JPH06261378A - Fault diagnosis device for gas combustion - Google Patents

Abstract

PURPOSE:To efficiently execute maintenance and to improve reliability by remotely monitoring and automatically diagnosing the operation situation of a co-generation equipment in plural customers by means of various machine type constitution. CONSTITUTION:In the co-generation equipment as a customer equipment 8, various kinds of data showing the operation situation are collected. Data are sampled at every minute and are added up at every hour. Added up data are collected as daily report data for 24 hours once a day, and they are transmitted to a center-side system 2A. A light fault set point is provided and data for one hour is immediately transferred as abnormal data of a light fault when the fault occurs. When a serious fault by which the customer equipment 8 trips occurs, high speed sampling data provided for a different system is immediately transferred. A fault diagnosis system 3 which is online-connected with the center- side systems 2A and 2B by 'Ethernet' 4 loads an expert system, and remotely monitors/automatically diagnoses the customer equipment 8.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a failure diagnosis device for predicting and diagnosing a failure of gas combustion equipment such as a co-generation system.

[0002]

2. Description of the Related Art In recent years, global environmental problems have become serious,
The gas combustion equipment is also required to pollute the surroundings as much as possible and to effectively use energy. A co-generation system drives a generator with the shaft output of a heat engine such as a gas engine or gas turbine to supply electricity, and the exhaust heat at that time is also used effectively as a heat source for heating and cooling, hot water supply, etc. Is. As a prior art of such a co-generation system, for example, there is JP-A-4-52401 by the applicant of the present application.

Such a co-generation system is provided in, for example, a hospital, a school, a large building or a store, and a gas supplier keeps a normal operating state while remotely monitoring.

[0004]

In gas combustion equipment, especially in cogeneration systems, stable operation is indispensable while responding to demand fluctuations for each customer. Especially in hospitals and the like, it is absolutely necessary to avoid a situation in which equipment breaks down and to stop it, and to recover it immediately.

However, in the past, only the maintenance has been performed by periodic inspection, and the cause of the abnormality or the like has only been investigated by dispatching the operator to the site by telephone communication that the abnormality has occurred. Therefore, for example, it is impossible to perform maintenance at an appropriate time while observing the deterioration tendency. Also, when an abnormality occurs, the response is slow. further,
There is a problem that maintenance is inefficient because the situation is unknown until the site is visited.

An object of the present invention is to provide a failure diagnosing device for gas combustion equipment, which is capable of performing automatic diagnosis by remote monitoring for various types of gas combustion equipment.

[0007]

DISCLOSURE OF THE INVENTION The present invention provides a structure of gas combustion equipment in a failure diagnosis apparatus for gas combustion equipment for predicting and diagnosing a failure based on remote monitoring data transferred from a plurality of gas combustion equipment. A failure diagnosis device for gas combustion equipment, characterized in that it includes a processing means equipped with an expert system for performing diagnosis processing of remote monitoring data in accordance with a knowledge base systematized from above.

Further, in each of the gas combustion equipment of the present invention, a sampling means for sampling data at a predetermined first time interval, a first memory for storing the collected data for at least a predetermined second time, and a second memory The arithmetic means for arithmetically processing the data stored in the memory for each time, the second memory for storing the arithmetic output from the arithmetic means, and the data collected by the sampling means are within the normal operating range of the gas combustion equipment. When the warning range set in advance is exceeded, the storage content of the first memory is transferred as remote monitoring data, and when the collected data remains within the warning range, the storage content of the second memory is changed to
Processing means for transmitting remote monitoring data every predetermined third time is included.

Further, the gas combustion facility of the present invention is characterized by constituting a co-generation system.

[0010]

According to the present invention, the processing means is equipped with an expert system for diagnosing remote monitoring data transferred from a plurality of gas combustion equipment. The diagnostic process is performed according to a knowledge base systematically structured from the gas combustion facility. Since the knowledge base is structured from the structure of the gas combustion equipment, it is easy to find the deterioration tendency and take countermeasures, and it is possible to perform automatic diagnosis using remote monitoring data.

Further, according to the present invention, as remote monitoring data from each gas combustion facility, when the data exceeds a preset warning range within the normal operating range of the gas combustion facility, a predetermined first time interval is set. The data collected in step 2 is transferred only for the second hour. As a result, it is possible to effectively take measures based on automatic diagnosis while the operation of the gas combustion equipment is within the normal range. When the data stays within the warning range, the data calculated every second time is the third data.
Transferred every hour. As a result, the tendency of the operation over a relatively long period of time can be grasped, and it becomes possible to predict failure and take preventive measures.

According to the present invention, the gas combustion equipment is
Configure a generation system. The cogeneration system basically needs to be operated continuously. Since the operating status of the equipment can be automatically diagnosed based on the remote monitoring data, maintenance can be performed efficiently and the reliability of the cogeneration system can be improved.

[0013]

1 shows a schematic system configuration of an embodiment of the present invention. A failure diagnosis device 1 is provided at the remote monitoring center of the gas combustion facility. The failure diagnosis device 1 includes a customer monitoring system 2 and a failure prediction / diagnosis system 3. The customer monitoring system comprises, for example, a center side system (A system) 2A and a center side system (B system) 2B, each of which is realized by an industrial personal computer system. The center side system 2A and the center side system 2B are connected to the failure diagnosis system 3 via an Ethernet 4, which is a kind of local area network. Center side system 2A,
The output from 2B is also given to the large display 6 via the interface bus 5. The center side systems 2A and 2B are connected to the customer facility 8 side via modems 7A and 7B. The customer facility 8 constitutes a co-generation system based on, for example, a gas engine or a gas turbine. The outputs from the various sensors are processed by the customer side system 9, which acts as a data logger. The customer side system 9 is realized by an industrial personal computer, and a modem 10 is connected to it. The modem 10 is a customer monitoring system 2 of the failure diagnosis device 1 via the public line 11.
Connected to.

The daily report data and the abnormal data are transferred from the customer facility 8 to the failure diagnosis device 1 and are subjected to the failure diagnosis. The customer-side system 9 collects data, for example, every minute as the first time interval, and always stores 24 hours of data in the built-in memory. The data stored in the memory is subjected to a calculation process, such as a mean value calculation or a cumulative calculation process, every second hour, and the calculation processing result is stored again in the memory. The calculation results for 24 hours are also stored in the memory and are transferred as daily report data once a day.

The abnormal data includes major failure data and minor failure data. A major failure is a serious failure condition in which the gas combustion equipment trips and stops. The customer facility 8 is provided with a function of performing high-speed sampling of data, for example, every 0.1 seconds, and accumulating the data for a predetermined time, in addition to the above-mentioned data collection means for every minute. When a serious failure occurs, the high-speed sampled data before and after the failure is immediately transferred so that the cause of the failure can be easily investigated.

The minor failure of the abnormal data is stored in the memory when data exceeding the minor failure set value, which is a preset warning range, is generated within the normal operating range of the gas combustion equipment. Data for each minute is transferred for one hour. As a result, it becomes possible to grasp the tendency for the abnormality to occur before the abnormality actually occurs.

2 and 3 show a schematic configuration of the expert system realized by the fault diagnosis system 3 shown in FIG. The expert shell 13 includes a knowledge base 14, an inference engine 15 and a man-machine interface 16. In the expert shell 13, a knowledge base 14 is constructed by the procedural processing in the IF, THEN format based on the knowledge of a skilled expert. The inference engine 15 follows the predetermined schedule, or the man-machine interface 1
The knowledge base 14 is activated in accordance with the instruction given from 6, and the inference is advanced by referring to the external data 17. As a result, it is possible to automatically make a high-level judgment close to the judgment based on the know-how of the expert.

In FIG. 3, the expert shell 13
Together with the application 18 constitute an expert system 19. A ready-made system can be used as the expert system 19. Such an off-the-shelf expert system runs on UNIX, which is the operating system (OS) of the workstation. From the Ethernet 4 which is a kind of local area network (LAN), the application 18
Remote monitoring data is given to. The application 18 creates what corresponds to the external data 17 shown in FIG. 2 and gives it to the expert shell 13.

4 and 5 show a diagnostic mechanism in the embodiment shown in FIG. FIG. 4 shows a diagnosis based on daily report data, and FIG. 5 shows a diagnosis based on light failure data.

In the diagnosis by the daily report data shown in FIG. 4, (1) From the customer to the remote monitoring center. Daily report data for each customer is transferred according to a predetermined schedule. That is, the customer side system 9 transmits 24 hours worth of data to the monitoring center at a predetermined time for each customer via the modem 10 and the public line 11. In the remote monitoring center, for example, the system 2A on the center side receives the data via the modem 7A and stores the received data in the built-in hard disk (HD).

(2) From the remote monitoring center to the failure diagnosis system. The daily report data from each customer is sent as the equipment operation data of each customer to the failure diagnosis system 3 connected online from the center side system 2A via the Ethernet 4. The sent data is stored in the HD in the failure diagnosis system 3. Based on this data, the failure diagnosis system 3 analyzes and diagnoses the equipment operation status of each customer once a day.

The diagnosis shown in FIG. 5 upon occurrence of a slight abnormality is performed when any one of the sensors has an abnormality as shown in FIG.

(1) When a slight abnormality occurs, an alarm is issued from the customer side system 10 to the center side system 2A to which it is connected.

(2) The light failure data transfer is an interrupt process, and if any one of the sensors has an abnormality, the data for one hour of all the sensors is transferred. Such a minor failure is, for example, as shown in FIG. 6, an upper limit UA and a lower limit L.
Set for warning in the A range. In this case, the original failure is the upper limit UB and the lower limit LB, and if the range of the upper limit UX and the lower limit LX is exceeded, the safety device or the like provided in the customer facility will be in a trip state. Therefore, the range of the upper limit UX and the lower limit LX is called a major failure, and the range of the upper limit UB and the lower limit LB is called a minor failure, whereas the range of the upper limit UA and the lower limit LA is called a minor failure.

FIG. 7 shows a configuration of a cogeneration system as an example of the customer side equipment 8 shown in FIG. In the embodiment shown in FIG. 1, the system configuration of the customer equipment 8 as shown in FIG. 7 is displayed on the cathode ray tube (CRT) of the man-machine interface 16 shown in FIGS. 2 and 3 according to the operation of the operator. Further, it is possible to display the change state of the output of the sensor provided in each system.

In the co-generation system shown in FIG. 7, air is taken into the air cleaner 21 through the air supply line and cleaned. The air cleaned by the air cleaner 21 is sent to the carburetor 23 via the booster (B) of the supercharger 22. The carburetor 23 is supplied with city gas, which is a fuel gas, from a fuel gas line via a gas regulator 24 and is mixed with air. An air bypass valve 25 is provided in parallel with the carburetor 23, and the air is bypassed by opening the valve. The mixer from the carburetor 23 is cooled by the secondary cooling water in the aftercooler 26 and supplied to the gas engine 27. A generator 28 is directly connected to the gas engine 27 to take out gas combustion energy as electric power,
Extracts heat energy from exhaust gas. The exhaust gas is guided to the turbine (T) of the supercharger 22 and rotationally drives the supercharger 22. An exhaust bypass valve 29 is connected in parallel to the turbine of the supercharger 22. The exhaust gas is finally guided to the catalyst tank 30 to reduce the content of nitrogen oxides and the like.

A temperature sensor is provided in the air supply line,
The room temperature 31 is measured. A gas pressure sensor is provided on the inlet side of the carburetor 23 and measures the boost pressure 32. The opening degree of the air bypass valve 25 is also measured. A pressure gauge is provided in the fuel gas line to measure the gas inlet pressure 33. Temperature sensors are provided at the inlets and outlets of the secondary cooling water of the aftercooler 26 to measure the secondary cooling water engine inlet temperature 34 and the secondary cooling water engine outlet temperature 35, respectively. A pressure sensor is provided on the outlet side of the aftercooler 26 to measure the air-fuel mixture pressure 36.

The gas engine 27 includes an after cooler 2
The air-fuel mixture is supplied via 6, and primary cooling water and lubricating oil are also supplied. Temperature sensors are provided at the inlets and outlets of the primary cooling water to measure the primary cooling water engine inlet temperature 37 and the primary cooling water engine outlet temperature 38, respectively. Lubricating oil inlets and outlets are also provided with temperature sensors to measure the lubricating oil engine inlet temperature 39 and the lubricating oil engine outlet temperature 40, respectively. The governor output 41 and the cylinder exhaust gas temperature 42 are measured as the output characteristics of the gas engine 27 itself.

The output characteristic of the generator 28 is that the electric power is 4
3, voltage 44, current 45 and frequency 46 are measured. In order to supplement the power generated by the cogeneration system, the power that is purchased and received from commercial power is 4
7, measured as voltage 48, current 49 and frequency 50.

On the exhaust gas side, a supercharger inlet exhaust gas temperature 51 is measured by a temperature sensor provided on the turbine inlet side of the supercharger 22. An O2 sensor output 52 is measured from the oxygen gas sensor provided on the outlet side of the turbine. Further, on the downstream side, the exhaust gas temperature 53 in the catalyst tank is measured.

8 and 9 show examples of diagnostic rules.
FIG. 8 shows the contents of the knowledge base 14 shown in FIGS. 2 and 3, and FIG. 9 shows an example of application of diagnostic rules to the cogeneration system shown in FIG. The knowledge base structure 54 shown in FIG. 8 includes a gas engine rule 55,
Gas turbine rule 56 and failure case rule 57
Consists of. In remote monitoring of cogeneration systems, it is necessary to deal with the operating conditions of equipment of multiple customers of various model configurations, but from the principles of cogeneration systems, rules 55 for gas engines and rules 56 for gas turbines are adopted. Broadly classify. That is, FIG.
In the co-generation system using the gas engine 27 as shown in Fig. 5, even if various configurations are different, it is possible to determine the arrangement of the sensor with the greatest common divisor, and the diagnosis based on the output of the sensor can be ruled. Is.
For sensors that are not used by actual customers, their output need not be taken into consideration. On the other hand, the co-generation system using a gas turbine has a significantly different configuration. Therefore, as the knowledge base structure 54, a general knowledge is systematized and summarized, and a rule 55 for gas engine and a system of gas turbine and related equipment as systematic diagnosis know-how of the gas engine and related equipment. Each of the gas turbine rules 56 is provided as a computerized diagnostic know-how. The knowledge base structure 54 further includes, as a failure case management database, data on actual failure cases that have occurred in the past, and a reproducible failure card rule 57 that includes the contents of the trouble cards. Such a failure case rule 57 is searched at high speed by a keyword search.

FIG. 9 represents the diagnostic rules in the form of a fault tree. For example, in FIG. 7, when both the condition a1 that the outlet temperature of the lubricating oil engine is high (H) and the condition a2 that the cylinder exhaust gas temperature is high (H) are satisfied, the diagnostic result indicated by b is obtained. That is, the ignition timing is bad and the exhaust valve is bad. As the related sensors, the catalyst tank temperature, the air bypass valve opening degree, the booster pressure, the O2 sensor output and the like are shown.

As the condition a1 in which the lubricating oil engine outlet temperature is H, there are other conditions in which the primary cooling water engine inlet temperature indicated by a3 is H, or the lubricating oil engine inlet temperature indicated by a4 is H. There are cases where it holds. In these cases, different diagnostic results will be obtained. Also, the conditions a1 and a2
Even if is satisfied, the condition that the governor output at a5 is H or the condition that the generated power at a6 is H may also be satisfied. When the governor output of condition a5 is H, whether the inlet pressure of the fuel gas gas engine (GE) shown in condition a7 is H, or the mixture pressure shown in condition a8 is H, etc. There may be a condition of. The diagnostic results differ depending on each of those conditions,
New diagnostic results are prepared based on expert know-how.

The diagnosis result is graphically displayed on the display screen of the man-machine interface 16 shown in FIGS. The display screen can be easily switched by the operator operating the mouse, and a plurality of pieces of information can be simultaneously viewed on the window screen. For example, the system configuration for each customer as shown in FIG. 7 can be displayed, and a trend display showing the tendency of the data to be diagnosed, a process value display, a diagnostic message, etc. can be developed on each window screen. In this way, it is possible to examine the diagnostic content from various angles with a simple operation using the mouse.

By using the remote monitoring system as described above, if it is found that, for example, the vibration of the turbine tends to increase, it is possible to prevent the trip by making an early adjustment. Further, when a serious failure occurs in which the engine temperature rises due to the valve operation of the cooling water pipe, the data before and after the occurrence of the serious failure are transferred at high speed, so that the cause can be easily investigated.
Also, when the customer equipment was commissioned, the circuit breaker had tripped for unknown reasons, but when investigating the cause, it was found that there was a setting error in the insufficient power relay of the generator.

[0036]

As described above, according to the present invention, since remote monitoring data from a plurality of gas combustion facilities is automatically diagnosed by using an expert system, it is possible to predict the occurrence of a failure and prevent it from occurring. . As a result, a plurality of gas combustion facilities can be efficiently monitored remotely, and the reliability of the gas combustion facilities can be improved.

Further, according to the present invention, each gas combustion equipment is
The data is collected at the first time interval, and the result of the arithmetic processing at the second time interval is transferred at the third time interval. By the expert system automatically diagnosing the transferred remote monitoring data, it is possible to grasp the tendency of deterioration of characteristics and perform maintenance or the like at an appropriate time. When the data exceeds the warning range, the data stored in the first memory at each first time interval is transferred as remote monitoring data. Since detailed information can be known up to the point beyond the warning range, it is possible to analyze the cause of abnormality. By reflecting the result in the work at the site, the response at the time of abnormality is accelerated, and the efficiency of maintenance is further improved.

Further, according to the present invention, the cogeneration system can be operated with high reliability.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic system configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of an expert shell 13 formed in the failure diagnosis system 3 shown in FIG.

FIG. 3 is a block diagram showing a software configuration of the failure diagnosis system 3 shown in FIG.

FIG. 4 is a block diagram showing a data flow when the diagnostic system shown in FIG. 1 makes a diagnosis based on daily report data.

FIG. 5 is a block diagram showing a data flow when the diagnostic system shown in FIG. 1 makes a diagnosis based on light and minor failure data.

FIG. 6 is a graph showing a state in which a malfunction occurs easily.

FIG. 7 is a system configuration diagram of a co-generation system as an example of the customer-side facility 8 shown in FIG.

8 is a block diagram showing a configuration of a knowledge base in the failure diagnosis system 3 shown in FIG.

9 is a fault tree diagram showing the contents of the gas engine rule 55 shown in FIG. 8. FIG.

[Explanation of symbols]

 1 Failure Diagnosis Device 2 Customer Monitoring System 2A, 2B Center Side System 3 Failure Diagnosis System 4 Ethernet 6 Large Display 8 Customer Equipment 9 Customer Side System 11 Public Line 13 Expert Shell 14 Knowledge Base 15 Inference Engine 16 Man-Machine Interface 19 Expert System 22 Supercharger 23 Carburetor 25 Air bypass valve 27 Gas engine 28 Generator 55 Gas engine rule 56 Gas turbine rule 57 Failure case rule

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Mikio Osaki 4-1-2, Hirano-cho, Chuo-ku, Osaka City, Osaka Prefecture Osaka Gas Co., Ltd. (72) Inventor, Shozo Takeuchi 4-chome, Hirano-cho, Chuo-ku, Osaka-shi, Osaka No. 1 and 2 in Osaka Gas Co., Ltd.

Claims (3)

[Claims] 1. A failure diagnosing device for a gas combustion equipment, which predicts and diagnoses a failure based on remote monitoring data transmitted from a plurality of gas combustion equipment, and a knowledge base systematically structured from the structure of the gas combustion equipment. According to the invention, there is provided a failure diagnosis device for gas combustion equipment, comprising a processing means on which an expert system for performing remote monitoring data diagnosis processing is mounted. 2. Each gas combustion facility has a sampling means for sampling data at a predetermined first time interval, a first memory for storing the collected data at least for a predetermined second time, and every second time. The arithmetic means for arithmetically processing the data stored in the memory, the second memory for storing the arithmetic output from the arithmetic means, and the data collected by the sampling means are stored in advance within the normal operating range of the gas combustion equipment. When the set warning range is exceeded, the storage content of the first memory is transferred as remote monitoring data, and when the collected data remains within the warning range, the storage content of the second memory is set to a predetermined third time. The fault diagnosis device for gas combustion equipment according to claim 1, further comprising: a processing unit that transfers the data as remote monitoring data. 3. The failure diagnosis device for gas combustion equipment according to claim 1, wherein the gas combustion equipment constitutes a co-generation system.