JP5653786B2 - Anomaly detection device - Google Patents

Description

  The present invention relates to an apparatus for detecting that some kind of abnormality has occurred in equipment.

  In various plants such as a gas turbine power plant, a nuclear power plant, or a chemical plant, the presence or absence of abnormality is monitored during operation of the plant. For example, Patent Document 1 discloses an operational index (operational) that represents the performance of a target machine, including the steps of collecting operational data from one or more machines and calculating an exceptional anomaly score from these operational data. A method for determining whether or not (metric) has an abnormal value is described. Patent Document 1 describes that an exceptional abnormality score (EAS) is calculated for technical data (for example, operation sensor data).

JP 2009-075081 A (0010, 0013)

  Although the technique described in Patent Document 1 calculates an exceptional abnormality score from operation sensor data, the operating state of devices that cannot obtain data directly with a sensor or the devices that are difficult to obtain data directly with a sensor It is difficult to detect the occurrence of abnormality in the operating state. This invention is made in view of the above, Comprising: It aims at detecting the abnormality of the driving | running state of the apparatus which a measurement is difficult or cannot measure with a sensor among the apparatuses which a gas turbine has.

  In order to solve the above-described problems and achieve the object, the present invention provides an operation of the gas turbine based on an operation condition of the gas turbine and a sensor output value output from a sensor that detects an operation state of the gas turbine. An abnormality of the gas turbine is detected based on a first estimation unit that obtains a first operation index indicating a state at a plurality of timings and a set of the first operation index estimated by the first estimation unit. The abnormality information calculation unit for obtaining the first abnormality detection information for comparing the abnormality detection information obtained by the abnormality information calculation unit with a predetermined threshold value to determine whether there is an abnormality in the gas turbine. And an abnormality detection unit for detecting the abnormality.

  This abnormality detection device obtains the first operation index by estimating it at a predetermined timing from a physical model of the gas turbine, and the abnormality detection information obtained by statistically processing the obtained set of first operation indices. Based on the above, the presence or absence of abnormality of the gas turbine is detected. Thus, since this abnormality detection apparatus calculates | requires a 1st operation parameter | index from the physical model etc. of a gas turbine, even if it is the operation state of the apparatus which the detection by a sensor is difficult or cannot detect, it detects an abnormality. Can do.

  As a desirable mode of the present invention, the first estimation unit may calculate the gas from the correspondence relationship between the first operation index and the sensor output value for detecting the operation condition of the gas turbine and the operation state of the gas turbine. It is preferable to obtain the first operation index during rated operation of the turbine. Thus, by converting the first operation index into a value at the time of rated operation, the first operation index obtained at different timings can be evaluated under the same standard as at the time of rated operation.

  As a desirable aspect of the present invention, from the abnormality detection information obtained by the abnormality information calculation unit and information that an abnormality has occurred in the gas turbine detected by the abnormality detection unit, the gas turbine in a predetermined period It is preferable to have an information processing unit that determines the frequency of occurrence of abnormality. In this way, it is possible to determine whether the abnormality occurring in the gas turbine is acute or chronic.

  As a desirable mode of the present invention, using the first operation index estimated by the first estimation unit, a second operation index indicating an operation state of the gas turbine different from the first operation index is obtained. A second estimation unit configured to perform estimation at a plurality of timings, wherein the abnormality information calculation unit corresponds to the set of the second driving index obtained by the second estimating unit and the second driving index; Based on the sensor output value output from the sensor that detects the operating state of the gas turbine, second abnormality detection information for detecting an abnormality of the gas turbine is obtained, and the abnormality detection unit calculates the abnormality information. It is preferable to detect the presence or absence of an abnormality of the gas turbine by comparing the second abnormality detection information obtained by the section with a predetermined threshold value set in advance. By doing in this way, the abnormality detection of the gas turbine by a more various method is realizable.

  The present invention can detect an abnormality in the operating state of devices that are difficult or cannot be measured by sensors among the devices that the gas turbine has.

FIG. 1 is a schematic diagram illustrating a configuration example of the abnormality detection apparatus according to the first embodiment. FIG. 2 is a control block diagram of the abnormality detection apparatus according to the first embodiment. FIG. 3 is a conceptual diagram illustrating an example of a correspondence relationship between the first operation index and the sensor output value of the gas turbine. FIG. 4 is a conceptual diagram illustrating an example of a correspondence relationship between the first operation index and the operation conditions of the gas turbine. FIG. 5 is a conceptual diagram illustrating an example of a correspondence relationship between the first operation index, the operation condition of the gas turbine, and the sensor output value of the gas turbine. FIG. 6 is an explanatory diagram of a technique for obtaining the first abnormality detection information. FIG. 7 is a schematic diagram illustrating a configuration example of the abnormality detection apparatus according to the second embodiment. FIG. 8 is a control block diagram of the abnormality detection apparatus according to the second embodiment. FIG. 9 is a conceptual diagram showing the frequency of occurrence of abnormality in the gas turbine. FIG. 10 is a schematic diagram illustrating a configuration example of the abnormality detection apparatus according to the third embodiment. FIG. 11 is a control block diagram of the abnormality detection apparatus according to the third embodiment.

  DESCRIPTION OF EMBODIMENTS Embodiments (embodiments) for carrying out the present invention will be described in detail with reference to the drawings. The present invention is not limited by the contents described in the following embodiments. The constituent elements described below include those that can be easily assumed by those skilled in the art and those that are substantially the same. Furthermore, the constituent elements described below can be appropriately combined. Although this embodiment demonstrates the example which applied this invention to the state monitoring of the power plant of a gas turbine, the object which can apply this invention is not limited to this. For example, the present invention can be applied to any plant that needs to detect an abnormal operation state of a plant that is difficult or impossible to measure, such as a nuclear power plant or a chemical plant.

(Embodiment 1)
FIG. 1 is a schematic diagram illustrating a configuration example of the abnormality detection apparatus according to the first embodiment. The abnormality detection device 10 monitors the state of the gas turbine 6 included in the gas turbine power plant 1 (particularly the state during operation) and detects an abnormality of the gas turbine 6. The gas turbine 6 to be monitored by the abnormality detection device 10 is a drive source for driving the generator 5 of the gas turbine power plant 1. The generator 5 is driven by the gas turbine 6 to generate electric power.

  The gas turbine 6 includes a compressor 2, a combustor 3, and a turbine 4 that rotates the compressor 2. The air drawn from the intake port of the compressor 2 is compressed by the compressor 2 and is led to the combustor 3 as high-temperature and high-pressure air. In the combustor 3, fuel is supplied to high-temperature and high-pressure air and burned. The fuel combusted in the combustor 3 is supplied to the turbine 4 as a high-temperature and high-pressure combustion gas, and rotates it. The output shaft of the gas turbine 6, that is, the rotating shafts of the turbine 4 and the compressor 2 are connected to the generator 5. With such a structure, the output obtained by operating the gas turbine 6 and rotating the turbine 4 is transmitted to the generator 5. With such a structure, the gas turbine 6 drives the generator 5 to cause the generator 5 to generate electric power.

  In the present embodiment, the gas turbine power plant 1 drives the generator 5 with the gas turbine 6. In addition to the gas turbine 6, the gas turbine power plant 1 drives the generator 5 with a steam turbine driven by the exhaust heat of the gas turbine 6. Also good. That is, the gas turbine power plant 1 may be a combined cycle plant.

  The abnormality detection device 10 monitors the state of the gas turbine 6. In the present embodiment, the abnormality detection device 10 monitors the state of one gas turbine 6, but may monitor the state of a plurality of gas turbines 6. The abnormality detection device 10 is, for example, a computer, and includes an input / output unit 11, a processing unit 12, and a storage unit 13. The abnormality detection apparatus 10 may use a so-called personal computer, or may combine a CPU (Central Processing Unit) and a memory.

  The gas turbine 6 has a plurality of sensors 7 as means for detecting its own operating state (operating state detecting means). Examples of the type of sensor 7 include a fuel temperature sensor that detects the temperature of fuel supplied to the gas turbine 6, a compressor outlet temperature sensor that detects the outlet temperature of the compressor 2, and the temperature of exhaust gas discharged from the turbine 4. There are an exhaust gas temperature sensor to detect, a rotation speed sensor to detect the rotation speed of the turbine 4, and the like. In the present embodiment, two sensors 7 are shown, but the type and number of sensors 7 are appropriately changed according to the gas turbine 6. In general, one gas turbine 6 has 50 to 60 sensors 7.

  The output of the sensor 7 (sensor output) is input to the input / output unit 11 of the abnormality detection device 10. The sensor output is sent to the input / output unit 11 of the abnormality detection device 10 in the form of an electrical signal. The processing unit 12 acquires the sensor output from the input / output unit 11 and detects an abnormality of the gas turbine 6 based on the value of the sensor output (sensor output value). The processing unit 12 is, for example, a CPU, and sequentially reads and interprets an instruction sequence called a program (computer program) existing on the storage unit 13, and moves or processes data according to the result.

  Note that the processing unit 12 may be realized by dedicated hardware. Further, a computer program for realizing the function of the processing unit 12 is recorded on a computer-readable recording medium, and the computer program recorded on the recording medium is read into the computer system and executed, thereby executing this embodiment. You may perform the process sequence of the plant state monitoring method which concerns. The “computer system” here includes an OS and hardware such as peripheral devices.

  The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a recording device such as a hard disk built in a computer system. Furthermore, “computer-readable recording medium” means that a computer program is dynamically held for a short time, like a communication line when a computer program is transmitted via a communication line such as the Internet or a telephone line. What is held, and what holds a computer program for a certain period of time, such as a volatile memory inside a computer system serving as a server or a client in that case. Further, the computer program may be for realizing a part of the above-described functions, and even if the functions described above can be realized in combination with a computer program already recorded in the computer system. Good.

  In the present embodiment, a method for detecting an abnormality in the gas turbine 6 (an abnormality detection method) can be realized by executing a computer program prepared in advance on a computer such as a personal computer or a workstation. This computer program can be distributed via a communication line such as the Internet. The computer program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. May be.

  The processing unit 12 includes a first estimation unit 12a, an abnormality information calculation unit 12b, and an abnormality detection unit 12c. The first estimation unit 12a is a first operation index indicating the operation state of the gas turbine 6 from the operation condition of the gas turbine 6 and the sensor output value output from the sensor 7 that detects the operation state of the gas turbine 6. Are obtained at a plurality of timings. The abnormality information calculation unit 12b obtains first abnormality detection information for detecting an abnormality of the gas turbine 6 based on the first set of operating indices estimated by the first estimation unit 12a. The abnormality detection unit 12c compares the abnormality detection information obtained by the abnormality information calculation unit 12b with a predetermined threshold value, and detects whether there is an abnormality in the gas turbine 6.

  In addition to the sensor 7, a control panel 14 that is an input / output unit is connected to the input / output unit 11 of the abnormality detection device 10. The control panel 14 is provided with a display 14D as a display means and an input means 14C for inputting a command to the abnormality detection device 10. The storage unit 13 of the abnormality detection apparatus 10 includes, for example, a volatile memory such as a RAM (Random Access Memory), a nonvolatile memory such as a ROM (Read Only Memory), a hard disk device, a magneto-optical disk device, a CD- A storage medium that can only be read, such as a ROM, or a combination of these can be configured.

  The storage unit 13 stores a computer program and data for realizing the abnormality detection method according to the present embodiment. The processing unit 12 implements the abnormality detection method according to the present embodiment using these computer programs and data. The storage unit 13 may be provided outside the abnormality detection device 10 so that the abnormality detection device 10 can access the storage unit 13 via a communication line. Next, a method in which the abnormality detection apparatus 10 detects an abnormality in the gas turbine 6 will be described.

  FIG. 2 is a control block diagram of the abnormality detection apparatus according to the first embodiment. The abnormality detection device 10 is useful for estimating and obtaining the operating state of the gas turbine 6 that is difficult or impossible to detect with the sensor 7. In order to realize this, as shown in FIG. 2, the abnormality detection apparatus 10 uses the sensor output value x from the sensor 7 and the operating condition y of the gas turbine 6 as input values of the first estimation unit 12a. The 1st estimation part 12a calculates | requires the 1st driving | operation parameter | index a which shows the driving | running state of a gas turbine from several sensor timings from the sensor output value x and the driving | running condition y.

  The sensor output value detects the operating state of the gas turbine 6. The sensor output value x includes the power generation amount of the gas turbine 6, the calorific value of the fuel supplied to the gas turbine 6, the flow rate, the temperature, the outlet temperature of the compressor 2 of the gas turbine 6, the outlet pressure, the intake air temperature, the gas turbine 6. Exhaust gas temperature, rotation speed, atmospheric pressure, fuel-air ratio, compressor 2 IGV (Inlet Guide Vane) angle, air bypass valve opening, combined cycle if combined cycle The amount of power generation, the flow rate of feed water to the steam turbine, etc. The operating condition y is an operation amount for the gas turbine 6, an outside air temperature, an outside air pressure, and the like.

  The first operation index a is an operation state of the gas turbine 6 that is difficult or cannot be detected by the sensor 7. The first operation index a is, for example, the inlet temperature (turbine inlet temperature) of the turbine 4 of the gas turbine 6. In addition, when the gas turbine power plant 1 shown in FIG. 1 is a single-shaft combined cycle plant, the first operation index a is the power generation amount of each of the steam turbine and the gas turbine that are components of the single-shaft combined cycle plant.

  When the first operation index a is the turbine inlet temperature of the gas turbine 6, the sensor output value x includes the power generation amount of the gas turbine 6, the calorific value of fuel supplied to the gas turbine 6, the flow rate, the temperature, and the gas turbine 6. The outlet temperature of the compressor 2, the outlet pressure, the intake air temperature, the exhaust gas temperature of the turbine 4 of the gas turbine 6, the rotational speed, the atmospheric pressure, the fuel-air ratio, the IGV (Inlet Guide Vane) angle of the compressor 2, An air bypass valve opening or the like is used. When the first operation index a is the power generation amount of each of the steam turbine and the gas turbine of the single-shaft combined cycle plant, the sensor output value x is the power generation amount in the combined cycle and the heat generation amount of the fuel supplied to the gas turbine 6. , Flow rate, temperature, outlet temperature of the compressor 2 of the gas turbine 6, outlet pressure, intake air temperature, exhaust gas temperature of the turbine 4 of the gas turbine 6, rotational speed, atmospheric pressure, fuel / air ratio, IGV (Inlet Guide) of the compressor 2 Vane: inlet guide blade) angle, air bypass valve opening, feed water flow rate to the steam turbine, and the like are used.

  The first estimation unit 12a that has acquired the sensor output value x and the operation condition y has the first operation index (for example, the turbine inlet temperature) a, the operation condition y of the gas turbine 6, and the sensor output value x of the gas turbine 6. The first operation index a during the rated operation of the gas turbine 6 is obtained from the above relationship. As the correspondence relationship, for example, a performance curve of the gas turbine 6 or a physical model of the gas turbine 6 can be used. The physical model of the gas turbine 6 can be created based on, for example, the thermal cycle of the gas turbine 6. The performance curve and the physical model can know how much the first operation index ac changes when the sensor output value x and the operation condition y deviate from the values at the rated operation.

  FIG. 3 is a conceptual diagram illustrating an example of a correspondence relationship between the first operation index and the sensor output value of the gas turbine. FIG. 4 is a conceptual diagram illustrating an example of a correspondence relationship between the first operation index and the operation conditions of the gas turbine. FIG. 5 is a conceptual diagram illustrating an example of a correspondence relationship between the first operation index, the operation condition of the gas turbine, and the sensor output value of the gas turbine. For example, from the physical model of the gas turbine 6, the correspondence between the first operation index a as shown by the solid line in FIG. 3 and the sensor output value x is obtained, and the first operation as shown by the solid line in FIG. Assume that the correspondence between the index a and the operating condition y is obtained. FIG. 5 shows the correspondence between the first driving index a, the sensor output value x, and the driving condition y in three-dimensional coordinates.

  LM indicated by a solid line in FIGS. 3 to 5 indicates a correspondence relationship between the first operation index a, the sensor output value x, and the operation condition y. As indicated by LM, the first operation index a is a function of the sensor output value x and the operation condition y. Hereinafter, LM is referred to as a driving index estimation function. The driving index estimation function LM is stored in the storage unit 13 of the abnormality detection device 10.

  3 and 5, x (t) is the sensor output value x at time t, and y (t) in FIGS. 4 and 5 is the operating condition y at time t. 3 and 5 is a sensor output value x corresponding to the rated operation of the gas turbine 6, and yc of FIGS. 4 and 5 is an operating condition y at the rated operation of the gas turbine 6. FIG. 3 to 5 is a first operation index a at time t. The ac (t) in FIGS. 3 to 5 is a first value obtained by matching the first operation index a (t) in x (t) and y (t) with the value at the rated operation of the gas turbine 6. It is a driving index.

For example, the turbine inlet temperature T as the first operation index a is P ² / G ² . P is the turbine inlet pressure and G is the turbine inlet flow rate. In the above equation, the turbine inlet pressure P is determined by the gas turbine 6, and the turbine inlet flow rate G is the sum of the compressor discharge air amount and the fuel amount. The compressor discharge air amount is a function of the compressor outlet temperature, the compressor outlet pressure, the rotation speed, and the IGV angle, and the fuel amount is a function of the fuel flow rate. The generator end output or thermal efficiency value of the gas turbine 6 varies depending on operating conditions such as intake air temperature or atmospheric pressure. As these values change, the value of the turbine inlet temperature T also changes. In the present embodiment, values having different operating conditions are converted into values mapped to operating condition values during rated operation.

  The first estimation unit 12a that has acquired the sensor output value x (t) and the operation condition y (t) at time t1 reads the operation index estimation function LM from the storage unit 13. Then, the first estimation unit 12a maps the sensor output value x (t) and the operation condition y (t) to values at the rated operation using the operation index estimation function LM, and the first operation index. Adjust a to the value ac (t) during rated operation. The first estimation unit 12a outputs this value as the first driving index a (a = ac (t)).

  When the first operation index a is the turbine inlet temperature T, the turbine inlet temperature T is a balance between the heat energy of the turbine inlet and the work of the turbine 6, the amount of heat of exhaust gas, loss, etc., compression of the compression ratio and the corrected flow rate. It can be estimated from a plurality of equations such as the characteristics of the vessel 2. As described above, when the turbine inlet temperature T is a function of a plurality of parameters, the turbine inlet temperature T is set to a value other than, for example, one of the plurality of parameters (for example, the intake air temperature) during the rated operation. In this way, the turbine inlet temperature T can be expressed as a function of, for example, one of the plurality of parameters (intake temperature in this example).

  In this way, the first operation index a (= ac (t)) obtained at different timings is rated by converting the first operation index a (= ac (t)) into a value during rated operation. It can be evaluated under the same criteria of driving. Further, the abnormality detection device 10 can estimate the operating state of devices that are difficult or cannot be detected by the sensor, based on a physical model for estimation, and determine the presence or absence of abnormality. If the value is not converted to a value during rated operation, a value close to the operation condition is selected, and an abnormality is determined by comparison with a comparison target that meets each condition. However, since the abnormality detection device 10 converts the sensor output value x (t) and the operating condition y (t) into values at the rated operation, it is necessary to select the sensor output value x (t) when detecting the abnormality. There is no advantage. Moreover, there is an advantage that only one value to be compared (comparison value at rated operation) is sufficient.

  In this embodiment, the 1st estimation part 12a calculates | requires the 1st driving | operation parameter | index ac (t) at several timings. For example, the first estimation unit 12a acquires the sensor output value x (t) and the operation condition y (t) at a predetermined interval (for example, every 10 seconds), and sets the first operation index ac (t). Ask. Then, the abnormality information calculation unit 12b includes a set A {ac (t1), ac (t2),... Ac (tn) of a plurality of first driving indices ac (t) obtained within a predetermined time T. }, The first abnormality detection information G1 for detecting the abnormality of the gas turbine 6 is obtained (the first estimation unit 12a obtains the sensor output value x and the operating condition y for t1, t2, etc.). 1 is the time at which the driving index ac is obtained, and n is the number of data included in the set A). Therefore, the abnormality information calculation unit 12b outputs the first driving index ac (t) to the abnormality information calculation unit 12b. In the present embodiment, the first estimating unit 12a may output the first driving index ac (t) to the abnormality information calculating unit 12b every time it obtains one or the plurality of first driving indexes ac (t The set A of t) may be output to the abnormality information calculation unit 12b.

FIG. 6 is an explanatory diagram of a technique for obtaining the first abnormality detection information. The abnormality information calculation unit 12b receives the set A {ac (t1), ac (t2),... Ac (tn)} of the first driving index ac (t) as an input, and uses the statistical method to obtain the first abnormality. The detection information G1 is obtained. In the present embodiment, the first abnormality detection information G1 is obtained by Expression (1). m is a natural number, σ is a standard deviation of the set A obtained from the equation (2), μ is an average value of the set A, and ac is a first operation index a corresponding to the rated operation of the gas turbine 6. The value of m can be arbitrarily set according to the purpose at the time of detecting an abnormality. In Expression (2), n is the number of data included in the set A as described above.
G1 = m × σ− | μ−ac | (1)
σ = √ {1 / (n−1) × Σ (ac (ti) −μ) ² }, [i = 1 to n] (2)

  After obtaining the first abnormality detection information G1, the abnormality information calculation unit 12b outputs this to the abnormality detection unit 12c. The abnormality detection unit 12c compares the first abnormality detection information G1 with a predetermined threshold value (abnormality detection threshold value) Gc that is set in advance, and detects whether there is an abnormality in the gas turbine 6. And the abnormality detection part 12c outputs the signal (abnormality determination signal) J which shows the presence or absence of abnormality, and preserve | saves it at the memory | storage part 13, or outputs the abnormality determination signal J to the control panel 14 shown in FIG. The control panel 14 that has acquired the abnormality determination signal J displays on the display 14D, for example, that the abnormality has occurred in the gas turbine 6.

  The predetermined threshold Gc is obtained in advance based on information obtained from past operation of the gas turbine 6 or information obtained from operation of a gas turbine similar to the gas turbine 6 and stored in the storage unit 13 of the abnormality detection device 10. Is done. When detecting the presence or absence of an abnormality in the gas turbine 6, the abnormality detection unit 12c reads the abnormality detection threshold Gc from the storage unit 13 and compares it with the first abnormality detection information G1. As a result, if Gc ≦ G1, it is assumed that an abnormality has occurred in the gas turbine 6, and the abnormality detection unit 12c outputs the abnormality occurrence signal J (f) as the abnormality determination signal J. Further, if Gc> G1, it is assumed that no abnormality has occurred in the gas turbine 6, and the abnormality detection unit 12c outputs the normal operation signal J (c) as the abnormality determination signal J.

  As described above, the abnormality detection apparatus 10 obtains the first operation index a (= ac (t)) indicating the operation state of the gas turbine 6 by estimating it from the physical model of the gas turbine 6 at a predetermined timing. Based on the statistical value (abnormality detection information) obtained by statistically processing the obtained set of first operation indices a, the presence / absence of abnormality of the gas turbine 6 is detected. Thus, since the abnormality detection apparatus 10 calculates | requires the 1st driving | operation parameter | index a from the physical model etc. of the gas turbine 6, even if it is the 1st driving | operation parameter | index which cannot be detected with the sensor 7, it is abnormal. Can be detected. The configuration of the present embodiment can be appropriately applied to the following embodiments, and those having the same configuration as the present embodiment have the same operations and effects as the present embodiment.

(Embodiment 2)
FIG. 7 is a schematic diagram illustrating a configuration example of the abnormality detection apparatus according to the second embodiment. FIG. 8 is a control block diagram of the abnormality detection apparatus according to the second embodiment. FIG. 9 is a conceptual diagram showing the frequency of occurrence of abnormality in the gas turbine. In the abnormality detection device 10A of the second embodiment, in addition to the abnormality detection device 10 of the first embodiment, abnormality has occurred in the abnormality detection information obtained by the abnormality information calculation unit 12b and the gas turbine 6 detected by the abnormality detection unit 12c. The information processing unit 12d that obtains the frequency of occurrence of abnormality in the gas turbine 6 in a predetermined period is different from the information (abnormality determination signal). Next, the abnormality detection device 10A will be described in more detail.

  As shown in FIG. 8, the abnormality information calculation unit 12b acquires the first driving index a (= ac (t)) obtained by the first estimating unit 12a, and calculates the first driving index ac (t). Based on the set A, the abnormality detection information G1 is obtained. Then, the abnormality information calculation unit 12b outputs the obtained abnormality detection information G1 to the abnormality detection unit 12c and the information processing unit 12d. The abnormality detection unit 12c that has acquired the abnormality detection information G1 outputs an abnormality determination signal J to the information processing unit 12d.

  The information processing unit 12d obtains an abnormality occurrence frequency of the gas turbine 6 in a predetermined period from the abnormality detection information G1 and the abnormality determination signal J. For example, the information processing unit 12d uses the abnormality detection information G1 and the abnormality determination signal to generate an abnormality occurrence frequency of the gas turbine 6 with a predetermined time width such as seconds, minutes, and hours, that is, the abnormality occurrence signal J ( The occurrence frequency of f) or the average value of the first abnormality detection information G1 (F in FIG. 9) that is equal to or higher than the abnormality detection threshold Gc is calculated. The information processing unit 12d compares these calculated values with a predetermined threshold value stored in the storage unit 13 to thereby determine the type of abnormality (acute or chronic) that has occurred in the gas turbine 6. Or the like). For example, the type of abnormality that has occurred in the gas turbine 6 can be determined based on the number of times, the frequency, the pattern of exceeding the abnormality detection information G1 exceeding the abnormality detection threshold Gc, and the like.

  Thus, according to the present embodiment, it is possible to determine whether an abnormality that has occurred in the gas turbine 6 is acute or chronic. The configuration of the present embodiment can be appropriately applied to the following embodiments, and those having the same configuration as the present embodiment have the same operations and effects as the present embodiment.

(Embodiment 3)
FIG. 10 is a schematic diagram illustrating a configuration example of the abnormality detection apparatus according to the third embodiment. FIG. 11 is a control block diagram of the abnormality detection apparatus according to the third embodiment. The abnormality detection device 10B of the third embodiment uses the first driving index a (= ac (t)) estimated by the first estimation unit 12a in addition to the abnormality detection device 10 of the first embodiment. It has the 2nd estimation part 12e which estimates the 2nd driving | operation parameter | index q which shows the driving | running state of the gas turbine 6 different from the driving | operation index | exponent at several timings. Then, the abnormality information calculation unit 12b included in the abnormality detection device 10B indicates the set of the second operation index q obtained by the second estimation unit 12e and the operation state of the gas turbine 6 corresponding to the second operation index q. Based on the sensor output value xq output from the sensor 7 to be detected, second abnormality detection information G2 for detecting an abnormality of the gas turbine 6 is obtained. In addition, the abnormality detection unit 12c included in the abnormality detection device 10B compares the second abnormality detection information G2 obtained by the abnormality information calculation unit 12b with a predetermined threshold G2c set in advance, and detects an abnormality in the gas turbine 6. Detect the presence or absence.

  The first estimation unit 12a receives the sensor output value x and the operation condition y as input, obtains the first operation index a (= ac (t)) by the method described in the first embodiment, and obtains the second operation index a (= ac (t)). It outputs to the estimation part 12e. The second estimating unit 12e uses the first operation index a (= ac (t)) to obtain a second operation index q indicating the operation state of the gas turbine 6 different from the first operation index a. . The second estimation unit 12e obtains the second driving index q by the same method as the first estimation unit 12a. In the present embodiment, since the second estimation unit 12e receives only the first driving index a and does not input the driving condition y, the correspondence relationship between the second driving index q and the first driving index a. Thus, the second driving index q is obtained. In the present embodiment, the second estimation unit 12e may obtain the second operation index q by inputting the operation condition y or the sensor output value x in addition to the first operation index a. The second estimation unit 12e outputs the obtained second operation index q to the abnormality information calculation unit 12b.

  The abnormality information calculation unit 12b obtains second abnormality detection information G2 for detecting an abnormality of the gas turbine 6 based on the second set of operation indicators q and the sensor output value xq. As described above, the sensor output value xq is a value output from the sensor 7 that detects the operation state of the gas turbine 6 corresponding to the second operation index q. As the sensor output value xq, a value that is not used when the first estimation unit 12a obtains the first driving index a may be used.

  The abnormality information calculation unit 12b, for example, the average value of the set obtained by statistically processing a set of sensor output values xq acquired at a plurality of timings and the second driving index q obtained at a plurality of timings. A difference (average value difference) from the average value of the set obtained by statistically processing the set is obtained. Then, the abnormality information calculation unit 12b outputs the obtained average value difference to the abnormality detection unit 12c as second abnormality detection information G2. In addition, you may obtain | require 2nd abnormality detection information G2 by methods other than the method mentioned above. For example, the abnormality information calculation unit 12b uses the standard deviation, the maximum value, the minimum value, and the like obtained by statistically processing the set of sensor output values xq and the set of the second operation index q, to detect the second abnormality detection information. G2 may be obtained.

  The abnormality detection unit 12c compares the second abnormality detection information G2 with a predetermined threshold value G2c set in advance and stored in the storage unit 13, and detects the presence or absence of an abnormality in the gas turbine 6. As a result, if G2c ≦ G2, it is assumed that an abnormality has occurred in the portion related to the gas turbine 6, more specifically, the second abnormality detection information G2, and the abnormality detection unit 12c outputs the abnormality occurrence signal J ( f) is output as the abnormality determination signal J. Further, if G2c> G2, it is assumed that no abnormality has occurred in the gas turbine 6, and the abnormality detection unit 12c outputs the normal operation signal J (c) as the abnormality determination signal J. The predetermined threshold G2c is obtained in advance based on information obtained from past operation of the gas turbine 6 or information obtained from operation of a gas turbine similar to the gas turbine 6, and is stored in the storage unit 13 of the abnormality detection device 10. Is done.

  In the present embodiment, for example, when detecting an abnormality occurring in the nozzle clogging of the combustor 3 of the gas turbine 6, the input to the second estimation unit 12 e is the power generation amount of the gas turbine 6. The power generation amount of the gas turbine 6 is obtained by the first estimation unit 12a and is input to the second estimation unit 12e as the first operation index a. The second estimation unit 12e outputs BPT (Blade Pass Temperature: gas temperature at the final stage moving blade outlet gas of the turbine 4 of the gas turbine 6) as the second operation index q.

  The abnormality information calculation unit 12b receives the sensor output values xq obtained at a plurality of timings and the second operation index q obtained at the plurality of timings. The sensor output value xq at this time is the BPT detected by the sensor that detects the BPT. The abnormality information calculation unit 12b includes an average value of the set obtained by statistically processing the set of sensor output values xq, and an average value of the set obtained by statistically processing the set of second driving indicators q. Difference (average value difference) is obtained. Then, the abnormality information calculation unit 12b outputs the obtained average value difference to the abnormality detection unit 12c as second abnormality detection information G2. The abnormality detection unit 12c compares the abnormality detection information G2 with a predetermined threshold G2c obtained in advance for the BPT, and detects an abnormality in the BPT.

  In the present embodiment, the abnormality detection of the gas turbine 6 is performed by more various methods by using the first operation index a or using the sensor output value xq that is not used when obtaining the first operation index a. Can be realized.

DESCRIPTION OF SYMBOLS 1 Gas turbine power generation plant 2 Compressor 3 Combustor 4 Turbine 5 Generator 6 Gas turbine 7 Sensor 10, 10A, 10B Abnormality detection apparatus 11 Input / output part 12 Processing part 12a First estimation part 12b Abnormal information calculation part 12c Abnormality detection Unit 12d information processing unit 12e second estimation unit 13 storage unit

Claims (4)

And the sensor output value outputted from the sensor for detecting an operating condition and operating condition of the gas turbine of the gas turbine, correspondence defined operating indicator estimation function of the first operational indicator of the operating condition of the gas turbine Is a pre-stored storage unit;
Using the operation index estimation function, a first operation index in a specific operation state during steady operation is obtained at a plurality of timings from a sensor output value for detecting the operation state of the gas turbine and the operation conditions of the gas turbine. A first estimation unit;
Abnormal information calculation for obtaining first abnormality detection information for detecting abnormality of the gas turbine based on a set of first operation indices in a specific operation state during the steady operation estimated by the first estimation unit And
An abnormality detection unit that detects the presence or absence of an abnormality in the gas turbine by comparing the abnormality detection information obtained by the abnormality information calculation unit with a predetermined threshold value set in advance;
An abnormality detection device including: The abnormality detection device according to claim 1, wherein the specific operation state during the steady operation is during a rated operation .   Information for determining the occurrence frequency of abnormality of the gas turbine in a predetermined period from the abnormality detection information obtained by the abnormality information calculation unit and information that abnormality has occurred in the gas turbine detected by the abnormality detection unit The abnormality detection device according to claim 1, further comprising a processing unit. A second operation index indicating an operation state of the gas turbine different from the first operation index is estimated at a plurality of timings using the first operation index estimated by the first estimation unit. Has two estimators,
The abnormality information calculation unit is a sensor that is output from a sensor that detects a set of the second operation index obtained by the second estimation unit and an operation state of the gas turbine corresponding to the second operation index. Based on the output value, second abnormality detection information for detecting an abnormality of the gas turbine is obtained,
The abnormality detection unit detects the presence or absence of an abnormality in the gas turbine by comparing the second abnormality detection information obtained by the abnormality information calculation unit with a predetermined threshold value set in advance. The abnormality detection device according to any one of the above.

Images