JP2002073156A - Method and device for diagnosing gas turbine operating state - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for diagnosing gas turbine operating state, with which the performance deterioration or fault occurrence of a gas turbine can be highly accurately detected. SOLUTION: Operation data in the respective parts of a gas turbine G are detected, these detected operation data are standardized, and these standardized operation data are sampled for prescribed time. Then, moving average processing is applied to these sampled data and on the basis of these moving average data, the operating state of the gas turbine G is diagnosed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a method and an apparatus for diagnosing a gas turbine operating condition. In more detail, the performance of the gas turbine
The present invention relates to a gas turbine operation state diagnosis method and a diagnosis device for diagnosing based on operation data in various parts of a gas turbine.

[0002]

2. Description of the Related Art Conventionally, in order to diagnose the operating state of a gas turbine operated in a power plant or the like, operating data of a portion related to the performance of the gas turbine is detected during normal operation, and performance degradation is performed based on the detected value. Alternatively, various methods for detecting a failure (hereinafter, also simply referred to as performance degradation) have been proposed.

For example, Japanese Patent Application Laid-Open No. 11-3113 discloses that performance deterioration is detected by comparing current operation data of a gas turbine with a threshold value set based on operation performance data at the time of initial operation. .

However, in this method, there is a problem that the fluctuation range of the detection value of the operation data becomes too large, so that it is not possible to accurately detect the performance deterioration. That is, the operation data of the gas turbine usually fluctuates according to changes in various operation conditions such as the partial load factor and the outside air temperature. This is because the operation data may exceed the threshold.

[0005] Therefore, in order to detect the performance degradation of the gas turbine by comparing the operation data with such a large fluctuation with the threshold value, it is necessary to set a threshold value with low sensitivity in consideration of the fluctuation width. However, in this case, the result is that the performance degradation cannot be accurately detected.

In connection with this point, Japanese Patent Application Laid-Open No. H11-1551
No. 6 generates standardized data under common operating conditions from which influences due to differences in various operating conditions are excluded from the operating data (the specific generating method is not shown), and the standardized data has a threshold. We propose a method to detect performance degradation by setting values. However,
Gas turbine operation data (so-called process data)
Typically includes a periodic error, and due to various factors such as the response delay of the system and the response delay of the sensor to changes in the partial load factor and ambient conditions, a considerable amount of data is actually included in the standardized data. Usually, a range of fluctuation is observed. For this reason, it is still difficult to accurately detect the performance deterioration of the gas turbine simply by comparing the standardized data with the threshold.

As described above, there is no system capable of automatically detecting the performance deterioration and failure occurrence of the gas turbine with high accuracy in actuality. Therefore, the operating state of the gas turbine depends on the knowledge and know-how of a skilled engineer. The fact is that the diagnosis is made.

[0008]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems in the prior art, and is directed to a method for diagnosing a gas turbine operation state capable of detecting deterioration of performance and occurrence of a failure of a gas turbine with high accuracy. It is an object to provide a diagnostic device.

[0009]

According to the gas turbine operating condition diagnosis method of the present invention, the operating data of each part of the gas turbine sampled at predetermined time intervals is standardized by a predetermined process to create standardized data. And moving average processing of the created standardized data to create diagnostic data, and diagnosing the operating state of the gas turbine based on the created diagnostic data.

The method for diagnosing the operating condition of a gas turbine according to the present invention is, specifically, a procedure for standardizing operation data of each part of a gas turbine sampled at a preset time interval by a predetermined process to create standardized data. And a procedure for moving average processing the created standardized data to create diagnostic data, and a procedure for diagnosing the operating state of the gas turbine based on the created diagnostic data. And

In the gas turbine operating state diagnosis method of the present invention, a moving average time for performing a moving average process on the standardized data using the operating data in the initial state of the gas turbine may be set.

Further, in the gas turbine operating state diagnosis method of the present invention, the moving average time is set based on a settling time until a change in operating data due to a change in various operating conditions of the gas turbine is settled. Is also good.

Further, in the method for diagnosing the operating state of a gas turbine according to the present invention, the moving average time is set for each operation data of each part to be detected, and the moving average time is set to be the longest among the settling times for each operating condition. May be done.

Further, in the gas turbine operating state diagnosis method of the present invention, the moving average time set based on the settling time is corrected so that the average result based on the moving average time falls within a predetermined fluctuation range. Is also good.

Further, in the gas turbine operating condition diagnosis method of the present invention, the standardization of the operating data may be performed using a preset characteristic curve or characteristic function.

On the other hand, the gas turbine operating condition diagnostic apparatus of the present invention comprises an operating data sampling means for sampling operating data of each part of the gas turbine detected by the detecting unit at predetermined time intervals, and the operating data sampling means. Standardization means for standardizing the operation data sampled by a predetermined process to create standardized data, moving average processing means for moving average processing of the standardized data created by the standardization means to create diagnostic data, Operating state diagnostic means for diagnosing the operating state of the gas turbine based on the created diagnostic data.

Further, the gas turbine operating state diagnostic apparatus of the present invention includes a moving average time setting means, and the moving average time setting means comprises a moving average time for moving average processing of the standardized data by the moving average processing means. May be set using operation data in the initial state of the gas turbine.

Further, in the gas turbine operating condition diagnostic apparatus according to the present invention, the moving average time setting means sets the moving average time setting means based on a settling time until a change in operating data due to a change in various operating conditions of the gas turbine settles. The moving average time may be set.

Further, in the gas turbine operating condition diagnostic apparatus of the present invention, the moving average time setting means sets a moving average time for each operation data of each section to be detected, and the moving average time is set for each operating condition. May be the longest in the settling time.

Further, in the gas turbine operating condition diagnostic apparatus according to the present invention, the moving average time setting means determines the moving average time set based on the static time of the gas turbine as a predetermined result based on the moving average time. The correction may be made to be within the fluctuation range.

Further, in the gas turbine operating condition diagnostic apparatus of the present invention, the moving average processing means may hold the set moving average time.

Further, in the gas turbine operating condition diagnostic apparatus of the present invention, the standardizing means may have a characteristic curve / characteristic function setting section.

Thus, the gas turbine operating condition diagnostic apparatus of the present invention is provided in a gas turbine.

Further, the gas turbine and the gas turbine operating condition diagnostic apparatus of the present invention may be connected via communication means, and a plurality of gas turbines may be provided.

Furthermore, the operation data sampling means of the gas turbine operating state diagnostic device may be connected to other elements via communication means.

[0026]

Since the present invention is configured as described above, it is possible to reduce the influence of fluctuations in operating conditions, time delay until the process system reaches an equilibrium state following the fluctuation, and effects of periodic and mechanical error factors. Excluded from the operation data, the operation state of the gas turbine can be diagnosed with high accuracy.

[0027]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described based on embodiments with reference to the accompanying drawings, but the present invention is not limited to only such embodiments.

Embodiment 1 FIG. 1 shows a schematic configuration of a gas turbine operating condition diagnostic device (hereinafter simply referred to as a diagnostic device) according to Embodiment 1 of the present invention. The operation data (hereinafter, referred to as operation data) of each part of the gas turbine G, for example, output, thermal efficiency, exhaust temperature, compressor outlet temperature, outlet pressure, CO, NOx, intake flow rate, etc. are detected and sampled. It is assumed that the performance of the gas turbine G is deteriorated and whether or not a failure has occurred is diagnosed based on the operation data of the respective parts.

That is, the diagnostic apparatus A can use the operation data sampling means 11 for sampling the operation data of each part at a predetermined time interval, and use the operation data of each part sampled by the sampling means 11 as data for operating state diagnosis. As described above, the standardizing means 12 for converting the data into standardized data (hereinafter, referred to as “parts standardized data”) under the common operating conditions, and the moving average processing means 1 for performing the moving average processing for the respective parts standardized data.
3 and the settling time until the fluctuation of each part operation data due to the change of the operation condition of the gas turbine G is stabilized.
Moving average time setting means 14 for setting a moving average time for each of the standardized data of each part for performing a moving average process on the standardized data of each part using operation data in an initial state of the gas turbine G (hereinafter referred to as initial operation data of each part); , Moving average processed standardized data (hereinafter,
Operating state diagnostic means 15 for diagnosing the operating state of the gas turbine G using the moving average data of each part).

The detection unit 10 operates data of each part of the gas turbine, for example, the output OP of the gas turbine G, the thermal efficiency H
E, an exhaust gas temperature Te, a compressor outlet temperature Tc, a compressor outlet pressure Pc, CO, NOx, a concentration FP, and various sensors for detecting an intake flow rate SF.

The operation data sampling means 11 outputs operation data of each part of the gas turbine from the detection unit 10, for example,
Gas turbine G output OP, thermal efficiency HE, exhaust temperature T
e, compressor outlet temperature Tc, compressor outlet pressure Pc, CO,
It has a sampling unit 11a for sampling the NOx concentration FP and the intake flow rate SF, and a sampling cycle holding unit 11b for holding a preset sampling cycle in order to perform sampling by the sampling unit 11a at predetermined time intervals.

The standardizing means 12 has a characteristic curve / characteristic function setting section 12a in which a characteristic curve or characteristic function used for standardization is set in advance. That is, the standardization unit 12 converts the operation data of each part sampled by the operation data sampling unit 11 into standardized data using the characteristic curve set in the characteristic curve / characteristic function setting unit 12a.
Here, the characteristic curve refers to various operating conditions of the gas turbine G,
That is, the atmospheric pressure Pa, the atmospheric temperature Ta, and the atmospheric humidity H
The ambient conditions such as “a” and changes in operation parameters such as the guide vane angle Vd, the intake pressure loss PIDp, the exhaust pressure loss PEDp, the steam injection amount GS, and the water injection amount GW are reflected in the operation data of each part, for example, the output OP of the gas turbine G. The effect is analyzed and formalized, and standardization is performed using a standardization coefficient calculated based on this characteristic curve.

The characteristic curve or characteristic function is, specifically,
It is created by statistically processing data obtained by a variation test performed in a bench test of the gas turbine G.
Here, the variation test is a test for monitoring the gas turbine G while variously changing various operating conditions of the gas turbine G and grasping the performance of the gas turbine G.

Tables 1 and 2 show which operating conditions should be used when converting the operating data of each part into the standardized data of each part. Here, in the table, the single circle is
It indicates that it is necessary to use the characteristic curve (characteristic function) under the operating conditions when converting the operation data of each part to the standardized data of each part, and the horizontal line indicates that when converting the operation data of each part to the standardized data of each part. It shows that the need to use characteristic curves for operating conditions is low.

[0035]

[Table 1]

[0036]

[Table 2]

The moving average processing means 13 has a moving average time holding unit 13a for holding the moving average time set by the moving average time setting means 14, and the moving average time holding unit 1
Based on the moving average time held in 3a, each part standardized data is subjected to moving average processing to generate each part moving average data as diagnostic data. Here, this moving average processing is to take the average of the standardized data of each part within the moving average time.

The moving average time setting means 14 includes a static time setting unit 14a for holding a static time Ts until a variation in operation data of each part due to a change in operating conditions of the gas turbine G is stabilized as a static time table. When setting each moving average time for moving average processing of each part standardized data based on this settling time table, each part standardized data obtained by standardizing each part initial operation data (hereinafter referred to as each part initial standardized data) And a moving average time correction unit 14b that corrects the moving average time so that the fluctuation range of the data falls within the predetermined fluctuation range.

Tables 3 and 4 show that the static settling time holding unit 14a
Shows an example of the settling time table held in FIG. The unit of the numerical values in the table is seconds.

[0040]

[Table 3]

[0041]

[Table 4]

The measurement of the settling time Ts is performed, for example, as follows.

The settling time Ts of the fuel flow rate GL is derived using the measurement results of the fuel flow rate and the output, and the time until a change in the fuel flow rate is detected as an output change is set as the settling time Ts. Is done.

The derivation of the settling time Ts of the guide vane angle Vd is performed using the measurement result of the opening degree of the guide vane.
The time required for the combustion air to reach the turbine from the gas turbine inlet is set as the settling time Ts.

Table 5 summarizes this measuring method.

[0046]

[Table 5]

The operating state diagnosing means 15 includes the standardizing means 12
Diagnostic data generated by the moving average processing in the moving average processing means 13 (hereinafter, referred to as diagnostic data).
The operating state of the gas turbine G is diagnosed based on the moving average data of each part). Here, as shown in Table 6, for example, the presence or absence of a failure in the entire gas turbine G is diagnosed based on the output OP measured from the generator output, and the overall efficiency of the gas turbine G is determined based on the thermal efficiency HE measured from the fuel flow rate and the output. An appropriate operation data is selected depending on the item to be diagnosed, for example, the presence or absence of a failure is diagnosed.

[0048]

[Table 6]

Table 7 shows the diagnostic object adaptability of the operation data of each part. Here, a double circle indicates that the operation state of the relevant component of the gas turbine G can be satisfactorily diagnosed using the operation data, a single circle indicates that the diagnostic turbine has a medium diagnostic adaptability, and a triangle indicates the diagnosis. Indicates poor adaptability.

[0050]

[Table 7]

Hereinafter, the flow of processing executed by the diagnostic device A will be described with reference to FIG. Note that S1 to S1 in FIG.
Step numbers up to S6 are shown.

In step S1, the operation data sampling means 11 samples the operation data of each part by the sampling unit 11a at predetermined time intervals held in the sampling period holding unit 11b. Here, the operation data of each unit to be sampled is appropriately selected according to the diagnosis contents shown in Table 6.

In step S2, the standardization means 12 standardizes the operation data of each part sampled by the operation data sampling means 11. That is, standardized data of each part is generated.

In step S3, the moving average processing means 13
As a result, the moving average processing is performed on the standardized data of each unit generated by the standardizing means 12 using the moving average time held in the moving average time holding unit 13a.
The moving average time held in the moving average time holding unit 13a is set in a moving average time setting process described later in detail.

In step S4, the operating state diagnostic means 15
Thereby, the moving average data of each part as the diagnostic data is compared with the threshold value thereof, and the operating state of the gas turbine G is diagnosed.

In step S5, the result of diagnosis in step S4 is output.

Next, the moving average time setting processing will be described.

FIG. 3 shows an execution procedure of the moving average time setting process. Step numbers S11 to S15 in FIG. 3 indicate step numbers.

In step S11, the operation data sampling means 11 samples the operation data of each part in the initial state of the gas turbine G (initial operation data of each part).

In step S12, the standardization means 12 standardizes the initial operation data of each section, and generates the initial standardized data of each section.

In step S13, the moving average time setting means 14 temporarily sets the moving average time for each of the unit initial standardized data based on the settling time table held in the settling time holding unit 14a. Specifically, the longest settling time among the settling times for each of the operating conditions in Tables 3 and 4 is provisionally set as the moving average time for performing the moving average processing on each set of operating data.

For example, for the output OP, the moving average time is temporarily set to 300 seconds. The heat efficiency HE is 300 seconds, the exhaust temperature ET is 350 seconds, the compressor outlet temperature CT is 200 seconds, and the compressor outlet pressure C
200 seconds for P, 300 seconds for CO and NOx concentrations FP, and 50 seconds for the intake air flow rate SF are provisionally set as the moving average time. Note that the moving average time is not limited to the above, but is appropriately determined by the scale of the gas turbine G, the response times of various sensors, and the time constant of the control system.

In step S14, a moving average process of the standardized data of each unit is executed based on the moving average time provisionally set in step S13.

In step S15, the moving average time correction section 14b determines whether or not the average result in step S14 falls within a predetermined fluctuation range.

More specifically, the standardized data of each section is subjected to a moving average process based on the temporarily set moving average time, and the fluctuation range of the processing result within a predetermined time is within the data fluctuation range as shown in Table 8. It is determined whether there is. For example, in the case of the output OP, the moving average processing is executed based on the moving average time provisionally set for the output OP, and it is determined whether or not the fluctuation range of the calculated moving average result is within 5% of the rated output. .

Here, Table 8 shows that the reduction gear speed is 3600.
rpm, rotation speed 14000 rpm, power generation frequency 60
3 shows a relationship among a threshold value, an allowable error, and the data fluctuation width set for each part of operation data for an operating state diagnosis in an industrial medium-sized gas turbine of Hz.

In step S15, if the moving average result in step S14 does not fall within the data fluctuation range shown in Table 8, it is determined that an inappropriate moving average time has been temporarily set, and the process returns to step S13. The moving average time is corrected by temporarily setting the moving average time again. The correction of the moving average time is performed by selecting a moving average time that minimizes the influence of disturbance factors (noise) other than the settling time.

On the other hand, if the data fluctuation width of each part moving average data is within the data fluctuation width shown in Table 8, it is determined that an appropriate moving average time has been temporarily set, and the flow advances to step S16.

[0069]

[Table 8]

In step S16, step S13
Is output to the moving average processing means 13 as an optimal moving average time, and is held in the moving average time holding unit 13a.

As described above, according to the diagnostic apparatus T of the first embodiment, the operation data of each section sampled for diagnosing the operation state of the gas turbine G is converted into the standardized data of each section under the common operation conditions. A moving average time for performing a moving average process on each of the standardized data of each part based on a stabilization time until a fluctuation of each part of the operation data due to a change in the operating condition of the gas turbine G is stabilized using the initial operation data. Is set, the influence of the fluctuation of the operating conditions is removed from the operation data of each part, and the performance deterioration and failure occurrence of the gas turbine G can be detected with high accuracy using the operation data of each part measured during the normal operation. .

Further, when setting the moving average time, the moving average time is corrected so that the electrical noise due to the rotation of the engine is minimized. The average time can be set.

Embodiment 2 FIG. 4 shows a schematic configuration of a gas turbine operation state diagnosis system B according to Embodiment 2 of the present invention. The gas turbine operation state diagnosis system B shown in FIG. The gas turbine G
The operation data from the detection unit 10 provided in the diagnostic unit A of the first embodiment is transmitted through the communication unit 16 including an Internet communication network, a telephone line, a mobile phone, and the like. The gas turbine operation state diagnosis system B shown in FIG. 2B includes the operation data sampling unit 11 of the diagnosis device A of the first embodiment near the gas turbine G, and the operation data sampling unit 11 and the diagnosis device. The other elements constituting A, that is, the standardizing means 12, the moving average processing means 13, the moving average time setting means 14, and the operating state diagnostic means 15 (hereinafter, referred to as other elements of the diagnostic device A) are connected to the Internet communication network. , A telephone line, a mobile phone, and the like.
The remaining configuration is the same as in the first embodiment.

According to the gas turbine operation state diagnosis system B having such a configuration, the operation state can be diagnosed at a place remote from the gas turbine G.

As described above, according to the gas turbine operation state diagnosis system B of the second embodiment, the operation state of the gas turbine G and the performance deterioration / failure can be remotely monitored.

[0076]

Hereinafter, the present invention will be described more specifically with reference to more specific examples.

FIG. 5 shows changes over time in the operation data of each part of the gas turbine, specifically, the operation data (output OP) in which the standardization processing and the moving average processing have not been performed (Comparative Example 1). As shown in the figure, it is clear that the operating data has a large short-term fluctuation range, and it is not possible to accurately diagnose the operating state by simply comparing such operating data with a threshold value.

FIG. 6 shows how the standardized data of each part of the gas turbine, specifically, the operation data (output OP) subjected to the standardization process changes over time (Comparative Example 2). As shown in FIG. 5, although the standardized data has a smaller data fluctuation range than the detected data shown in FIG. Is below the threshold from around 2000 hours. Thus, it can be seen that simply comparing the standardized data with the threshold value makes it difficult to accurately detect the performance deterioration of the gas turbine.

FIG. 7 shows an example in which the diagnostic device A actually processes the operation data of each part of the gas turbine, specifically, the measurement result of the output OP (embodiment). As shown in FIG. 7, the moving average data subjected to the moving average processing by the diagnostic apparatus A does not become a dispersive value, and therefore, it is possible to sufficiently detect the performance deterioration or the occurrence of a failure by comparing with the threshold value. Data with high stability.

Further, there is a time when the output OP value falls below the threshold value when the operation time exceeds 5000 hours. As a result of inspection of the gas turbine, a problem such as clogging of the intake filter occurs. It was confirmed that. Therefore, it can be said that it can be confirmed that the operation state of the gas turbine can be diagnosed with high accuracy only by comparing the measured operation data of each part with the threshold value corresponding thereto.

As described above, the present invention has been described based on the embodiments and the examples. However, the present invention is not limited to the embodiments and the examples, and various modifications are possible. For example, in the embodiments and examples, a gas turbine is described as an example, but the application of the present invention is not limited to a gas turbine, and can be applied to various process devices. Can also be applied. Although the moving average time setting process is performed in the initial state, the process is not limited to the initial state, and may be performed at a time when the performance of the gas turbine G is guaranteed to be normal. For example, a bench test may be performed. A plurality of gas turbines may be diagnosed by one diagnostic device.

[0082]

As described in detail above, according to the present invention,
Accurate diagnosis of gas turbine operating conditions by eliminating from operating data fluctuations in operating conditions, time delays until the process system reaches an equilibrium state, and the effects of periodic and mechanical error factors. The excellent effect that it can be obtained is obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of a gas turbine operating state diagnostic device according to a first embodiment of the present invention.

FIG. 2 is a flowchart schematically showing a procedure of a diagnosis process executed by the gas turbine operation state diagnosis device.

FIG. 3 is a flowchart illustrating a flow of a moving average time setting process.

FIG. 4 is a block diagram illustrating a schematic configuration of a gas turbine operating state diagnosis system according to a second embodiment of the present invention;
(A) shows an example in which a gas turbine and a diagnostic device are connected via communication means, and (b) shows an example in which an operation data sampling means and other elements are connected via communication means. Is shown.

FIG. 5 is a graph showing a detection result of process amount data.

FIG. 6 is a graph showing a processing result of process amount data according to the related art.

FIG. 7 is a graph showing a processing result of process amount data according to an embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Detecting part 11 Operating data sampling means 11a Sampling part 11b Sampling period holding part 12 Standardization means 12a Characteristic curve / characteristic function setting part 13 Moving average processing means 13a Moving average time holding part 14 Moving average time setting means 14a Static time holding part 14b Moving average time correction unit 15 Operating state diagnostic means 16 Communication means A Diagnostic device B Gas turbine operating state diagnostic system G Gas turbine

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Eiichi Harada 1-1-1, Kawasaki-cho, Akashi-shi Kawasaki Heavy Industries, Ltd. Inside the Akashi Plant (72) Inventor Kazunori Sato 1-1-1, Kawasaki-cho, Akashi-shi Kawasaki Heavy Industries, Ltd. Inside the Akashi factory (72) Inventor Yoshihiko Ozaki 1-1, Kawasaki-cho, Akashi Kawasaki Heavy Industries, Ltd. F-term in the Akashi factory (reference) 5H223 AA02 BB01 DD07 EE06

Claims (18)

[Claims] An operation data of each part of a gas turbine sampled at a preset time interval is standardized by a predetermined process to generate standardized data, and the generated standardized data is subjected to a moving average process to perform diagnosis. A method for diagnosing a gas turbine operating condition, comprising creating data and diagnosing an operating condition of a gas turbine based on the created diagnostic data. 2. A procedure for standardizing operation data of each part of a gas turbine sampled at a preset time interval by a predetermined process to generate standardized data, and performing a moving average process on the generated standardized data. A method for diagnosing an operating state of a gas turbine, comprising: a step of creating diagnostic data; and a step of diagnosing an operating state of a gas turbine based on the created diagnostic data. 3. The method for diagnosing a gas turbine operation state according to claim 1, wherein a moving average time for moving average processing of the standardized data is set using operation data in an initial state of the gas turbine. . 4. The moving average time is set based on a settling time until a change in operation data due to a change in various operating conditions of the gas turbine is settled.
The method for diagnosing the operating state of a gas turbine according to claim 1. 5. The gas according to claim 4, wherein the moving average time is set for each operation data of each part to be detected, and is set to be the longest in the settling time for each operation condition. Turbine operating state diagnosis method. 6. The gas turbine operation according to claim 4, wherein the moving average time set based on the settling time is corrected so that the average result based on the moving average time falls within a predetermined fluctuation range. Condition diagnosis method. 7. The method according to claim 1, wherein the operation data is standardized using a preset characteristic curve or characteristic function. 8. An operation data sampling means for sampling operation data of each part of the gas turbine detected by the detection unit at predetermined time intervals, and operating data sampled by the operation data sampling means by a predetermined process. Standardizing means for standardizing to create standardized data; moving average processing means for moving average processing of the standardized data created by the standardizing means to create diagnostic data; gas based on the created diagnostic data; A gas turbine operating state diagnostic device, comprising: operating state diagnostic means for diagnosing an operating state of a turbine. 9. A moving average time setting means, wherein the moving average time setting means uses a moving average time for the moving average processing means to perform a moving average processing of the standardized data by using operation data in an initial state of the gas turbine. The gas turbine operating state diagnostic apparatus according to claim 8, wherein the apparatus is set. 10. The moving average time setting means sets a moving average time based on a settling time until a change in operation data due to a change in various operating conditions of the gas turbine is settled. 10. The gas turbine operating state diagnostic device according to claim 9. 11. The moving average time setting means sets a moving average time for each operation data to be detected, and sets the moving average time to be the longest of the settling times for each operation condition. The gas turbine operating state diagnostic device according to claim 10, wherein: 12. The moving average time setting means corrects a moving average time set based on a settling time of the gas turbine so that an average result based on the moving average time falls within a predetermined fluctuation range. The gas turbine operating state diagnostic device according to claim 10. 13. The gas turbine operating state diagnostic apparatus according to claim 8, wherein the moving average processing means holds the set moving average time. 14. The apparatus according to claim 8, wherein the standardizing means includes a characteristic curve / characteristic function setting unit. 15. A gas turbine comprising the gas turbine operating condition diagnostic device according to claim 8. 16. A system for diagnosing operating conditions of a gas turbine, wherein the gas turbine and the device for diagnosing operating conditions of a gas turbine according to claim 8 are connected via communication means. 17. The system according to claim 16, wherein a plurality of gas turbines are provided. 18. A gas turbine operating state characterized in that the operating data sampling means of the gas turbine operating state diagnostic device according to claim 8 and another element are connected via communication means. Diagnostic system.