KR920003344B1 - Maintenance systems of switching centre - Google Patents

Abstract

The circuit includes a message managing means (2) which carries out real time displays, message searches, record preservation, restorations, extractions, counting and summarizing. A message managing terminal (7) supplies switch connection table managing data to the user, and a disorder diagnosing means (1) analyzes the cause of disorder based on the message of the message managing means (2), and reports a required action to the user. A color monitor (5) and an audible warning means (4) informs the status of the occurrence of disorders to the user, and a disorder diagnosing terminal (6) furnishes analytical data on the disorders of the system.

Description

Maintenance method and device of electronic changer

1 is a schematic configuration diagram of the present invention.

2 is a functional block diagram of the apparatus.

3 is a detailed configuration diagram of a detailed internal functional module of the maintenance apparatus according to the present invention.

4 is a flow chart of a maintenance method according to the present invention.

* Explanation of symbols for main parts of the drawings

1: Fault Diagnosis Computer 2: Message Management Computer

3 & 8: Printer 4: Home Alarm

5: color monitor alarm display device 6: fault diagnosis terminal

7: message management, terminal 9: electronic exchange that is subject to maintenance

The present invention relates to a method and apparatus for performing maintenance on an exchange based on fault knowledge of an electronic switch and using a portion of artificial intelligence.

Among the conventional general network operation preservation devices, the exchange operation preservation device, in particular, stores data in a database in a certain format so that the data generated from the exchange can be easily analyzed for maintenance of the exchange. In addition to fault recovery and handling devices.

The problem of the prior art is that, first, the quality of troubleshooting has a significant relationship with the operator's knowledge, and therefore, the failure handling is different for each operator. Third, it is difficult to permanently preserve the knowledge possessed by the experience operator, so that knowledge of troubleshooting experience is lost in the absence of the operator. Fourth, the exchange structure necessary for troubleshooting is complicated and diverse. There is a lighting constraint that it takes time to consult a number of exchange manuals.

Accordingly, the present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide a new exchanger maintenance method and apparatus in which a part of artificial intelligence utilizes the features of an expert system.

The maintenance method of the present invention for achieving the above object, the first step of retrieving the relevant message according to the characteristics of the message unique number by analyzing the input message from the electronic switch, and after performing the first step, The second step of finding a message corresponding to the main cause of the alarm, determining the detailed failure type from the message, and suggesting a possible failure situation where the failure type can occur, and the failure cause hypothesis after performing the second step. The third step of interactively verifying the authenticity of the hypothesis generated by the verification, and after performing the third step, the fault location is determined using the structural knowledge of the electronic switchboard based on the input data of the operator. The fourth step is to output the troubleshooting measures based on the fault location.

The maintenance apparatus of the present invention is connected to a maintenance target electronic switch and a modem, collects and stores messages from the electronic switch, and manages a message database provided therein, real time display, message search, record preservation, Message management means for restoring, extracting, counting and summarizing, and message management means connected to the message management means to provide the operator with an exchange access table manager fee provided by the message management means in a multi-window work screen processing manner to the operator. The message management means includes a terminal, a message base connected to the message management means through a local area network to exchange messages in a packet stream manner, and a knowledge base storing structural knowledge and experience knowledge of the electronic switchgear 1 to be maintained. Analyze the cause of failure by inputting message received from It is connected to the fault diagnosis means for providing a countermeasure to the operator, the color monitor alarm means and an audible alarm means connected to the fault diagnosis means to convey the status of the fault occurrence to the operator in the form of audible and audible, It is provided with a fault diagnosis terminal for providing the operator with the fault diagnosis report, message management and message interpretation, the structure description of the electronic switchboard provided by the fault diagnosis means.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic configuration diagram of a maintenance apparatus according to the present invention. In the drawings, (1) is a fault diagnosis computer, (2) is a message management computer, (3) and (8) is a printer, (4) is a home alarm, (5) is a color monitor alarm display device, and (6) The fault diagnosis terminal (work seat), (7) is a message management terminal (work seat), (9) is an electric switchboard to be subjected to maintenance, and (10) shows a maintenance apparatus of the present invention, respectively.

The fault diagnosis computer 1 performs a fault diagnosis based on a knowledge base storing structural knowledge and empirical knowledge about the target electronic switch 1 of the present invention, and inputs a message received from the message manager 2. All the knowledge that needs to be diagnosed is based on the exchange knowledge base and other fault related information is requested and processed by the database in the message management computer (2). Functions provided to the operator include real-time fault monitoring and fault analysis to handle detailed fault diagnosis. Other supporting intelligences include message interpretation and electronic switchboard (9) structure description function, fault diagnosis report and message management. There exists a function for performing a function in the troubleshooting terminal 6 and the like.

The printer 3 connected to the failure diagnosis computer is in charge of outputting a failure diagnosis report.

The color monitor alarm display device 5 is responsible for visual information among the devices that provide a status of a failure occurrence in the operator, and the home alarm device 4 is responsible for an audible alarm.

Although the message management computer 2 is similar in function to a conventional switchboard operation maintenance device, the configuration method works by adding a function and a component for supporting the failure diagnosis computer 1, and an additional element is the message management terminal 7 To provide multiple work screens and interface with the fault diagnosis computer (1). The function of the message management computer (2) is the exchange message general management, which is connected to the target exchange (9) and the modem to collect and store messages, to manage the database, real-time display, message retrieval, record preservation, restoration, extraction, counting and There is a function of providing a switch connection table management and the like to the message management terminal 7 in a multi-window work screen processing method and providing a message management statistics report to the operator through the printer 8.

In addition, the apparatus used for the configuration of the message management computer 2 and the failure diagnosis computer 1 is a general-purpose mini-class computer, and the functions of each computer are performed by onboard software. In addition, the information communication connection between the two computers 1 and 2 is a local area network configuration, and the data communication protocol is TCP / IP. Based on this, the message transmission method is based on the packet stream method. It was made to meet the purpose of construction. This improves transmission performance for various information exchanges in two computer periods, and a loose coupling structure, which is a basic configuration that forms a mutual information exchange method between a database in the message management computer 2 and a knowledge processing database in the failure diagnosis computer 1. To achieve. This is one of the features of the apparatus of the present invention, which seeks to improve the quality of the important maintenance of the communication network by simultaneously maintaining general data processing and knowledge processing.

2 is a detailed configuration and connection diagram of the fault diagnosis computer 1 and the message management computer 2, in which 2a is an exchanger connector, 2b is a message collector, 2c is a message DB device, and 2d is a message real-time processing delivery. 2e is a message DB search transmission unit, 1a is a message real-time processing receiving unit, 1b is a message DB search receiving unit, 1c is a failure monitoring unit, 1d is a failure analysis unit, 1e is a knowledge base DB device, and 1f is a knowledge base management unit. Indicates.

The exchanger connector 2a is connected to the target electronic switch 9 by a modem and internally connected to the input / output (I / O) bus of the computer to message all messages inputted to a device that controls the external input / output device inside the computer. It is connected to transmit the message to the collection block 2 and to the real-time processing transfer unit 2d for processing for delivery to the failure diagnosis unit 1.

The function of the message collecting unit 2b is the central processing unit (CPU) of the computer. The message is collected and connected to the disk I / O bus and stored in the database in the message DB device 2c. In response to the request of the diagnostic computer 1, communication with the message DB retrieval transmission section 2e is connected.

The fault diagnosis computer 1 includes a message real-time processing receiver connected to the Ethernet real-time I / O bus for mutual communication corresponding to the message real-time processing transfer unit 2d and the message DB search transfer unit 2e. 1a) and a message DB search receiving unit 1b, which are connected to the message real-time processing receiving unit 1a and the message DB search receiving unit 1b by a CPU bus and monitor a failure in real time in real time. It is provided with a failure analysis unit (1d) for.

In addition, the failure monitoring unit 1c and the failure analysis unit 1d are connected to each other by the CPU BUS for information exchange, and information exchanged with each other achieves each operator knowledge inference.

The result knowledge of the failure monitoring unit 1c is stored in the knowledge base DB device 1e connected to the failure monitoring unit 1c by a disk input / output bus, and connected to the failure analysis unit 1d by a CPU bus. The knowledge base manager 1f connected to the knowledge base DB device 1e by a disk input / output bus transfers the monitoring knowledge for failure analysis to the knowledge base DB device. That is, the experience knowledge resulting from the failure analysis unit 1d is again stored in the knowledge base DB device 10 through the knowledge base management unit 1f in order to use it again at the next failure.

Such a connection configuration enables conversion to a device in which past processing experiences due to repeated failures are accumulated, and the present invention becomes intelligent as time passes.

3 is a detailed configuration diagram of a detailed internal function module of the maintenance apparatus according to the present invention.

In the figure, 21 is an exchanger connection module, 22 is a message collection and storage module, 23 is a message database, 24 is a message general management module, 25 is a work seat processing module, and 26 is a Message management terminal, 27 is a message real-time delivery processing module, 28 is a fault diagnosis query processing providing module, 29 is a report processing module, 30 and 45 a printer, 31 is a real-time message receiving The processing module, 32 is a failure analysis processing module, 36 is a knowledge base 37, a knowledge base management module, 38 is a fault diagnosis work seat processing module, 39 is a color monitor alarm expression processing module, 40 denotes an audible alarm processing module, 41 a report processing module, 42 a fault diagnosis terminal, 43 an alarm expression terminal, 44 an audible alarm device, and 45 a printer. The functions of modules and parts are as follows.

First, in the detailed configuration of the message management computer 2, the hardware connection of the output device of the switch 9 and the input device of the computer 2 is RS232C, in order to receive a message from the exchange, using a modem. At this time, the switch connection module 21 exists as a software routine for driving the input, and the module stores the input message itself in the order of input to the queue list.

Dozens of messages and the storage processing module 22 is responsible for the function of collecting, processing and storing the message in the message database of the message management computer 2 from the message queue list, accordingly the message real-time delivery processing module 27 ) Transmits the message to the fault diagnosis computer 1, and the message real-time reception processing module 31 receives the forwarded message at the fault diagnosis computer 1 side.

The message database 23 due to the message storage process is composed of attributes such as message generation time, type, type, name, location, etc., and the message generation time becomes a key. Message occurrence time and type are indexed respectively. The message occurrence time is the time (year, month, day, hour, minute, second) at which the message occurred at the exchange, and the type indicates the degree of importance of the message. The message type is indicated by the serial number according to the alarm or fault, and the location is the name of the fault unit.

The message general management module 24 which carries out the message information in the message database 23 performs the message database 23 such as creation, calling, regeneration, tape, preservation, disk preservation, and restoration of the message database 23. Management function and the ability to extract the message that meets the condition for the message stored in the database, the ability to record the extracted message in the specified file, the ability to count the message that meets the input condition, and the stored message Perform general functions on the message database, such as the ability to summarize over time.

In addition, the report processing module 29 processes the message general various statistics with the message information, the result is requested to the workability processing module 25 and output to the printer 30.

The message management task seat processing module 25 divides the general management and report generation and output sections into the message database 23 and the real-time processing of messages input from the exchange. Basic screen processing is based on window processing technique. There are three types of window types: base panel and operator and interactive window. The message management terminal 26 uses a color monitor, and inputs to the terminal are performed using a keyboard and a mouse.

The fault analysis query processing module 32 is a processing module that interacts with the fault diagnosis computer 1, and in addition to the message real-time processing 27, 31, a query request for message information is performed during the fault analysis process in the fault diagnosis computer 1; It is responsible for transmitting the query contents to the message management computer 2 when present.

The fault diagnosis query processing module 28 is present in the message management computer 2 and receives the contents of the query transmitted to the message management computer 2 to transmit information corresponding to the heterogeneity to the failure analysis query processing module 32. Play a role. In addition, the message management command processing module 33 present in the fault diagnosis computer 1 enables the command of the message general management module 24 to be used as the original command in the fault diagnosis terminal 42 of the fault diagnosis computer 1 as well.

The following is a description of the functional module in charge of the processing in the failure diagnosis computer 1, and describes the failure monitoring and analysis on the basis of the message unit of one unit of the message real-time reception processing module 31. The failure monitoring processing module 34 is a module which collectively controls the failure monitoring operation mode of the present invention. The fault condition is monitored in real time from the message output to the exchange, and if a fault is detected, a simple fault diagnosis is performed using only the knowledge necessary for emergency measures in the knowledge base 36, and the result is analyzed as a fault diagnosis task seat processing module ( The maintenance personnel are notified via the fault diagnosis terminal 42, the alarm expression terminal 43, the audible alarm device 44, and the printer 45 through 38).

All knowledge for diagnosing the actual fault is largely divided into structural knowledge and operational alarm knowledge of the electronic switchboard, and provides the knowledge necessary for the fault monitoring processing module 34 and the fault analysis processing module 35 to diagnose the fault. . The detailed flow of the process of diagnosing faults of the two modules is shown in FIG. 4, which will be described later.

In addition, when adding new knowledge, it is possible to use the knowledge base management module 37. Since diagnostic knowledge determines the accuracy of the diagnostic results of the device, further modification of the diagnostic knowledge must be managed and carried out efficiently. Therefore, the knowledge base management module 37 allows the knowledge manager to manage the knowledge base 36 through the fault diagnosis terminal 42. This module is composed of knowledge editor and knowledge tester. It is easy to modify and add knowledge with knowledge editor, and test of added or changed knowledge is done through knowledge tester.

The color monitor alarm expression processing module 39 receives the message unique number, the fault exchanger name, and the board name copper monitoring result from the fault monitoring processing module 34, and transmits the alarm state of the switch to the alarm expression terminal 43, that is, the color monitor. This function displays characters and pictures easily for the operator to understand.

Representation of characters is a mixture of Korean and English, and graphic function is used to express system structure easily. When the corresponding function board and unit are determined by failure analysis, the physical board knowledge is translated into the shelf and slot numbers of the rack to which the board belongs, using the physical structure knowledge of the knowledge base 36. Tells the operator the location of and the higher processor the board belongs to. And, the description of the structure of the exchange uses a menu selection method.

4 is a detailed flowchart showing each process from the maintenance method according to the present invention, that is, from failure monitoring and analysis to the final failure countermeasure.

The message output from the electronic switchgear 9 has a unique number, which has an implication that defines the nature of the message and the failure subsystem. If the message is in the number range of -100, the power system, the processor and OS system in the number range of 200-300, the subscriber line and relay line system in the number range of 400-500, and the switch network system in the number range of 600-700, 800 In the case of the -900 number, it is classified as maintenance.

Accordingly, based on the above classification criteria, the message inputted from the electronic switchgear 9 is analyzed (100). Since only fault messages and path messages are used for fault diagnosis, message filtering is performed at this stage.

Determine according to the characteristics of the input message unique number (101), in the case of a message having information that can determine the failure unit, after a step (102) of roughly determining the failure unit, and then retrieves the relevant message, The other messages go directly to the related message retrieval step 103. The kind of message that can go to the related message search 103 corresponds to the case of message class.

In the failure unit determination step 102, a failure unit having a broad meaning is determined using only the currently input message itself.

In the related message retrieval step 103, messages related to the currently input message are searched among the input messages and the message is searched based on the mutual priority relationship of the failures. Herein, the related message refers to a message generated at the same location or a message generated at a next lower module. Failure analysis is usually initiated with an alarm message and targets the failure message and the alarm message indicated before the alarm message occurred to find the message that caused the current failure that caused the current message.

Then, the message corresponding to the main cause of the alarm is found, and the detailed failure type is determined from this message (104). The failure type determined here is data for generating a hypothesis in the hypothesis setting step 105, which is a failure cause, which refers to a unit in the functional model of the sub-system.

In the failure cause hypothesis setting step 105, a hypothesis is suggested as a possible failure situation in which the failure type determined through the failure type determination step 104 may occur. At this time, since the generation order of the hypothesis becomes the hypothesis verification order of the failure cause hypothesis verification step 105 as the next processing step, the hypothesis is set in the order of occurrence frequency based on the operational experience to shorten the hypothesis verification time.

Then, the cause of failure hypothesis verification (106) to verify the authenticity of the generated hypothesis, and checks the system of the system interactively with the operator.

Provide the operator with the method and location of the check and wait for the operator to enter the test results to confirm the hypothesis from the input data. At this time, the verification of the hypothesis is performed by using a history of failure, a man-machine communication (MMC) result, and an exchange structure knowledge (108, 109). If any fault is found during this process, verification is completed without checking the rest of the hypothesis.

After the verification, the failure location is determined based on the operator's input data (107). At this time, the decision result is presented in the unit of board or connector, and the structural knowledge of the electronic switch is used to determine the fault location.

If the fault location is found, based on the suggested fault location, the fault corrective action is output (110), and the fault corrective action includes the fault repair procedure.

Therefore, the present invention made as described above, by utilizing the characteristics of the expert system, which is a part of the artificial intelligence and processing in relation to the exchange message based on the exchange structure knowledge and operator experience knowledge, there is an effect capable of accurate maintenance.

Claims (2)

In the maintenance method of the electronic switch 9 which is a communication network component, a first step (100 to 100) of analyzing a message input from the electronic switch 9 and searching for a related message according to the characteristic of the message unique number. 103) and, after performing the first steps (100 to 103), finds a message corresponding to the main cause of the alarm, determines the detailed failure type from the message, and suggests a possible failure situation where the failure type can occur. After performing the second step 104 and 105 and the second step 104 and 105, the third step (106, 108 and 109) to interactively verify the authenticity of the hypothesis generated by verifying the hypothesis generated by the cause of failure, and the third step ( 106,108,109) After performing the operation, the fault location is determined using the electronic switchboard and structural knowledge based on the operator's input data, and the fault repair measures are output based on the determined fault location. Maintenance way around the electronic exchange, characterized in that is carried out by a step 4 (107 110). In maintenance device of electronic switchboard, it is connected to maintenance target electronic switchboard (9) by modem and collects and stores messages from all electronic switchboard, and manages its own message database, real-time display, message search Message management means (2) for storing, restoring, extracting, counting, and summarizing records; and connected to the message management means (2), the message management means (2) being provided in a multi-window working screen processing scheme. A message management terminal 7 which provides an operator connection table manager fee to an operator, is connected to the message management means 2 and a local area network to send and receive a message in a packet stream manner, With a knowledge base that stores structural knowledge and experience knowledge for), the message received from the message management means (2) is inputted. Fault diagnosis means (1) which provides the operator with the cause of the analysis and processing of the fault, and a color monitor alert means (5) connected to the fault diagnosis means (1) to convey to the operator the status of the occurrence of the fault in a visible and audible form (5). ), Audible alarm means (4) and fault diagnosis means (1) connected to the fault diagnosis means (1) provided by the fault diagnosis means (1), the message management and message interpretation, the structure description of the electronic switch (9) A maintenance device for an electronic switchboard, characterized in that it comprises a troubleshooting terminal (6) for providing the operator with data.

Images