JP2854645B2 - Fault diagnosis device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Purpose of the Invention] (Industrial Application Field) The present invention relates to a failure diagnosis device for a power receiving and transforming device or the like.

(Prior art) For example, gas insulated switchgear (Gas Insulated
In the switchgear (GIS) fault diagnosis, there is a method in which an expert system is executed based on data of various sensors provided in the device to perform a fault diagnosis and obtain a result. In this case, in the GIS, the number of sensors is generally small, so the number of failure diagnosis rules is also small. Therefore, at present, rules for the expert system are created for each panel.

Therefore, even if there is only one rule database (referred to as a database storing rules) used in the expert system, there is no problem in failure diagnosis.

(Problems to be Solved by the Invention) However, as an application of the fault diagnosis of the GIS, power receiving and transforming equipment in general industry, for example, a cubic type GIS (C
-GIS), gas transformers, front-maintenance switchgear (SG), etc., when trying to diagnose failures by attaching sensors to a large number of devices, the number of sensors is enormous due to the large number of switchgears. It becomes something.

The number of sensors and the number of rules in the expert system correspond one-to-one with the number of sensors, and one or more rules correspond to one parameter. As a result, the number of rules is correspondingly increased, resulting in a huge number of rules.

However, when the number of sensors becomes a certain number or more and the number of rules increases, the number of rules stored in the rule database is limited, so that a failure diagnosis system cannot be constructed with a single database. The problem arises.

As an example, two C-GIS and two gas transformers
Basically, in the substation equipment with 20 SGs, if the number of failure diagnosis rules for each equipment is α, β, γ, the number of rules is created for each board, so a single rule database requires The number of rules is as large as α × 2 + β × 2 + γ × 20.

The present invention has been made in view of the above problems, and has as its object to construct a versatile failure diagnosis system by reducing the number of failure diagnosis rules.

[Means for Solving the Problems] According to the present invention for achieving the above object, data of various sensors provided in various devices is stored and held for each type, and then transmitted to a failure diagnosis unit. In the failure diagnosis device that determines the cause of equipment failure, a database of an expert program that starts in an abnormal state is set for each model, and the necessary data among the above data before the execution of the expert program is fetched and the parameters of the sensor are set. The gist of the present invention is to have a failure diagnosis unit that performs control and calculation for determining a sensor corresponding to a parameter output based on the set parameter after setting and executing the expert program.

(Operation) Create a rule database to be used by the fault diagnosis expert program for each type of power receiving and transforming equipment. Before and after execution of the expert program, set sensor parameters as data pre-processing and execute the expert program as post-processing. The sensor corresponding to the parameter output as a result of the failure diagnosis is determined based on the parameter set in the preprocessing. As a result,
The core part of the expert program can be standardized, so that the number of rules stored and used in the rule database can be reduced.

(Example) Here, an example of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram of the failure diagnosis device.

As shown in FIG. 1, reference numeral 1 denotes a sensor which outputs analog data. Although one sensor 1 is shown in FIG. 1 in a simulated manner, many sensors 1 are provided in each device for performing a failure diagnosis.

A local monitor 2 converts analog data from the sensor 1 into digital data. Reference numeral 3 denotes a programmable controller that stores and holds data at a predetermined address for each type of data.

In executing the expert system, the failure diagnosis unit 4 specifies a rule database name corresponding to the model set in the initialization file.

Further, the failure diagnosis unit 4 executes the preprocessing program 4a,
Execution of expert program 4b, post-processing program 4c
, And display / save 4d.

Among these, the preprocessing program 4a fetches necessary data from the programmable controller (PC) 3 in which the data is stored and sets it in the input file of the expert program, and sets parameters corresponding to each sensor. is there.

After performing the failure diagnosis by executing the expert program 4b, the post-processing program 4c outputs the failure cause from the output file in which the diagnosis result is output, and sets the sensor corresponding to the output parameter in the input file. The determination is based on the parameters.

The display / save 4d is for displaying the cause of the abnormality and identifying and saving the data.

The above is a brief description mainly on the specification of the rule database and the setting / determination of the parameters, which will be further described with reference to FIGS. 2 to 4.

Here, FIG. 2 is a simplified block diagram of the whole failure diagnosis apparatus, FIG. 3 is a hardware diagram of the failure diagnosis unit 4 which is the personal computer of FIG. 2, and FIG. 4 is a processing flow.

In FIG. 2, a number of sensors 1 provided in the main circuit device are connected to a PC 2. The PC2 has a function of converting an analog signal from the sensor 1 into a digital signal and transmitting the digital signal to the PC3.

The PC 3 has a function of storing and holding data at a predetermined address according to the type of data, and transmitting the stored and maintained data to the failure diagnosis unit 4.

The failure diagnosis unit 4 includes a controller 6 as shown in FIG.
Performs serial / parallel conversion as necessary to input data.

In the failure diagnosis unit 4, the RAM 9 stores data and also stores the processing procedure written in the floppy disk or the hard disk 7 and held in the form of a file when the system is started.

The CPU 8 performs control such as data identification and comparison, and performs control and calculation in accordance with a processing procedure.

Input / output to / from a CRT, a printer, or a keyboard 11 is performed via each controller 10.

The configuration is as described above. Next, the processing flow will be described with reference to FIG.

In step A, the system is initialized. In step B, the data of the sensor 1 taken and stored in the PC 3 is input to the CPU 8 via the controller 6.

In step C, the CPU 8 identifies the input data and the sensor in association with each other, and stores the data at a predetermined address in the RAM 9.

In step D, the CPU 8 compares the data with the upper and lower limit values preset for each sensor in the RAM 9 and determines whether the CPU 8 is in an abnormal state or a normal state.

When a normal state is determined in step D, the process returns to step B, and when an abnormal state is determined, step E is performed.
Move on to

In Step E, the board number corresponding to the model of the substation equipment in which the abnormality has occurred is specified based on the abnormal state data. In this case, the rule database corresponding to the specified model is set in the initialization file in the RAM 9.

In step F, the preprocessing program is executed from the board number specified in step E, and the parameters and data of the related sensors are set in the input file in the RAM 9.

In step G, the expert program is executed. Next, in step H, the failure diagnosis result is output to an output file in the RAM 9. In this case, the output file is output in a format in which the subject of the failure, the content of the failure, and the reliability thereof are determined by the expert program. As a reason for the inference, parameters, their contents, and set values are also output.

In step I, the post-processing program is executed, and based on the parameters set in the input file in step F, the sensor corresponding to the parameter output as the failure diagnosis result by the execution of the expert program in step G is identified.

Then, in step J, based on the contents of the output file, a failure diagnosis result
To be displayed.

 Thereafter, the process returns to step B.

Thus, for example, two C-GIS and two gas transformers
Basically, in a substation device with 20 SGs, when the number of failure judgment rules for each device is α, β, γ, a rule database is set for each model, so the number of rules in the expert system is α + β + γ, and the number of rules can be greatly reduced.

[Effects of the Invention] A rule database used in a failure diagnosis expert program is created for each type of power receiving and transforming equipment, and before and after the execution of the expert program, sensor parameters are set as data pre-processing, and the expert program is used as post-processing. Is determined based on the parameters set in the pre-processing, the sensors corresponding to the parameters output as the result of the failure diagnosis resulting from executing. As a result,
Even if the number of main circuit devices increases and the number of sensors increases accordingly, the core part of the expert program can be standardized, and the number of rules stored and used in the rule database can be reduced, limiting the limit. Never go over.

[Brief description of the drawings]

1 to 4 show an embodiment of the present invention, FIG. 1 is a schematic block diagram of a failure diagnosis device, FIG. 2 is a simplified block diagram of hardware of the failure diagnosis device, and FIG. FIG. 2 is a block diagram of the failure diagnosis unit, and FIG. 4 is a processing flowchart. 1 ... Sensor, 2,3 ... PC, 4 ... Fault diagnosis unit, 6,10 ... Controller, 8 ... CPU, 9 ... RAM.

Claims (1)

(57) [Claims] An expert program which is activated in an abnormal state in a failure diagnosis device for storing and holding data of various sensors provided in various devices for each type and transmitting the data to a failure diagnosis unit to diagnose the cause of the failure of the device. The database is set for each model, and before the execution of the expert program, the necessary data among the above data is fetched and the parameters of the sensor are set. After the execution of the expert program, output is performed based on the set parameters. A failure diagnosis device comprising a failure diagnosis unit that performs control / calculation for determining a sensor corresponding to a parameter set.