] The text 202 corresponding to the object quantity field is: "A total of devices were scanned."

[0176] The text 203 corresponding to the number field of the target inspection object of the object category is: "core equipment station, important equipment station, ordinary equipment station."

[0177] The text 204 corresponding to the indicator quantity field is: "Total number of inspection indicators."

[0178] The text 205 corresponding to the normal target inspection index quantity field is: "Normal inspection index total items."

[0179] The text 206 corresponding to the abnormal target inspection index quantity field is: "abnormal inspection index total number of items."

[0180] The text 207 corresponding to the distribution field of the abnormal target inspection index in the target inspection object is: "The abnormal inspection index is distributed in the following devices:".

[0181] The embodiment of the present application can fill in the contents to be filled in the fields in the report template according to the inspection result table to obtain the inspection result text.

[0182] In another optional implementation, the process of generating the inspection result report based on the result data may also include: generating corresponding prompt text for any target inspection object whose logical relationship between any two target inspection indicators is inconsistent with the indicator results. The prompt text may prompt the user to judge whether there is an abnormality in the target inspection object with inconsistent indicator results based on experience, thereby improving the accuracy of abnormality judgment. Examples of prompt text may include: "The indicator results of packet loss and optical attenuation of device A are inconsistent, please check" and the like.

[0183] The inspection result report of the embodiment of the present application may include: an inspection result table, an inspection result text and a prompt text. Figure 3 , shows a schematic diagram of an inspection result report of an embodiment of the present application, which may specifically include: inspection time 301, task category 302, inspection result table 303, inspection result text 3004 and prompt text 305.

[0184] It should be noted that the embodiment of the present application can use a large language model and a logic processing model to process the inspection task instructions in real time. Therefore, the difference between the output time of the result data and the generation time of the inspection result report can be ignored. Therefore, the inspection time can be the generation time of the inspection result report.

[0185] The embodiment of the present application can display the inspection result report. Alternatively, the embodiment of the present application can send the inspection result report to a preset user using a preset medium such as an email. For example, the preset user can be an operation and maintenance personnel or a decision-maker.

[0186] In summary, the inspection method of the communication equipment in the embodiment of the present application uses a large language model and a logic processing model to execute the inspection task instructions to obtain the corresponding result data, and generates an inspection result report based on the above result data.

[0187] Since the inspection task instruction can reflect the inspection requirements, the execution process of the instructions in the inspection task instruction includes: a means of analyzing the collection results of the equipment information according to the inspection requirements. On this basis, the embodiment of the present application can obtain result data and inspection result reports that meet the inspection requirements based on the analysis of the collection results; therefore, the embodiment of the present application can improve the accuracy of the inspection result report. Accurate inspection result reports can provide reliable data support for operation and maintenance personnel and help quickly locate problems or abnormal conditions in communication equipment.

[0188] Furthermore, the embodiment of the present application utilizes a large language model and a logic processing model to automatically execute inspection task instructions, thereby reducing the time and cost of manual operations and improving inspection efficiency. This automated and intelligent processing method makes the inspection process more efficient and accurate.

[0189] In addition, the embodiment of the present application uses PLC to realize the collaborative work of the large language model and the logical processing model. The above collaborative work can give play to the respective advantages of the large language model and the logical processing model. Among them, the large language model is good at processing and understanding natural language, and can accurately obtain log data from the database for the first instruction, and perform text analysis and understanding. The logical processing model can perform precise logical and arithmetic operations for the second instruction. Therefore, the collaborative work of the embodiment of the present application can improve the accuracy of the result data and the inspection result report.

[0190] It should be noted that, for the method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

[0191] Based on the above embodiment, this embodiment further provides a communication equipment inspection device, referring to Figure 4 The inspection device of the communication equipment may specifically include: a receiving module 401, an instruction execution module 402 and a report generation module 403.

[0192] The receiving module 401 is used to receive an inspection task instruction; the inspection task instruction includes: a first instruction and a second instruction;

[0193] The instruction execution module 402 is used to execute the inspection task instruction by using the large language model and the logic processing model to obtain result data corresponding to the inspection task instruction;

[0194] A report generation module 403 is used to generate an inspection result report based on the result data;

[0195] The instruction execution module 402 specifically includes:

[0196] A reading module 421 is used to read the current instruction in the inspection task instruction by using a large language model;

[0197] The first instruction execution module 422 is used for obtaining the first instruction result corresponding to the first instruction from the log of the database by the large language model if the current instruction is the first instruction; the log of the database records the collection result of the device information of the communication device;

[0198] A second instruction execution module 423 is used for, if the current instruction is a second instruction, sending the second instruction to the programmable logic controller PLC by the large language model, so that the PLC sends the second instruction to the logic processing model, and the logic processing model executes the second instruction to obtain a corresponding second instruction result, and receives the second instruction result forwarded by the PLC;

[0199] The fusion module 424 is used to fuse the first instruction result and the second instruction result by using a large language model to obtain result data corresponding to the inspection task instruction.

[0200] Optionally, the first instruction execution module 422 specifically includes:

[0201] An indicator instruction execution module, used for determining the number of indicators according to the task category of the inspection task when the first instruction is an inspection indicator instruction, and obtaining the target inspection indicator corresponding to the number of indicators from the inspection indicator database according to the weight of the inspection indicator; different task categories correspond to different numbers of indicators;

[0202] an object instruction execution module, for determining the target object category of the inspection object according to the task category of the inspection task when the first instruction is an inspection object instruction, and obtaining the target inspection object corresponding to the target object category from the inspection object database according to the object category of the inspection object; different task categories correspond to different target object categories;

[0203] The indicator value instruction execution module is used to obtain the indicator value of the target inspection indicator corresponding to the target inspection object from the indicator value database when the first instruction is an inspection indicator value instruction.

[0204] Optionally, the first instruction execution module 422 may further include:

[0205] An associated object acquisition module, used to determine, in the case where the first instruction is an associated relationship filtering instruction, a first associated target inspection object having an associated relationship among the target inspection objects according to an associated relationship database;

[0206] The associated index value acquisition module is used to acquire the index value of the target inspection index corresponding to the first associated target inspection object from the index value of the target inspection index corresponding to the target inspection object.

[0207] Optionally, the first instruction execution module 422 may further include:

[0208] An extended instruction execution module, used to determine a second associated target inspection object corresponding to the target inspection object according to an associated relationship database when the first instruction is an associated relationship extended instruction;

[0209] The extended indicator value acquisition module is used to obtain the indicator value of the target inspection indicator corresponding to the second associated target inspection object from the indicator value database.

[0210] Optionally, the first instruction result includes: the last first instruction result; the last first instruction result includes: the indicator value of the target inspection indicator corresponding to the target inspection object;

[0211] The fusion module is specifically used to add a second instruction result to the indicator value of the target inspection indicator corresponding to the target inspection object by the large language model.

[0212] Optionally, the second instruction is a logic judgment instruction;

[0213] The logic judgment instruction includes:

[0214] The first logic judgment instruction is used to judge whether the indicator result is abnormal or normal according to the indicator value and indicator threshold of the target inspection indicator corresponding to the target inspection object;

[0215] The second logic judgment instruction judges the logical relationship between the indicator results of any two target inspection indicators;

[0216] The third logic judgment instruction is used to determine the abnormal probability corresponding to the target inspection object when the logical relationship between the indicator results of any two target inspection indicators of the target inspection object is inconsistent.

[0217] Optionally, the logic processing model executes the second instruction to obtain a corresponding second instruction result, including:

[0218] The logic processing model obtains the most recent N indicator values ​​for any two target inspection indicators whose indicator results corresponding to the target inspection objects are inconsistent; N is a positive integer greater than 1;

[0219] According to the index values ​​of the last N times, determine the dynamic thresholds corresponding to any two target inspection indicators;

[0220] Compare the most recent N indicator values ​​with the dynamic threshold and collect statistics on the comparison results;

[0221] According to the statistical results of the comparison results, the abnormal probability corresponding to the target inspection object is determined.

[0222] Optionally, the report generating module specifically includes:

[0223] A table generation module is used to generate a corresponding inspection result table according to the result data; the inspection result table includes: a first field corresponding to the first instruction result, and a second field corresponding to the second instruction result;

[0224] The first field includes: a target inspection object field, a target inspection index field, and an index value field; the second field includes: a logical relationship field between any two target inspection indexes corresponding to the target inspection object, and an abnormal probability field corresponding to the target inspection object.

[0225] Optionally, the report generating module may further include:

[0226] A template acquisition module is used to acquire a report template; the fields in the report template include: a task category field of the inspection task, an object quantity field, a quantity field of target inspection objects of the object category, an indicator quantity field, a normal target inspection indicator quantity field, an abnormal target inspection indicator quantity field, and a distribution field of abnormal target inspection indicators in the target inspection object;

[0227] The filling module is used to fill in the fields in the report template according to the inspection result table to obtain the inspection result text.

[0228] Optionally, the report generating module may further include:

[0229] The prompt text generation module is used to generate corresponding prompt text for any target inspection object whose logical relationship between any two target inspection indicators is inconsistent with the indicator results.

[0230] In summary, the inspection device of the communication equipment in the embodiment of the present application uses a large language model and a logic processing model to execute inspection task instructions to obtain corresponding result data, and generates an inspection result report based on the above result data.

[0231] Since the inspection task instruction can reflect the inspection requirements, the execution process of the instructions in the inspection task instruction includes: a means of analyzing the collection results of the equipment information according to the inspection requirements. On this basis, the embodiment of the present application can obtain result data and inspection result reports that meet the inspection requirements based on the analysis of the collection results; therefore, the embodiment of the present application can improve the accuracy of the inspection result report. Accurate inspection result reports can provide reliable data support for operation and maintenance personnel and help quickly locate problems or abnormal conditions in communication equipment.

[0232] Furthermore, the embodiment of the present application utilizes a large language model and a logic processing model to automatically execute inspection task instructions, thereby reducing the time and cost of manual operations and improving inspection efficiency. This automated and intelligent processing method makes the inspection process more efficient and accurate.

[0233] In addition, the embodiment of the present application uses PLC to realize the collaborative work of the large language model and the logical processing model. The above collaborative work can give play to the respective advantages of the large language model and the logical processing model. Among them, the large language model is good at processing and understanding natural language, and can accurately obtain log data from the database for the first instruction, and perform text analysis and understanding. The logical processing model can perform precise logical and arithmetic operations for the second instruction. Therefore, the collaborative work of the embodiment of the present application can improve the accuracy of the result data and the inspection result report.

[0234] The embodiment of the present application also provides a non-volatile readable storage medium, which stores one or more modules (programs). When the one or more modules are applied to a device, the device can execute instructions (instructions) of each method step in the embodiment of the present application.

[0235] The present application embodiment provides one or more machine-readable media on which instructions are stored, and when executed by one or more processors, an electronic device executes one or more of the methods described in the above embodiments. In the present application embodiment, the electronic device includes various types of devices such as terminal devices and server terminals (clusters).

[0236] The embodiments of the present disclosure may be implemented as a device that uses any appropriate hardware, firmware, software, or any combination thereof to perform desired configurations. The device may include: electronic devices such as terminal devices and server (cluster). Figure 5 An exemplary apparatus 2100 that can be used to implement various embodiments described in this application is schematically shown.

[0237] For one embodiment, Figure 5An exemplary apparatus 2100 is shown having one or more processors 2102, a control module (chip set) 2104 coupled to at least one of the (one or more) processors 2102, a memory 2106 coupled to the control module 2104, a non-volatile memory / storage device 2108 coupled to the control module 2104, one or more input / output devices 2110 coupled to the control module 2104, and a network interface 2112 coupled to the control module 2104.

[0238] The processor 2102 may include one or more single-core or multi-core processors, and the processor 2102 may include any combination of general-purpose processors or special-purpose processors (such as graphics processors, application processors, baseband processors, etc.). In some embodiments, the device 2100 can be used as a terminal device, a server (cluster), and other devices described in the embodiments of the present application.

[0239] In some embodiments, the apparatus 2100 may include one or more computer-readable media (e.g., memory 2106 or non-volatile memory / storage 2108) having instructions 2114 and one or more processors 2102 combined with the one or more computer-readable media and configured to execute the instructions 2114 to implement a module to perform the actions described in the present disclosure.

[0240] For one embodiment, the control module 2104 may include any suitable interface controller to provide any suitable interface to at least one of the processor(s) 2102 and / or any suitable device or component in communication with the control module 2104 .

[0241] The control module 2104 may include a memory controller module to provide an interface to the memory 2106. The memory controller module may be a hardware module, a software module, and / or a firmware module.

[0242] The memory 2106 may be used, for example, to load and store data and / or instructions 2114 for the device 2100. For one embodiment, the memory 2106 may include any suitable volatile memory, such as a suitable DRAM (Dynamic Random Access Memory). In some embodiments, the memory 2106 may include a double data rate type four synchronous dynamic random access memory.

[0243] For one embodiment, the control module 2104 may include one or more input / output controllers to provide an interface to the non-volatile memory / storage device 2108 and the input / output device(s) 2110 .

[0244] For example, non-volatile memory / storage 2108 may be used to store data and / or instructions 2114. Non-volatile memory / storage 2108 may include any suitable non-volatile memory (e.g., flash memory) and / or may include any suitable non-volatile storage device(s) (e.g., one or more hard disk drives, one or more optical disk drives, and / or one or more digital versatile optical disk drives).

[0245] The non-volatile memory / storage device 2108 may include storage resources that are physically part of the device on which the apparatus 2100 is installed, or it may be accessible to the device without being part of the device. For example, the non-volatile memory / storage device 2108 may be accessed via (one or more) input / output devices 2110 over a network.

[0246] The input / output device(s) 2110 may provide an interface for the apparatus 2100 to communicate with any other appropriate device, and the input / output device 2110 may include a communication component, an audio component, a sensor component, etc. The network interface 2112 may provide an interface for the apparatus 2100 to communicate through one or more networks, and the apparatus 2100 may wirelessly communicate with one or more components of a wireless network according to any of one or more wireless network standards and / or protocols, for example, accessing a wireless network based on a communication standard, such as WiFi (Wireless Fidelity), 2G (2nd Generation Wireless Telephone Technology), 3G (3rd Generation Wireless Telephone Technology), 4G (4th Generation Wireless Telephone Technology), 5G (5th Generation Wireless Telephone Technology), etc., or a combination thereof for wireless communication.

[0247] For one embodiment, at least one of the processor(s) 2102 may be packaged together with the logic of one or more controllers (e.g., a memory controller module) of the control module 2104. For one embodiment, at least one of the processor(s) 2102 may be packaged together with the logic of one or more controllers of the control module 2104 to form a system-in-package. For one embodiment, at least one of the processor(s) 2102 may be integrated on the same die with the logic of one or more controllers of the control module 2104. For one embodiment, at least one of the processor(s) 2102 may be integrated on the same die with the logic of one or more controllers of the control module 2104 to form a system-on-chip.

[0248] In various embodiments, the device 2100 may be, but is not limited to, a terminal device such as a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). In various embodiments, the device 2100 may have more or fewer components and / or different architectures. For example, in some embodiments, the device 2100 includes one or more cameras, a keyboard, a liquid crystal display screen (including a touch screen display), a non-volatile memory port, multiple antennas, a graphics chip, a dedicated integrated circuit, and a speaker.

[0249] Among them, the main control chip can be used as a processor or control module in the detection device, sensor data, location information, etc. are stored in a memory or non-volatile memory / storage device, the sensor group can be used as an input / output device, and the communication interface may include a network interface.

[0250] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0251] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0252] The present application embodiment is described with reference to the flowchart and / or block diagram of the method, terminal device (system) and computer program product according to the embodiment of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, and the combination of the process and / or box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device produce a device for implementing the function specified in one process or multiple processes in the flowchart and / or one box or multiple boxes in the block diagram.

[0253] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0254] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce computer-implemented processing, so that the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0255] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0256] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0257] The above is a detailed introduction to a communication equipment inspection method and device, an electronic device and a machine-readable medium provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for inspecting communication equipment, characterized in that: The method comprises: Receiving an inspection task instruction; the inspection task instruction includes: a first instruction and a second instruction; Utilizing the large language model and the logic processing model, the inspection task instruction is executed to obtain result data corresponding to the inspection task instruction; Generate an inspection result report based on the result data; The use of a large language model and a logic processing model to execute the inspection task instruction includes: The large language model reads the current instruction in the inspection task instruction; If the current instruction is the first instruction, the large language model obtains the first instruction result corresponding to the first instruction from the log of the database; the log of the database records the collection result of the device information of the communication device; If the current instruction is a second instruction, the large language model sends the second instruction to the programmable logic controller PLC, so that the PLC sends the second instruction to the logic processing model, and the logic processing model executes the second instruction to obtain a corresponding second instruction result, and receives the second instruction result forwarded by the PLC; The large language model fuses the first instruction result and the second instruction result to obtain result data corresponding to the inspection task instruction.

2. The method according to claim 1, characterized in that The large language model obtains a first instruction result corresponding to the first instruction from a log of a database, including: In the case where the first instruction is an inspection indicator instruction, the number of indicators is determined according to the task category of the inspection task, and the target inspection indicator corresponding to the number of indicators is obtained from the inspection indicator database according to the weight of the inspection indicator; different task categories correspond to different numbers of indicators; In the case where the first instruction is an inspection object instruction, the target object category of the inspection object is determined according to the task category of the inspection task, and the target inspection object corresponding to the target object category is obtained from the inspection object database according to the object category of the inspection object; different task categories correspond to different target object categories; In the case where the first instruction is an inspection index value instruction, the index value of the target inspection index corresponding to the target inspection object is obtained from the index value database.

3. The method according to claim 2, characterized in that The large language model obtains a first instruction result corresponding to the first instruction from a log of a database, further comprising: In the case where the first instruction is an association relationship filtering instruction, determining a first associated target inspection object having an association relationship among the target inspection objects according to an association relationship database; The index value of the target inspection index corresponding to the first associated target inspection object is obtained from the index value of the target inspection index corresponding to the target inspection object.

4. The method according to claim 2, characterized in that: The large language model obtains a first instruction result corresponding to the first instruction from a log of a database, further comprising: In the case where the first instruction is an association relationship extension instruction, determining a second associated target inspection object corresponding to the target inspection object according to an association relationship database; The index value of the target inspection index corresponding to the second associated target inspection object is obtained from the index value database.

5. The method according to claim 2, characterized in that: The first instruction result includes: the last first instruction result; the last first instruction result includes: the index value of the target inspection index corresponding to the target inspection object; The large language model fuses the first instruction result and the second instruction result, including: The large language model adds a second instruction result to the indicator value of the target inspection indicator corresponding to the target inspection object.

6. The method according to claim 2, characterized in that The second instruction is a logic judgment instruction; The logic judgment instruction includes: The first logic judgment instruction is used to judge whether the indicator result is abnormal or normal according to the indicator value and indicator threshold of the target inspection indicator corresponding to the target inspection object; The second logic judgment instruction judges the logical relationship between the indicator results of any two target inspection indicators; The third logic judgment instruction is used to determine the abnormal probability corresponding to the target inspection object when the logical relationship between the indicator results of any two target inspection indicators of the target inspection object is inconsistent.

7. The method according to claim 2, characterized in that The logic processing model executes the second instruction to obtain a corresponding second instruction result, including: The logic processing model obtains the most recent N indicator values ​​for any two target inspection indicators whose indicator results corresponding to the target inspection objects are inconsistent; N is a positive integer greater than 1; According to the index values ​​of the last N times, determine the dynamic thresholds corresponding to any two target inspection indicators; Compare the most recent N indicator values ​​with the dynamic threshold and collect statistics on the comparison results; According to the statistical results of the comparison results, the abnormal probability corresponding to the target inspection object is determined.

8. The method according to any one of claims 1 to 7, characterized in that: Generating an inspection result report according to the result data includes: Generate a corresponding inspection result table according to the result data; the inspection result table includes: a first field corresponding to the first instruction result, and a second field corresponding to the second instruction result; The first field includes: a target inspection object field, a target inspection index field, and an index value field; the second field includes: a logical relationship field between any two target inspection indexes corresponding to the target inspection object, and an abnormal probability field corresponding to the target inspection object.

9. The method according to any one of claims 1 to 7, characterized in that: The generating of the inspection result report according to the result data also includes: Obtain a report template; the fields in the report template include: a task category field of the inspection task, an object quantity field, a quantity field of target inspection objects of the object category, an indicator quantity field, a normal target inspection indicator quantity field, an abnormal target inspection indicator quantity field, and a distribution field of abnormal target inspection indicators in the target inspection objects; According to the inspection result table, the fields in the report template are filled in to obtain the inspection result text.

10. The method according to any one of claims 1 to 7, characterized in that: The generating of the inspection result report according to the result data also includes: For any two target inspection indicators whose logical relationship is inconsistent with the target inspection object, a corresponding prompt text is generated.

11. A patrol inspection device for communication equipment, characterized in that: The device comprises: A receiving module, used for receiving an inspection task instruction; the inspection task instruction includes: a first instruction and a second instruction; An instruction execution module, used to execute the inspection task instruction by using a large language model and a logic processing model to obtain result data corresponding to the inspection task instruction; A report generation module, used to generate an inspection result report based on the result data; Wherein, the instruction execution module includes: A reading module, used for reading the current instruction in the inspection task instruction by a large language model; A first instruction execution module, configured to obtain, by the large language model, a first instruction result corresponding to the first instruction from a log of a database if the current instruction is a first instruction; the log of the database records a collection result of device information of the communication device; A second instruction execution module, configured to, if the current instruction is a second instruction, send the second instruction to a programmable logic controller (PLC) by the large language model, so that the PLC sends the second instruction to the logic processing model, and the logic processing model executes the second instruction to obtain a corresponding second instruction result, and receive the second instruction result transferred by the PLC; The fusion module is used to fuse the first instruction result and the second instruction result by using a large language model to obtain result data corresponding to the inspection task instruction.

12. An electronic device, characterized in that: include: processor; and A memory having executable codes stored thereon, which, when executed, causes the processor to execute the method according to any one of claims 1 to 10.

13. A machine-readable medium having executable codes stored thereon, which, when executed, causes a processor to perform the method according to any one of claims 1 to 10.

Images