JP2003074851A - Eigenvalue-prediction method in combustion equipment and evaluation method of response factor between measuring devices - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an eigenvalue-prediction method in combustion equipment and evaluation method of response magnification factor between measuring devices, in which a reference measuring-device is used downstream of flames to obtain a correct value and an object measuring-device is used upstream of the flames, and the measured values are compared with each other to calculate automatically a correct value in regulating fuel-air ratio by detecting oscillating combustion and pressure fluctuation in combustion equipment such as a gas turbine and rocket-propulsion engine. SOLUTION: A reference measuring-device is placed on a high-temperature side of flames in a combustor such as a gas turbine and a rocket-propulsion engine, and an object measuring-device is placed on a low temperature side. A response magnification factor between the reference and object measuring-devices is calculated from measurement results of the devices. Unusual data is filtered by modified Thompson τ-method. Residual data is further filtered by 2σ to obtain variance of the data, and further a regression line is calculated to obtain a calibration factor. Thus, the measurement can be carried out singly with the object measuring-device.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring eigenvalues such as internal pressure in a combustor such as a gas turbine or a rocket propulsion device or pressure due to acoustic characteristics, and measuring the relationship between heat quantity, heat density and damping. It is a thing.

[0002]

2. Description of the Related Art A combustor such as a gas turbine or a rocket propulsor has a combustor 11 attached to a casing 12 of the gas turbine, as shown in FIG. There is. And this combustor 11
In the outer cylinder 16, the fuel nozzle 13, the inner cylinder 14, and the tail cylinder 1 are provided.
5, a tail pipe 15 is provided with a bypass elbow 17, a bypass valve 18, a bypass valve variable mechanism 19
Is provided. 20 is an air compressor, and the compressor discharge air 21 compressed here flows to the passenger compartment 12,
It is introduced as combustion air into the combustor 11 from the upstream side of the fuel nozzle 13 by passing around the combustor 11 as shown by an arrow. Then, in the combustor 11, the fuel fed through the fuel nozzle 13 is burned, and the combustion gas thereof is sent to the gas turbine 22 to drive it.

In the combustor 11 of such a gas turbine, combustion vibrations and pressure fluctuations are generated and reflected inside the constituent members of the combustor 11, such as the inner cylinder 14, the tail cylinder 15, and the outer cylinder 16, to cause large vibrations. There is concern that the energy may become energy and cause fatigue damage to the main body of the combustor 11, and further damage to peripheral components such as turbine blades. Therefore, such combustion vibrations and pressure fluctuations are detected by using a pressure sensor, a microphone, etc., and the distribution ratio of fuel and air, pilot ratio, bypass valve opening, etc. are adjusted.

In order to accurately measure such combustion vibrations and pressure fluctuations, a measuring device such as a pressure sensor or a microphone for detecting these combustion vibrations and pressure fluctuations is provided on the downstream side of the flame in the combustor. Although it is preferable to install it, the wake side of the flame is hot and it is necessary to use a pressure sensor or a microphone that can withstand this high temperature. However, measuring instruments such as pressure sensors and microphones that can withstand such high temperatures are generally expensive.

[0005]

Therefore, the first object of the present invention is to make it possible to carry out the measurement by using an inexpensive measuring device such as a sensor or a microphone which cannot withstand a high temperature.

However, inexpensive sensors and microphones that cannot withstand high temperatures cannot be installed on the downstream side of the flame where accurate measurement values can be obtained. Therefore, install an inexpensive sensor or microphone that cannot withstand high temperatures as the target measuring instrument in a place that does not become hot, such as the upstream side of the flame, and use a pressure sensor or microphone that can withstand high temperature as a reference measuring instrument on the downstream side of the flame. It is necessary to install the method and calculate the calibration magnification by comparing the measurement values of the target measurement equipment and the reference measurement equipment to calculate the response magnification. However, the response magnification between the reference measuring instrument and the target measuring instrument generally fluctuates over a wide range in general, and the calculation of the calibration magnification from the varied response multiplying factor is conventionally done with the measured value of the target measuring instrument and the reference value. The ratio of the measured values of the measuring device is obtained and plotted, and an experienced technician eliminates the abnormal value in it and calculates the relationship,
It was inefficient and time consuming because it was done with the method of determining the magnification almost by intuition.

In order to deal with such a situation, for example, in Japanese Unexamined Patent Publication No. 8-166820, the distance between the first quartile and the second quartile of the cumulative frequency of measured values is H1 and the second quartile. A method of detecting an abnormal value by setting a distance between the quantile and the third quartile as H2 and setting a range for judging an abnormal value based on H1 and H2 is shown. That is, in determining an abnormal value of a measured value, conventionally, a method using standard deviation or a relative cumulative frequency of 25%, 50%, and 75% is used as a first quartile and a second quartile, respectively. , The third quartile is calculated as the quartile, and the distance from the first quartile to the third quartile is used as the hinge width, and the hinge width from each quartile is 3 Methods such as the quartile method, in which measured values that are more than twice as large as the abnormal value are used, were used when the frequency distribution of measured values is symmetrical, but when the frequencies are biased. Was not very effective. Therefore, by setting a range for determining an abnormal value based on the above H1 and H2, it is possible to deal with the deviation of the measured values.

However, this Japanese Patent Laid-Open No. 8-166820
Although the method disclosed in the publication has the advantage that it is effective even when the frequency distribution of measured values is asymmetrical, it shows a method of determining an abnormal value included in measured values from 2σ and 3σ of a normal distribution. However, there is no mention of calculating the calibration magnification as described above.

Therefore, in the present invention, a second object is to provide a calibration method capable of automatically and accurately calculating the calibration magnifications of the target measuring instrument and the reference measuring instrument.

[0010]

In order to solve the first problem, in the present invention, an inexpensive sensor or microphone that cannot withstand high temperature is installed as a target measuring device in a place where the temperature does not become high, such as the upstream side of a flame. Then, install a pressure sensor or microphone that can withstand high temperature as a reference measuring instrument on the downstream side of the flame, compare the measured values of this target measuring instrument and the reference measuring instrument to calculate the calibration magnification, and then calculate this calibration magnification. By using it, it was possible to measure the eigenvalues such as the internal pressure in the combustor and the pressure due to the acoustic characteristics only with the target measuring device, so that an accurate value could be obtained with an inexpensive measuring device.

In order to solve the above-mentioned second problem, in the present invention, an inexpensive sensor or microphone that cannot withstand high temperature is installed as a target measuring instrument in a place where the temperature does not become high, such as the upstream side of the flame, and the temperature becomes high. Install a pressure sensor or microphone that can withstand as a reference measuring instrument on the downstream side of the flame, calculate the response magnification from the measured values of this target measuring instrument and the reference measuring instrument, and correct the abnormal value of this response magnification with the Thompson τ method. Exclude and repeat the process of finding the variance of the remaining data and eliminating the abnormal value by 2σ, creating a regression line with the finally remaining data, and using the slope as the calibration value magnification, The calibration magnification can be calculated automatically.

By doing so, an inexperienced technician can eliminate the abnormal value in the measured values, calculate the relationship, and determine the magnification by intuition, which is inefficient. Anyone can quickly and automatically calculate an accurate calibration magnification without using a time-consuming method.

In order to solve the first problem, claim 1
Is a method invention, which is a method of predicting eigenvalues such as internal pressure, sound pressure, and heat quantity in a combustor, in which a reference measuring instrument is used on the high temperature side of a flame generated in the combustor and a target measuring instrument is used on the low temperature side. Installed, calculate the calibration magnification for the reference measuring instrument in the target measuring instrument from the measurement results of the reference measuring instrument and the target measuring instrument, and use only the target measuring instrument installed on the low temperature side of the flame using the calibration factor The feature is that the value can be predicted.

By doing so, it is possible to perform accurate measurement with an inexpensive pressure sensor or microphone, without using an expensive pressure sensor or microphone that can withstand high temperatures as described above, which has a great economic effect. To bring.

Next, in order to solve the second problem, a second aspect is provided.
Is also a method invention, which is a method for evaluating response magnification between a plurality of measuring instruments when measuring eigenvalues such as pressure due to internal pressure or acoustic characteristics in the combustor, in which a flame is generated in the combustor. The reference measuring device at the high temperature side position of the flame, the target measuring device is installed at the low temperature side position, and after calculating the response magnification of the reference measuring device and the target measuring device from the measurement results of the reference measuring device and the target measuring device, The abnormal data of the response magnification is filtered by the modified Thompson τ method, the variance of the remaining data is obtained, and the regression straight line is calculated by filtering by 2σ, and the slope of the regression straight line is used as the calibration magnification of the target measuring instrument. It is characterized by

By thus filtering the measurement results using the modified Thompson τ method, when the number of measurement results exceeds a certain level, the probability of excluding normal measurement values is set to 5%, and the abnormal values are automatically and accurately determined. Further, by calculating the regression line after filtering using the variance, it is possible to calculate the accurate calibration magnification completely automatically without bothering an experienced person.

Then, the response multiplying factor is defined in claims 3 and 4.
As described above, the plurality of measuring devices are pressure sensors for measuring the internal pressure in the combustor, and the response magnification of the pressure sensor provided at the target measuring device installation position in the combustor is evaluated. . The plurality of measuring devices are microphones for measuring acoustic characteristics of the combustor when the combustion is not performed, and the response magnification of the microphone provided at the target measuring device installation position in the combustor is evaluated. As a result, it is possible to obtain a calibration magnification that can accurately measure the pressure and acoustic characteristics in the combustor such as the gas turbine and the rocket propulsion device.

The response magnification evaluation method is defined in claim 5.
As described in, it is a method of evaluating the response magnification of the attenuation amount of the acoustic system corresponding to the heat quantity and heat density in the combustor, the heat quantity and heat density in the combustor and the sound pressure for each frequency are measured, and the heat quantity After calculating the response magnification of the attenuation amount by the heat generation density and the frequency of sound pressure, the abnormal data of the response magnification is filtered by the modified Thompson τ method to obtain the variance of the remaining data.
After the filtering with σ, a regression line is calculated, and the slope of the regression line is used as the response magnification.

By doing so, the response magnification of the attenuation amount of the acoustic system corresponding to the amount of heat and the heat generation density in the combustor such as the gas turbine and the rocket propulsion device can be completely eliminated without bothering experienced persons as described above. It can be calculated automatically and accurately.

[0020]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be exemplarily described in detail below with reference to the drawings. However, the dimensions, materials, shapes, relative positions, and the like of the constituent parts described in this embodiment are not intended to limit the scope of the present invention thereto, unless otherwise specified, and are merely It is only an example.

FIG. 1 schematically shows a combustor such as the gas turbine or rocket propulsor shown at 11 in FIG. 11 and a state of a flame generated by the combustor, and a pressure sensor or a microphone for detecting combustion vibration or pressure fluctuation. FIG. 1 is an explanatory view showing a place where a measuring instrument such as is installed, 1 is a combustor such as a gas turbine or a rocket propulsor, 2 is a flame, 3 is a pressure sensor installed on the downstream side of the flame and capable of withstanding high temperatures, Reference measuring devices 4 such as a microphone are target measuring devices such as a pressure sensor and a microphone installed upstream of the flame that cannot withstand high temperatures.

In the present invention, the target measuring device 4 such as a pressure sensor or a microphone, which cannot withstand a high temperature, is installed on the upstream side of the flame 2 in the combustor 1 such as a gas turbine or rocket propulsor. It is calibrated by a reference measuring device 3 such as a pressure sensor and a microphone that can withstand high temperature installed on the wake side of the engine, and in actual use, combustion vibration and pressure fluctuations are measured only by the target measuring device 4 without using the reference measuring device 3. I was able to predict.

By doing so, it is possible to accurately predict the eigenvalue using only a low-priced measuring instrument without using an expensive pressure sensor or microphone that can withstand high temperatures.
There are significant economic benefits.

The reference measuring device 3 installed as shown in FIG.
2 shows a graph in which the frequency spectrum of the pressure in the combustor 1 such as a gas turbine or a rocket propulsion device is plotted by the target measuring device 4. Of these, (A) is the measured value P ₀ (i) of the reference measuring device 3 (i: 1, 2, 3, ... i)
Graph (B) is the measured value P ₁ (i) of the target measuring instrument 4
(I: 1, 2, 3, ... i), where i is the same graph shows the values of the reference measuring device 3 and the target measuring device 4 at the same time point. In each graph, the horizontal axis is frequency (H
z), the vertical axis represents pressure (Pa), and the graph of FIG. 2 corresponds to each of a plurality of measurement values i (i = 1, 2, 3, ... I) of each measuring device.

As can be seen from FIG. 2, the pressure in the combustor 1 such as a gas turbine or a rocket propulsion device has peak values at various frequencies depending on the combustion condition at each time. If the measurement time point i of the container 4 is the same, the peak value will be approximately the same frequency. Therefore, the ratio of the peak values P ₀ (i) and P ₁ (i) where i is equal, that is, the magnification A (i) (i: 1, 2, 3, ...
When the measurement value P ₀ is plotted on the horizontal axis and the measurement value P ₁ of the target measuring device 4 is plotted on the vertical axis on the plane of P ₀ and P ₁ , the variation is plotted in this plane as shown in FIG.

[0026] Therefore, to eliminate the outliers from the plotted values is not determined calibration factor further obtain the relation of the measured values P ₁ of the reference measuring device 3 measurements P ₀ and the target instrument 4 However, as described above, according to the experience of engineers in the past, for example, a value in the range indicated by 30 in FIG. 3 is set as an effective region, and values outside this region 30 are values that remain as abnormal values and have a relationship. It was done by calculating and intuition to determine the magnification. Therefore, in the present invention,
By the response magnification evaluation method between measuring instruments as shown in the schematic flow chart of FIG. 4, anyone can quickly and accurately obtain an accurate calibration magnification without using such an inefficient and time-consuming method. It was possible to calculate.

That is, in the response magnification evaluation method between the measuring devices in the combustor such as the gas turbine and the rocket propulsion device of the present invention, first, the reference measuring device 3 is set to the limit value which is predetermined in the first step S1. the measured value P ₀ by filtering, the response magnification of the measured value P ₀ of the reference measuring device 3 that this filtering at the second step S2, a measured value P ₁ of the object measuring instrument 4 corresponding to the measured value P ₀ a Find (n). Then, in the third step S3, this response magnification A
(N) is filtered by the modified Thompson τ method to eliminate the abnormal value and replace the remaining response magnification A (n) with A (m), and further calculate the variance of the remaining data in the fourth step S4 to obtain 2σ. Repeat to eliminate abnormal values with. The remaining A (m) is replaced with A (k), a regression line is created in the fifth step S5 using this A (k), and the slope of this regression line is calculated in the sixth step S6. The calibration factor can be automatically calculated as the calibration value.

First, the first step S1 is to filter the measured value P ₀ (i) of the reference measuring device 3 using a predetermined limit value α, and this limit value α is It is a value that excludes those that exceed the range empirically known. α <P ₀ (i) (α: empirical value) (1) Then, assuming that the number of valid data remaining in (1) is n, the n number of data in the next second step S2. P
The response magnification A (n) with the measured value P ₁ (n) of the target measuring device 4 corresponding to ₀ (n) is calculated by the following equation (2).

[Equation 1]

The response magnification A (n) is filtered by the modified Thompson τ method in the third step S3. The modified Thompson τ method is used by the Japan Society of Mechanical Engineers on November 25, 1987. As described in detail in “Measurement Uncertainty” on pages 22 to 23 of the day, when the number of measurement results exceeds a certain level, the probability of excluding normal measurement values is set to 5% and automatically. , It is possible to accurately exclude outliers.

To explain the outline of this modified Thompson τ method, if there is a data consisting of N measurement values X _i , the precision S and the average value are

[Equation 2]

[Equation 3] Becomes If the j-th measured value X _j is suspected to be an abnormal value, the absolute value of the difference between X _j and the average value is

[Equation 4] Becomes Here, using Table 1, 5 for the size N of the material
The value of τ at the% significance level is determined, which limits the probability of excluding normal measurements to 5%. (The probability of not excluding outliers is not constant and depends on the size of the material.)

[0031]

[Table 1]

An abnormal value is judged by the difference δ according to the equation (5),
This is done by comparing the product τS of τ and the equation (3).・ If δ is equal to or larger than τS, X _j is an abnormal value. ・ If δ is smaller than τS, X _j is not an abnormal value. Thus, when an abnormal value is excluded, the measured value excluding the abnormal value is accurate. The degree S and the average value are recalculated, and δ is also recalculated.
Similarly, the abnormal value is determined and eliminated until the abnormal value disappears.

When the abnormal value is eliminated by the modified Thompson τ method in this way, the remaining data in the response magnification A (n) is replaced with A (m) (m = 1, 2, 3, ... M),
This time, the variance is obtained and filtered by 2σ in the fourth step S4. For example, as shown in FIG. 5, the measured value P ₀ of the reference measuring device 3 is plotted on the horizontal axis and the target measuring device 4
Plotting the measurement result measured values P ₁ to the plane of P ₀ and P ₁ that the longitudinal axis of, since the data, such as 50 by the third step S3 is eliminated, the value of the remaining A (m) The variance is calculated, and σ is calculated from η and ξ which are coordinate-converted by the following equation (6). Then, the abnormal value 51 is eliminated by 2σ, the same operation is performed on the remaining data, and the abnormal value 52 is eliminated.

[Equation 5]

When there is no abnormal value excluded by 2σ in this way, the remaining data is A (k) (k = 1,
2, 3, ... K), and by this A (k), a regression line 60 passing through the origin as shown in FIG. 6 is drawn in the fifth step S5, and this regression line is drawn in the sixth step S6. The slope is the calibration magnification.

By doing so, an accurate calibration magnification can be obtained by simply proceeding with the calculation according to the flow shown in FIG. 4, and the measured value of the target measuring instrument and the reference measuring instrument as in the conventional case can be obtained. The ratio of the measured values is calculated and plotted, and an experienced technician eliminates outliers among them and calculates the relationship to determine the multiplication factor, which is inefficient. It is not necessary to use such a method, and by doing so, an inexpensive measuring device that cannot withstand high temperatures can be used, and a great economic effect can be brought about.

The reference measuring device 3 and the target measuring device 4 described with reference to FIG. 1 have been described by taking the case where the flame 2 is generated as an example, but as shown in FIG.
Is installed, measurement is performed when the combustor such as a gas turbine or rocket propulsion is not burning, and the calibration value is obtained by the flow of FIG. 4 described above. Measurements can be performed only with unbearable microphones.

In the above description, the present invention has been described by taking as an example the case of measuring the pressure and acoustic characteristics in the combustor such as a gas turbine or rocket propulsion device. In the above, the attenuation amount of the acoustic system changes according to the amount of heat and the heat generation density, and the amount of attenuation also affects the above-mentioned combustion oscillation and pressure fluctuation. Therefore, although the measurement of this attenuation amount also has an important meaning, this measured value also often greatly varies, and an accurate attenuation amount can be obtained by using the response magnification evaluation method shown in FIG.

The damping ratio in the combustor such as a gas turbine or rocket propulsion device is ζ, the heat generation fluctuation is q ', the total heat quantity is Q,
When the heat generation density is θ, the sound pressure fluctuation is p ′, and the acoustic space is V, these relationships are expressed by the following equation (7).

[Equation 6] That is, assuming that the distribution of the heat generation fluctuation q ′ is in the section A of FIG. 8, the heat generation fluctuation q ′ is a function of the heat quantity Q, and if it is simply proportionally calculated, the larger the heat quantity Q is from the equation (7). It can be seen that the damping ratio ζ changes.

When the same total heat quantity Q is applied to the sections A and B of FIG. 8, the section A, that is, the space volume is smaller than the section B, so that the heat generation density θ becomes large.
From the equation (7), the damping ratio ζ is determined by the sound pressure fluctuation p ′ and the heat generation fluctuation q ′.
Since it can be expressed by the product of, the narrower the interval, the greater the influence on the stability of the system (damping ratio ζ). In an extreme case, when the heat generation fluctuation q ′ is distributed in a section that crosses plus and minus of the sound pressure fluctuation p ′, there is a portion to be canceled, so that the efficiency affecting the stability (damping ratio ζ) is poor. Become.

Now, as shown in FIG. 9, the frequency spectrum of the sound pressure in the combustor 1 such as a gas turbine or a rocket propeller is plotted on a plane with the frequency (Hz) on the horizontal axis and the pressure (Pa) on the vertical axis. Then, assuming that the frequency of the peak value is f ₀ and the frequency of the pressure that is 1 / √2 of the peak value is f ₁ and f ₂ , the damping frequency (damping ratio) ζ is expressed by the following equation (8).

[Equation 7] Therefore, by measuring a plurality of these values and calculating the response magnification between the heat quantity and the damping ratio according to the flow shown in FIG. 4, it is possible to derive a fairly accurate relationship between the heat quantity and the damping ratio from the varied values.

As described above, the relationship between the heat generation density and the damping ratio is the same. That is, although the heat generation density θ can be derived from the sound pressure by the equation (7), the response magnification α (i) is calculated by the following equation (9) from the heat generation density θ and the damping ratio ζ derived by the equation (8). ) (I = 1, 2, 3, ...
When i) is calculated and plotted on the plane of θ and ζ as shown in FIG. 10, this value also varies like the relationship between the heat quantity and the damping ratio. Therefore, by calculating this response magnification α (i) according to the flow shown in FIG. 4, it is possible to derive a fairly accurate relationship between the heat generation density and the damping ratio, which have varied.

[0042]

As described above, according to the present invention, an inexpensive sensor or microphone, which cannot withstand high temperature, is used as a target measuring device in a combustor such as a gas turbine or a rocket propulsion device, and a high temperature such as an upstream side of a flame is detected. Install in a place where it does not occur, install a pressure sensor or microphone that can withstand high temperatures as a reference measuring instrument on the downstream side of the flame, and calculate the calibration magnification of both measuring instruments so that measurement can be performed only with the target measuring instrument. Therefore, it is possible to accurately predict the eigenvalue in the combustor with an inexpensive measuring device without using an expensive pressure sensor or microphone that can withstand high temperatures, which brings a great economic effect.

Further, in calculating the calibration magnification, the response magnification is calculated from the measured values of the target measuring instrument and the reference measuring instrument installed in the combustor such as the gas turbine or the rocket propulsion device, and the abnormal value of this response multiplying factor is calculated. Is eliminated by the modified Thompson τ method, the variance of the remaining data is obtained, and the abnormal value is eliminated by 2σ, and a regression line is finally created by the remaining data and the slope thereof is used as the calibration value magnification. Since an accurate calibration magnification can be automatically calculated in this way, an experienced technician, as in the past, eliminates abnormal values in the measured values, calculates the relationship, and uses the intuition to calculate the magnification. Anyone can quickly and automatically calculate an accurate calibration magnification without using an inefficient and time-consuming method of determining, and a great economic effect can be obtained.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing an embodiment of a measuring method in a combustor such as a gas turbine or a rocket propulsion device according to the present invention.

FIG. 2 is a graph in which a frequency spectrum of pressure in a combustor in a gas turbine, a rocket propulsion device or the like measured by a reference measuring device and a target measuring device is plotted.

FIG. 3 is a graph in which peak values of pressure in a combustor in a gas turbine, a rocket propulsion device, and the like measured by a reference measuring device and a target measuring device are plotted.

FIG. 4 is a schematic flow chart of a measuring method in a combustor such as a gas turbine or a rocket propulsion device according to the present invention.

FIG. 5 is an explanatory diagram for obtaining variance from measured values and excluding abnormal values by 2σ.

FIG. 6 is an explanatory diagram for obtaining a regression line from the remaining data after excluding abnormal values from measured values.

FIG. 7 is an explanatory diagram showing a place where a microphone or the like for measuring acoustic characteristics is installed.

FIG. 8 is an explanatory diagram showing sound pressure and spatial distance.

FIG. 9 is a diagram for explaining a damping frequency.

FIG. 10 is a diagram showing a response magnification α (i) plotted on a plane of a heat generation density θ and a damping ratio ζ.

FIG. 11 is a sectional view of the vicinity of a combustor in a gas turbine.

[Explanation of symbols]

1 Combustors such as gas turbines and rocket thrusters 2 flames 3 Standard measuring instrument 4 Target measuring instruments

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G01L 7/00 G01L 7/00 LF term (reference) 2F055 AA27 BB12 CC59 FF49 GG49 2G064 AA15 AB16 AB23 BA28 BD02 CC47

Claims (5)

[Claims] 1. A method for predicting eigenvalues of internal pressure, sound pressure, heat quantity, etc. in a combustor, wherein a reference measuring instrument is installed on the high temperature side of a flame generated in the combustor and a target measuring instrument is installed on the low temperature side. ,
Calculate the calibration factor for the reference measuring device in the target measuring device from the measurement results of the reference measuring device and the target measuring device, and use the calibration factor to predict the value on the high temperature side only with the target measuring device installed on the low temperature side of the flame. A method for predicting an eigenvalue in a combustor characterized by being able to do so. 2. A method for evaluating response magnification between a plurality of measuring instruments when measuring eigenvalues such as internal pressure in a combustor and pressure due to acoustic characteristics, which is a flame when a flame is generated in the combustor. Install the reference measuring device at the high temperature side position and the target measuring device at the low temperature side position, calculate the response magnification of the reference measuring device and the target measuring device from the measurement results of the reference measuring device and the target measuring device, and then correct Thompson By filtering the abnormal data of the response magnification by the τ method, further obtaining the variance of the remaining data and filtering by 2σ to calculate a regression line, and using the slope of the regression line as the calibration magnification of the target measuring device. A method for evaluating response magnification between measuring instruments in a characteristic combustor. 3. The pressure sensor for measuring the internal pressure in a combustor, wherein the plurality of measuring devices evaluate the response magnification of a pressure sensor provided in the target measuring device installation position in the combustor. Item 2. A response magnification evaluation method between measuring instruments in the combustor described in Item 2. 4. The microphone is a microphone for measuring acoustic characteristics of a combustor when the combustor is not burning, and the response magnification of a microphone provided at the target measuring instrument installation position in the combustor is evaluated. The method for evaluating response magnification between measuring devices in a combustor according to claim 2. 5. A method for evaluating a response magnification of an attenuation amount of an acoustic system corresponding to a heat quantity or a heat generation density in a combustor, wherein the heat quantity or the heat generation density in the combustor and the sound pressure for each frequency are measured, and the heat quantity or After calculating the response magnification of the attenuation amount by the heat density and the frequency of the sound pressure, filter the abnormal data of the response magnification by the modified Thompson τ method, obtain the variance of the remaining data, filter by 2σ, and then calculate the regression line. A response magnification evaluation method for a combustor, wherein the slope of the regression line is used as the response magnification.