JPH05141266A - Diagnostic device for trouble in fuel supply system - Google Patents

Abstract

PURPOSE:To accurately diagnose abnormality, when it is generated in the actually adjusting valve opening including even the case that an opening meter is in trouble, by diagnosing whether the abnormality is found or not in action of a fuel valve from a mutual relation of a generator output, turbine outlet temperature and a rotational speed. CONSTITUTION:A temperature gage 10 for detecting a turbine outlet temperature, wattmeter 12 and a rotational speedometer 65 are provided in a gas turbine electric power plant, and output signals of these gage and meters are input to a diagnostic device 14. Here, a preliminary value Tr of normal operational temperature of the turbine outlet temperature is calculated from a generator output P, and a degree of shifting the actual outlet temperature T relating to this reference value Tr is represented by the first membership function M1. A degree of shifting a rotational speed (n) from a control target value nr is represented by the second membership function M2. A membership value M3 of representing a degree of generating a phenomenon, in which a fuel adjusting valve 64 is in trouble to shift the opening from a required value, is obtained by a minimum value of M1, M2 to diagnose trouble generation, when M3 is approximate to 1, and normal when M3 is approximate to 0.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for diagnosing a fuel valve failure in a gas turbine power generator.

[0002]

2. Description of the Related Art FIG. 5 is a system diagram showing an example of a gas turbine power plant to be diagnosed by a failure diagnosis device of the present invention. The compressor 1, turbine 8 and generator 4 are rotating shafts 3.
Are connected by and rotate integrally. The compressor 1 sucks in air, compresses it to a high pressure, discharges it, and then discharges it to the combustor 5 through the pipe 2. Fuel is injected into the combustor 5, and the fuel is mixed with air and burned to generate high-temperature and high-pressure combustion gas. The combustion gas flows into the turbine 8 through the pipe 7 and performs work while passing through the inside of the turbine while being expanded, thereby rotating the turbine 8. This rotation causes the generator 4 to rotate and generate electric power.

The turbine speed is kept constant by adjusting the fuel injection quantity to adjust the heat drop or work in the turbine. The control principle is as follows. First, the fuel is discharged from the pump 62 and the pipe 63
Through to the combustor 5. A fuel control valve 64 is installed in the pipe 63. The rotation speed n is measured by the rotation speed meter 65 and transmitted to the controller 68 through the transmission line 66. The controller 68 calculates the control valve opening φ _G, which is the operation amount for the negative feedback control by the well-known proportional-plus-integral operation, based on the following arithmetic expression [Equation 1], and the fuel control valve through the transmission line 67. Tell 64. The fuel control valve 64 is operated by a well-known servo mechanism or the like so that the opening becomes φ _G, and the opening φ
_The fuel flow rate is adjusted according to _G.

[0004]

[Equation 1]

In the gas turbine configured as described above, there is a possibility that the fuel control valve 64 does not operate according to the opening degree demanded by the controller 68, causing a failure.
Therefore, conventionally, a diagnostic method such as comparing an actual opening of the fuel control valve 64 with an opening meter and a required opening value and determining an abnormality if they do not match is used. It was being done.

[0006]

In the conventional failure diagnosis system, when a failure occurs in the opening degree meter of the fuel control valve, the valve opening degree is controlled based on the erroneous opening degree signal. However, even if the actual valve opening degree reaches a value different from the control required value, there is a problem that it cannot be diagnosed as an abnormality.

An object of the present invention is to diagnose when the actual control valve opening becomes abnormal, including the case where the opening meter fails as described above.

[0008]

In order to solve the above-mentioned problems, the present invention is directed to a gas turbine power generator in which an output detector for detecting the output of the generator and a normal operation based on the detected value of the output detector are used. A calculator for calculating the reference value of the operating temperature, a temperature detector for detecting the turbine outlet temperature, a rotation speed detector for detecting the rotation speed of the turbine shaft, a deviation between the reference temperature and the detection temperature, and a control target. A minimum value selector that matches the deviation between the rotation speed and the detected rotation speed and selects the smaller value is provided, and the presence or absence of a failure in the fuel supply system is diagnosed based on the selected minimum value. The present invention proposes a fuel supply system failure diagnosis device characterized by the above.

[0009]

Since the present invention has the above-mentioned configuration, instead of monitoring the valve opening signal as in the conventional case, the presence or absence of abnormality in the valve operation is diagnosed from the interrelationship between the generator output, the turbine outlet temperature, and the rotation speed. That is, the generator output P calculated reference value T _r of the normal operating temperature of the turbine outlet temperature, the reference value T _r
The degree of deviation of the actual turbine outlet temperature T from the first
The membership function M ₁ of
Represents the degree of deviation from the control target value n _r by the second membership function M ₂ . And the fuel control valve failed,
The membership value M ₃ representing a degree of occurrence of events opening deviates from the required value, determined by the minimum value of the M ₁ and M _2, a diagnosis of failure when M ₃ is close to 1, the M ₃ 0
It is diagnosed as normal when it is close to. In this way, the failure of the fuel control valve is diagnosed without monitoring the opening degree meter of the fuel control valve.

[0010]

FIG. 1 is a system diagram showing an example of a gas turbine power plant to which an embodiment of the present invention is applied, and FIG. 2 is a diagram showing a concrete arithmetic circuit.

The failure diagnosis apparatus of this embodiment is arranged in a gas turbine power plant as shown in FIG. The turbine outlet temperature T is measured by the thermometer 10, and its signal is transmitted to the diagnostic device 14 through the transmission line 11. On the other hand, the generator output P is measured by the power meter 12, and its signal is transmitted through the transmission line 13 to the diagnostic device 14 described above.
Is transmitted to. The rotation speed n is transmitted from the rotation speed meter 65 to the diagnostic device 14 through a transmission line 69.

Next, a concrete embodiment of the diagnostic device 14 is shown in FIG.
Will be described. The generator output P is input to the function unit 17 via the transmission line 13. The function unit 17 is provided with the relationship between the generator output P and the reference outlet temperature T _r in the normal operating state of the turbine, and outputs T _r to the input of P, which is then transmitted to the transmission line 18 Is transmitted to the subtractor 19. On the other hand, the turbine outlet temperature measured by the thermometer 10 is transmitted to the subtractor 19 through the transmission line 11. In the subtractor 19, the deviation of T from T _r is calculated by the following equation and transmitted to the function unit 21 through the transmission line 20.

[0013]

[Equation 2]

In the function unit 21, the membership value M ₁ representing the degree of abnormality of ΔT is calculated by the membership function according to the following formula [Equation 3], and the M ₁ value is transmitted to the minimum value selector 25 through the transmission line 22. ..

[0015]

[Equation 3]

On the other hand, the rotation speed n is controlled by the operation of the controller 68 so as to maintain a constant rotation speed n _r .
Therefore, the membership value M, which indicates the degree of abnormality in the rotation speed,
₂ can be calculated by the membership function according to the following equation [Equation 4].

[0017]

[Equation 4]

The value of M ₂ is transmitted to the minimum value selector 25 through the transmission line 24. In the minimum value selector 25, M
Compare ₁ with M ₂ and select the smaller value as M ₃ . That is,

[0019]

[Equation 5]

It is well known fuzzy that M ₃ is a membership value indicating the certainty of the event X in which the turbine outlet temperature T is an abnormal value and the rotation speed n is also an abnormal value. It is in line with logic. And it is clear from the following cause-effect analysis that the event X is caused by the failure of the fuel control valve. That is, if the fuel control valve fails, the fuel supply amount becomes either too large or too small. Then, either the turbine outlet temperature is too high and the rotation speed is too high, or the turbine outlet temperature is too low and the rotation speed is too low. Here, it is assumed that the control system failure is detected by self-check outside the scope of the present invention. The above causal relationship is the phenomenon that the turbine outlet temperature decreases and the rotational speed transiently increases when the generator load decreases, or the turbine outlet temperature increases and the rotational speed increases when the generator load increases. Is a characteristic symptom that can be distinguished from a phenomenon in which is decreased transiently.

Therefore, the occurrence of the event X and the failure of the fuel control valve can be associated one to one. Therefore, the membership value M ₃ that represents the certainty that the event X occurred
Can also be considered as the likelihood that a fuel regulator valve failure has occurred. The value of M ₃ takes a value in the range of 0 to 1, with 0 being normal and 1 being a failure. In the middle of the range, the closer the value is to 1, the greater the probability that a failure has occurred. Therefore, by transmitting the value of M _{3 to} the display 16 through the transmission line 15 and displaying it, a useful diagnosis can be performed.

Here, two membership values M₁, M₂
Of the expressions [Equation 3] and [Equation 4], and the occurrence of failure
Membership value M that indicates the certainty of₃As M₁
And M ₂More on the reasons for choosing the minimum value of
Reveal

First, the equation [Formula 3] for calculating the M ₁ value is shown in the graph of FIG. In order to obtain this figure, the shape of the membership function M ₁ is defined by the formula [Equation 3]. The meaning is as follows.

When the turbine is operating normally, the turbine outlet temperature T is equal to the reference temperature T _r , and ΔT = T−
T _r = 0. At this time, M ₁ = 0.

However, if the turbine is abnormal, the turbine outlet temperature deviates from the reference temperature, and ΔT is not zero.

When ΔT> a or ΔT <b, it can be clearly understood by the characteristic analysis that there is an abnormality. At this time, since the membership value indicating the degree of abnormality of ΔT is 1, it is defined as M ₁ = 1.

When b ≦ ΔT ≦ 0 or 0 <ΔT ≦ a, it is between the normal and the obvious abnormality, and the degree of abnormality is expressed by a value of 0 to 1 in proportion to the absolute value of ΔT. , The functions of ΔT / b and ΔT / a are used respectively.

Next, the equation [Formula 4] for calculating the M ₂ value is shown in FIG. To obtain this figure, the shape of the membership function M ₂ is defined by the formula [Equation 4]. The meaning is as follows.

When the turbine is in a normal operating condition, the rotational speed n is equal to the reference rotational speed n _r , and n = n _r . At this time, M ₂ = 0.

However, if the turbine is abnormal, the turbine speed deviates from the reference value n _r .

When n> c or n <d, it can be understood by characteristic analysis that the abnormality is obvious. Then n
Since the membership value indicating the degree of abnormality of is 1,
Define M ₂ = 1.

When n _r <n ≤ c or d ≤ n ≤ n _r , it is between normal and obvious abnormality, and the degree of abnormality-likeness is 0 in proportion to the absolute value of (n-n _r ). The functions (n-n _r ) / (c-n _r ) and (n-n _r ) / (d-n _r ) are used to represent the values ˜1 respectively.

As mentioned above, M₁Is the turbine outlet temperature
The degree of unusualness as seen from the above is expressed by a numerical value of 0 to 1.
And 0 is normal and 1 is obvious abnormality.
It Also M₂Is the anomaly of the turbine speed
The degree is expressed by a numerical value of 0 to 1, and this is also
A value of 0 indicates normal, and a value of 1 indicates a clear abnormality. And fuel
When a control valve failure occurs, the symptom is the turbine outlet temperature.
Both the degree and the number of rotations deviate from the reference value, indicating an abnormal value.
This is as described above. Therefore, M₁Is 0
M without₂If is 0, the symptom requirement of fuel control valve failure
Lacking, M again ₂M is not 0₁If is 0, then
It will lack beaming requirements.

Therefore, the smaller value of M ₁ and M _{2 is} M _3.
As can be seen, the M ₃ value is not 0 only when both M ₁ and M ₂ satisfy the requirement, and otherwise the M ₃ value remains a small value (minimum value). From this property, it is useful to define the M ₃ value as a membership value that represents the likelihood of a fuel control valve failure.

[0035]

According to the present invention, instead of monitoring the fuel control valve opening signal, the generator output, turbine outlet temperature,
From the mutual relation of the number of rotations, it is possible to diagnose the presence or absence of valve failure simply by processing the signal of the existing sensor. In this way, even in a plant that does not use a valve opening meter, a failure of the fuel control valve can be diagnosed.

Further, even if the valve opening becomes abnormal due to a failure of the valve opening meter, this can be diagnosed. It should be noted that the failure of the valve opening meter cannot be diagnosed simply by monitoring the opening measurement value. However, including this case, according to the failure diagnosis device of the present invention, an abnormality of the valve itself is detected. Diagnosis is possible, and the range of diagnoseable failures is greatly expanded.

[Brief description of drawings]

FIG. 1 is a system diagram showing a gas turbine power plant to which an embodiment of the present invention is applied.

FIG. 2 is a diagram illustrating a specific arithmetic circuit in the above embodiment.

FIG. 3 is a diagram showing a first membership function.

FIG. 4 is a diagram showing a second membership function.

FIG. 5 is a system diagram showing an example of a gas turbine power plant to be diagnosed by the failure diagnosis device of the present invention.

[Explanation of symbols]

 1 Compressor 3 Rotating shaft 4 Generator 5 Combustor 8 Turbine 10 Thermometer 12 Power meter 14 Diagnostic device 16 Display device 17,21,23 Function unit 19 Subtractor 25 Minimum value selector 62 Fuel pump 64 Fuel control valve 65 Rotation Number Total 68 Rotation speed controller

Claims (1)

[Claims] 1. In a gas turbine power generator, an output detector for detecting an output of a generator, an arithmetic unit for calculating a reference value of a normal operating temperature based on a detection value of the output detector, and a turbine outlet temperature. The temperature detector that detects the rotation speed, the rotation speed detector that detects the rotation speed of the turbine shaft, the deviation between the reference temperature and the detected temperature, and the deviation between the control target rotation speed and the detected rotation speed. And a minimum value selector for selecting, and the fuel supply system failure diagnosis device is configured to diagnose whether or not there is a failure in the fuel supply system based on the selected minimum value.