Patent Grant
7797113
The present application claims priority from Japanese patent application JP 2007-214932 filed on Aug. 21, 2007, the content of which is hereby incorporated by reference into this application.
1. Field of the Invention
The present invention relates to a gas turbine performance analysis method and a gas turbine performance analysis system.
2. Description of Related Art
Gas turbine performance significantly changes due to not only operating conditions such as fuel flow rate but also atmospheric conditions such as inlet air temperature, it cannot be simply evaluated by using only absolute values. To solve this problem, there is a general method wherein the operating conditions of actual gas turbine are inputted into a performance calculation program of the gas turbine, and computed performance values, that is, proper predicted values are compared with actual measured values to conduct an evaluation of the gas turbine performance.
Actually, since the computed values obtained by the gas turbine performance calculation program never completely matches the actual measured values, to eliminate deviation, some adjustment is made to the computation method. This is generally called “tuning” or “model adjusting.” For example, Patent Document 1 has proposed a method wherein specifications of a model are adjusted so that the computed value of the performance model matches the actual measured value, and the computed value is compared with the measured value based on the adjustment results, thereby analyzing the presence or absence of deviation.
Patent Document 1: Japanese Patent Application Laid-Open No. 2001-174366
However, conventional adjustment methods including Patent Document 1 have problems in that they tend to be complicated, as described below, require a large amount of analysis, so taking enormous time and labor. For example, with regard to a certain measurement item, let specifications of an item related to the gas turbine facility be adjusted so that the actual measured value matches the computed value. In this case, with regard to the target measurement item, the computed value matches the actual measured value, however, due to the cause-effect relationships of phenomena occurring inside the gas turbine, the computed value of other measurement items also changes and does not match the actual measured value in some cases. To avoid such ramifications, it is necessary to adjust another item's specifications. However, this adjustment may cause computed values of another measurement items to be changed, and there is a possibility that the computed value may not match the actual measured value. It is not clarified yet how to deal with the successive adjustments of specifications and which criteria are the best to cut down less influential items for putting an end to a barrage of adjustments.
Therefore, to cope with the situations, it is necessary to closely examine the complicated cause-effect relationships among actual phenomena and their influential range, properly understand such overall complicated relationships as which items of specifications is to be adjusted with which measured value it can deviate and as other measured values are to be affected to what degree the adjustment is made, and determine appropriate adjusting procedures. To do so, a large amount of complicated preliminary analysis, time and labor are required.
It is an object of the present invention to provide a gas turbine performance analysis method and a gas turbine performance analysis system which can accurately analyze gas turbine performance by adjusting a small amount of factors without executing complicated procedures.
A gas turbine performance analysis system according to the present invention comprising: a performance computation module for receiving each of actual measured values of an inlet air temperature introduced to compressor, a compressor pressure ratio and a fuel flow rate supplied to a gas turbine in the actual gas turbine and calculating and outputting values of a gas turbine's power output and exhaust temperature exhausted from the gas turbine based on these actual measured values; and a performance estimation module for evaluating performance of the gas turbine based on the deviation between the values of power output and exhaust temperature outputted from the performance computation module and the actual measured values of power output and exhaust temperature; wherein the gas turbine performance analysis system further comprising:
an adjustment module for calculating a corrected value of the fuel flow rate based on the actual measured values of inlet air temperature, compressor pressure ratio and fuel flow rate, and inputting the corrected fuel flow rate value into the performance computation module.
Furthermore, in a gas turbine performance analysis method according to the present invention comprising steps of: inputting each of measured values of an inlet air temperature introduced to compressor, a compressor pressure ratio and a fuel flow rate supplied to a gas turbine in the actual gas turbine into a gas turbine performance computation process, calculating values of a gas turbine's power output and exhaust temperature exhausted from the gas turbine based on these actual measured values in the gas turbine performance computation process, and evaluating performance of the gas turbine in a performance estimation process based on the deviation between the computed values of power output and exhaust temperature and the actual measured values of power output and exhaust temperature; wherein the gas turbine performance analysis method further comprising steps of:
calculating a corrected value of fuel flow rate in an adjustment process based on actual measured values of inlet air temperature, compressor pressure ratio, and fuel flow rate according to a predetermined functional relation, and inputting the corrected fuel flow rate value into the performance computation process for evaluating performance of the gas turbine.
A gas turbine performance analysis system according to the present invention can have an arithmetic module (hereafter, referred to as an adjustment function setting module) which sets a functional relation of the adjustment module so that the deviation of values of power output and exhaust temperature, which were calculated by inputting time-series data of a certain time period about the actual measured values of inlet air temperature, compressor pressure ratio and fuel flow rate or data about a plurality of operating conditions (hereafter, referred to as a plurality of data sets), and the actual measured values is within the predetermined allowable criteria.
The adjustment module executes a correction calculation of a fuel flow rate based on the actual measured values of inlet air temperature, compressor pressure ratio and fuel flow rate according to the predetermined functional relation, replaces the fuel flow rate value measured in the actual equipment with the corrected fuel flow rate value, and inputs the corrected fuel flow rate value into the performance computation module.
According to the present invention, it is possible to accurately analyze gas turbine performance by adjusting only a few factors without executing complicated procedures. According to the present invention, various factors related to the deviation between the actual measured value and the computed value of the performance can be efficiently and simply adjusted in a comprehensive manner by correcting the data of only one item using input information of three items without executing complicated procedures that use a large amount of information. Furthermore, the information necessary for the correction is measurement data of only five items.
In the gas turbine performance calculation, we analyzed factors that cause deviations between the computed value and the actual measured value, and we noted measured deviations in energy input and output balance as one of dominating factors. We then analyzed factors that influence the imbalance and found significantly influential factors. Based on this, we established a model adjustment means for eliminating an error caused by energy imbalance and accordingly we created the present invention. Hereafter, the present invention will be described in detail.
(Analysis of Conventional Methods and Creation of a Model)
To compare with the present invention, first, a conventional gas turbine performance analysis system will be described.
The cycle theory of the conventional analysis system 60 is as described below.
The actual equipment measurement information 1 is input information which is measured values of an actual gas turbine's service conditions and operating state. In addition to the operating conditions including a fuel flow rate being supplied to the gas turbine, the service conditions also include environmental conditions, such as atmospheric temperature, pressure and humidity, which affect gas turbine performance. The measured values of those environmental conditions are not always measured values obtained at a location where actual equipment is installed, but other data such as data provided by a nearby meteorological observatory can also be used instead of data about the actual equipment.
The actual equipment measurement information 1 can be used in two ways: used as input information 2 and used as evaluation information 3. Typical input information 2 includes inlet air temperature of which inlet air is supplied to the compressor of the gas turbine, compressor pressure ratio and fuel flow rate of which fuel is supplied to the gas turbine and is used as service condition information inputted into the performance computation module 7. Typical evaluation information 3 includes power output generated by a generator of the gas turbine and exhaust temperature of which exhaust gas is exhausted from the gas turbine and are used to evaluate the condition of the actual equipment by being compared with the power output and the exhaust temperature calculated by the performance computation module 7.
The input means 5 receives input information 2 included in the actual equipment measurement information 1 as electronic data from the actual gas turbine or a data server connected to the actual gas turbine via a cable or wireless communication line and then inputs the information into the performance computation module 7. Specifically, input information 2 is received by the information receiving apparatus via a communication circuit, the received information is converted into an input format compatible with the performance computation module 7 by the information conversion means, and then the information is inputted. Input information 2 is separated from the actual equipment measurement information 1 at an arbitrary stage during the period prior to being received by the input means 5 till the information is inputted into the performance computation module 7.
Furthermore, in the same manner, the input means 5 receives evaluation information 3 included in the actual equipment measurement information 1 as electronic data from the actual gas turbine or a data server connected to the actual gas turbine via a cable or wireless communication line and then inputs the information into the performance estimation module 9. Specifically, evaluation information 3 is received by the information receiving apparatus via a communication circuit, the received information is converted into an input format compatible with the performance computation module 9 by the information conversion means, and then the information is inputted. Evaluation information 3 is separated from the actual equipment measurement information 1 at an arbitrary stage during the period prior to being received by the input means 5 till the information is inputted into the performance estimation module 9.
The performance computation module 7 receives information 2, which is the actual measured value to be used for performance computation, from the input means 5, computes expected performance of the gas turbine, and outputs the computation result information 8. The computation result information 8 typically includes power output and exhaust temperature. The performance computation module 7 implements known computation methods, such as computation of thermodynamical or aerodynamical cycles or an artificial-intelligence algorithm including a neural network, to be used as a computer program or a dedicated computation module.
Moreover, in the embodiment of the present invention, a device described as a “module” indicates, as an implemented means, a known computation means realized by an electronic means, such as computer programs, arithmetic chips, or dedicated circuits. Furthermore, the device can operate as a single module, and it can also be one component of a plurality of circuits that configure a program or a module. For example, the performance computation module 7 and the performance estimation module 9, described later, have different module names to clearly describe functional configurations, however, they do not have to be two different pieces of hardware and actually, in general, they are implemented as sub-modules that configure one analysis program.
Computation result information 8 calculated by the performance computation module 7 is outputted as electronic data and inputted into the performance estimation module 9.
The performance estimation module 9 evaluates the values of power output and exhaust temperature in the computation result information 8 based on the deviation from information 3 which is the actual measured value used for evaluation. For example, with regard to the power output or the exhaust temperature, if deviation between the value of the computation result information 8 and the value of the information 3 which is the actual measured value used for evaluation exceeds a predetermined threshold, information to trigger an alarm is created. Alternatively, by reading past storage results stored in the storage means 10 described next, a determination of the presence or absence of anomaly based on time-series change, a quantification of the degree of anomaly, or a computation of change in performance can also be executed. In the present invention, in some cases, various operations of the performance estimation module 9 are collectively called the “comparative evaluation between expectation and measurement,” and the resulting information is collectively called “compared data.”
Compared data of the performance estimation module 9 is stored in the storage means 10. Typical stored information is actual equipment measurement information 1 and computation result information 8. The storage means 10 is a known information storage means, such as hard disks, optical disks, memories and the like. Generally, every time the performance estimation module 9 is executed, compared data is sequentially stored in the storage means 10.
Compared data of the performance estimation module 9 or information stored in the storage means 10 is outputted by the output means 11 to be used as analysis information 13. The output means 11 is a means for outputting information from the analysis system 60 to the outside, and specifically, it is a display screen such as monitors, a file transfer means such as file transfer protocols, a file output means for outputting a data file, an information recording means such as hard disks, or a means for outputting data from an electronic medium to another medium such as printers.
To analyze the effectiveness, accuracy, and problems of the conventional method, an example in which gas turbine performance is actually evaluated is shown in
This indicates that energy input and output are not balanced. In other words, total input energy of inlet air and fuel which are to be inputted into the gas turbine facility is not equivalent to total output energy of power output, exhaust gas and loss which are outputted from the gas turbine facility. That is, if the actual measured values are used as criteria, computation results indicate that the output energy is much larger than the input energy in area A, and the output energy is insufficient in area B. However, in a computation model of the performance computation module 7 in this case, computation is executed so that energy input and output can be balanced. This indicates that the energy input and output are not balanced with regard to the actual measured values, and not in the computation model of the performance computation module 7. And this is the point we noted.
One of possible causes for the deviation of energy balance between input and output with regard to the actual measured values is the exhaust temperature. The exhaust temperature is measured at more than a dozen points in the circumferential direction at the turbine outlet of the gas turbine, and a median or a mean value is represented as the value and used for the performance evaluation and the plant operation control. However, actually, there are complicated factors, such as the temperature decrease from the turbine outlet to the exhaust gas duct, the temperature distribution in the circumferential direction and in the diameter direction on the cross-section where thermocouples are set, and voltecs flow. Therefore, the above-mentioned representative value does not match the thermodynamically balanced true value of exhaust temperature.
However, it is extremely difficult to estimate a true value of the exhaust temperature by considering all of those complicated factors. Accordingly, it is necessary to provide a simple and efficient method that can solve the problems with the deviation of energy balance so that accurate performance evaluation can be achieved.
While considering the above-mentioned complicated phenomena about the exhaust temperature, to find a simple and practical solution, first, we found the dominating factors that affect the deviation of energy balance by conducting thermodynamical analysis and then established a model adjustment method.
First of all, to find dominating factors that affect the deviation of energy balance, as described below, we organized relevant factors along the power cycle of the gas turbine, simplified the factors, and then identified the dominating factors.
First, based on the gas turbine cycle theory, deviation ΔH of the measured energy balance between input and output energy was formed into a model, as shown in equation (1), as a function of three elements: compressor efficiency ηc that affects compressor work, turbine efficiency ηt that affects turbine work, and fuel energy input QF that accounts for most of the energy input. Herein, f represents a function and will do the same hereafter.
ΔH=f(ηc, ηt, QF) Equation (1)
The compressor efficiency and the turbine efficiency on the right-hand side of equation (1) can thermodynamically be calculated based on temperatures and pressures at the each inlet and each outlet of the compressor and the turbine. However, to simplify the equation, we omitted the exit temperature, replaced the turbine's pressure ratio with the compressor's pressure ratio, and assumed the relation between equations (2) and (3). Herein, Tci represents an inlet air temperature, Tti represents a turbine inlet temperature, and φc represents a compressor pressure ratio.
ηc=f(Tci, Φc) Equation (2)
ηt=f(Tti, Φc) Equation (3)
We considered turbine inlet temperature Tti on the right-hand side of equation (3) to be equation (4) based on the energy balance between input and output energy. Herein, Tcd represents a compressor outlet temperature.
Tti=f(Tcd, QF) Equation (4)
Compressor outlet temperature Tcd on the right-hand side of the equation (4) can thermodynamically be shown as equation (5).
Tcd=f(Tci, ηc)| Equation (5)
(1) through (5) give equation (6).
ΔH=f(Tci, Φc, QF) Equation (6)
Thus, we found roughly three dominating factors which affect the deviation of the energy balance between input and output energy: inlet air temperature of the compressor, compressor pressure ratio, and fuel energy.
Next, by using the inlet air temperature, compressor pressure ratio, and fuel energy which are the dominating factors thus detected, we studied a method of correcting the deviation of energy balance. To correct the energy imbalance, we specifically studied and examined which factor should be corrected and decided to correct the fuel flow rate.
The reason of this is as described below. To correct energy imbalance, it is necessary to correct either the inside of the main body of the model or the energy input or output. Since it is a given fact that energy input and output are balanced in the model, adjusting and altering the inside of the main body would ignore this premise and therefore, such adjustments and alterations would be improper. Furthermore, the power output and exhaust temperature which are outputs are used to analyze degradation and anomaly of the gas turbine performance. Therefore, correcting those outputs is also improper. Accordingly, either the fuel flow rate or the inlet air flow rate both of which are inputs must be corrected. Of those, because the amount of energy of the fuel flow rate is significantly great and influential, we chose the fuel flow rate as an item to be corrected.
Thus, we decided to use the fuel flow rate as an item to be corrected and established a correction formula to correct the deviation of energy balance as shown below. In equation (8), we simplified the function and assumed a linear relation, and in both equations (7) and (8), we replaced the fuel energy, which must be calculated, with the fuel flow rate that can be directly measured.
Herein, Gf denotes a fuel flow rate, Gf_cor denotes a corrected fuel flow rate, γ denotes a correction coefficient, ai(1=1,2,3) denotes a coefficient, and b denotes a constant term.
Based on the above-mentioned analysis,
The actual measured values of the gas turbine, which is to be inputted for gas turbine performance analysis system 4, are followings as shown in
The adjustment module 6 is a computation module in which functions of equations (7) and (8) are implemented. In this module, according to those calculation formulas, computation is executed to correct the actual measured value of the fuel flow rate by using the function of the inlet air temperature of the compressor, compressor pressure ratio and the fuel flow rate. Then, among performance computation input data, the original value of the fuel flow rate is replaced with the corrected value of the fuel flow rate (corrected fuel flow rate) and then the value is inputted into the performance computation module 7.
Herein, specific contents of equations (7) and (8), which are content data of function f of equation (8), coefficients a1 to a3, and the value of constant term b, are predetermined by the adjustment function setting module 12. As shown in
A plurality of data sets 20 are time-series data of a specific period or data about a plurality of different operating conditions with regard to the actual measured values of the gas turbine including inlet air temperature of compressor, compressor pressure ratio, fuel flow rate, power output of generator, and exhaust temperature. Hereafter, a set of actual measured values of the gas turbine including inlet air temperature of compressor, compressor pressure ratio, fuel flow rate, power output of generator, and exhaust temperature measured under a specific condition or at a point in time is called a data set, and sets of values measured under a plurality of conditions or at a plurality of points of time are called a plurality of data sets. Furthermore, the inlet air temperature, compressor pressure ratio, and the fuel flow rate are collectively called input information 2, and the power output and the exhaust temperature are collectively called evaluation information 3.
Moreover, function f shown in the above equation (8), even if it is not formed into an equation, includes an algorithm in which items shown in parentheses on the right-hand side are included as input and items shown in parentheses on the left-hand side are included in output. Therefore, setting the adjustment function in the adjustment function setting module 12 is not limited to the determination of the equation that expresses function f of equation (8) and the constant such as coefficient included in the equation, but, in the broad sense, even if an algorithm is not explicitly formed into an equation, determination of the set value necessary to execute a computation is included. In other words, setting the adjustment function means the overall settings necessary to execute a computation that outputs, when inputting inlet air temperature, compressor pressure ration, and fuel flow rate, the corrected fuel flow rate.
(Processing Flow Executed by the Adjustment Function Setting Module).
To explain operations of the adjustment function setting module 12 in detail,
In the performance computation process 22 which is a part of the loop wherein operation proceeds to the next data set in process 28 (described later), gas turbine performance computation is executed by using a single data set at that point (included in a plurality of data sets 20) as input data. The contents of the performance computation are the same as those of the computation executed by the above-mentioned performance computation module 7 (
The performance computation results obtained in performance computation process 22, which are the computed values of the power output and the exhaust temperature, are compared with the actual measured values included in the corresponding data set, that is, evaluation information 2, and it is determined in determination process 23 whether the deviation between the computed value and the actual measured value is within predetermined allowable criteria.
If the determined result is false, in subsequent revision process 24, the value of the fuel flow rate is revised, and the original fuel value included in the corresponding data set is replaced with the revised fuel flow rate value, and then the performance computation is executed again. The revision of the fuel flow rate and the performance computation by using the revised value are repeatedly executed until the determined result in determination process 23 becomes true.
The predetermined allowable criteria in the above-mentioned determination process 23 indicate, for example, whether the deviation rate or the absolute value of the deviation between the actual measured value and the computed value with regard to power output and exhaust temperature, respectively, is within predetermined upper and lower limits. Furthermore, it is desirable for it to be determined whether the sum of the deviation rates is within predetermined upper and lower limits. The criteria in the determination process 23 are not limited to the above, but it is acceptable for it to be determined whether the degree of deviation between the actual measured value and the computed value with regard to the power output and the exhaust temperature, respectively, is within an allowable range by using the predetermined criteria.
As an example of the fuel flow rate correction methods in the revision process 24, the deviation between the computed value and the actual measured value with regard to the power output and the exhaust temperature, respectively, is converted into an equivalent amount of energy (herein, the exhaust temperature is calculated as enthalpy of exhaust gas), and correction coefficient γ is calculated by using summation ΔH_er of the deviations and the fuel energy input QF as shown in equation (9). And then, by inputting the obtained value into equation (7), it is possible to determine the corrected value of the fuel flow rate. Herein, k denotes a coefficient used to adjust the number of computations and accuracy.
γ=1−k×ΔH—er/QF Equation (9)
Herein, the reason why γ was described as a “revision” coefficient is that the coefficient is different from the “correction” coefficient described related to the equations (7) and (8). That is, the “correction” coefficient described in the equations (7) and (8) is a “correction” coefficient that is used to calculate the “corrected value” of the fuel flow rate by using a predetermined function. On the other hand, the “revision” coefficient is a coefficient related to the “corrected value” of the fuel flow rate which has been obtained by executing revisions by means of iterative computation by seeking convergence so that the computed value matches the actual measured value. Accordingly, different names have been given to those coefficients.
The above-mentioned fuel flow rate correction method is merely an example, and as long as the amount of revision of the fuel flow rate is obtained to detect the value that satisfies the determination criteria in determination process 23, other methods, such as a known optimization calculation and a convergence calculation method can be used. Alternatively, instead of providing a determination with regard to an individual data set as shown in the determination process 23, contents of the functions of the above equations (7) and (8) can be set according to the criteria which judge whether an index value showing a deviation in the all data sets, such as a mean value or a maximum value of the deviation between the computed power output and exhaust temperature and the actual measured values with regard to a plurality of data sets, is within a predetermined allowable range. In other words, optimization or convergence calculation method can be used wherein an equation of a function or an undetermined constant is determined as stated above, or a set value for operating an algorithm is determined.
When the determined result in the determination process 23 becomes true as the result of the revision of the fuel flow rate and the performance computation by using the revised value, the corrected value of the fuel flow rate and the value of correction coefficient γ are outputted by output process 25 and sent to data storage process 26. At this time, values of inlet air temperature, compressor pressure ratio, and original fuel flow rate of each data set are also sent and stored in the data storage process 26.
Operations from process 22 to process 25 are repeatedly controlled by data-set loop-end determination process 27 and process 28 wherein operation proceeds to the next data set until processing of all of the plurality of data sets 20 is completed.
When repeated operations of all data sets have been completed, in process 29 which is the latter half part, the function of equation (8) is identified by fitting based on the correction coefficient γ of all data sets and the data about the corresponding inlet air temperature, compressor pressure ratio, and fuel flow rate stored in the data storage process 26. In the simplest fitting method, the least squares regression is applied to the linear function of the second equation of equation (8) and the equation's coefficient and constant term are determined. However, an equation of the function is not limited to the linear function, and any function is available as long as correction coefficient γ is expressed as a function of the inlet air temperature, compressor pressure ratio, and the fuel flow rate. Furthermore, the fitting method is not limited to the linear least squares regression, and it can be a known optimization or a regression analysis method wherein an undetermined constant of the function can be determined so that the deviation between the computed value and the actual measured value is small.
Herein, to simplify the explanation of the function's equation, a correction coefficient is provided on the left-hand side of equation (8). However, it goes without saying that the fuel flow rate can be directly computed by providing a corrected fuel flow rate on the left-hand side instead of using a correction coefficient. In this case, the function setting procedure is executed by the same procedure as mentioned above in which a correction coefficient is provided on the left-hand side.
(Schematic Diagram of the Adjustment Function Setting Module)
At least three elements described below are implemented in the backward calculation module 31. Firstly, there is implemented a convergence calculation loop comprising a function to activate and execute the performance computation process 22 (performance computation module 7) to calculate gas turbine performance in order to activate the performance computation module 7 so that the corrected value of the fuel flow rate or the correction coefficient can be calculated and determined, determination process 23 to determine computation result, and revision process 24 to revise the fuel flow rate if the obtained determined result is false. Secondary, output process 25 is implemented wherein the corrected value of the fuel flow rate or the correction coefficient calculated by means of the convergence calculation is outputted along with values of inlet air temperature, compressor pressure ratio, and original fuel flow rate of each corresponding data set. Thirdly, the loop control module (processes 27, 28) is implemented which repeatedly executes convergence calculations of all of the plurality of data sets 20.
The performance computation module 7 is an implementing module which executes performance computation in the performance computation process 22, and as described in the explanation of
The data retention means 32 is a means for executing the data storage process 26 wherein values of correction coefficient γ of all data sets calculated by the backward calculation module 31 are stored along with values of the inlet air temperature, compressor pressure ratio, and original fuel flow rate of each corresponding data set. The data retention means 32 is typically implemented as a temporary storage area such as a computer memory, however, it is not limited to it and can be a medium which electronically stores data.
The adjustment function identification module 33 implements an arithmetic procedure of process 29 wherein adjustment function setting information is identified. A specific set value for the undetermined constant of the function of equation (8) is determined by computation executed in process 29 and outputted as adjustment function setting information 21.
(Performance Analysis Execution Method by Using a Configured Adjustment Module)
In the gas turbine performance analysis system according to the embodiment of the present invention, adjustment function setting information 21 (
The processing flow comprises a correction computation process 41 in which actual measured values of the actual equipment's inlet air temperature Ta, compressor pressure ratio P, fuel flow rate F, power output E, and exhaust temperature Te (hereafter, it is called an actual equipment's data set 40) are inputted and the corrected value of the fuel flow rate is calculated by using the adjustment module 6 configured as explained in
In the correction computation process 41, actual measured values of the actual equipment's inlet air temperature, compressor pressure ratio, and fuel flow rate included in the actual equipment's data set are inputted into the adjustment module 6 configured as explained in
In the subsequent performance computation process 22, among the actual equipment's data set's performance computation input data (inlet air temperature, compressor pressure ratio, fuel flow rate), the value of the fuel flow rate is replaced with the value corrected in the correction computation process 41 and inputted. Specifically, the performance computation module 7 is executed based on the input data, gas turbine performance is computed, and items including power output and exhaust temperature are calculated. Moreover, the computation results will become computation result information 8 described in the explanation about
Thus, the computed values of power output and exhaust temperature obtained in the performance computation process 22 are compared in evaluation process 43 with each actual measured value, that is, the actual measured values of power output and exhaust temperature for evaluation included in the actual equipment's data set 40 and evaluated. Specifically, this evaluation is executed by the above-mentioned performance estimation module 9 (
Result information evaluated in the evaluation process 43 or information provided for evaluation, that is, information of each item of the actual equipment's data set 40 and computation result information 8 (
The result information thus evaluated in the evaluation process 43 or information stored in the data storage process 44 is outputted in the subsequent result output process 45. Specifically, data is outputted by the above-mentioned output means 11 (
Computation in correction computation process 41 and performance computation process 22, subsequent evaluation of the computed values and the actual measured values (evaluation process 43), data storage (data storage process 44), and result output (result output process 45) are repeatedly executed with regard to the actual equipment's data set 40 of a certain period or the overall data set. The repetition of operations is controlled by repeating process 47 in which operation proceeds to the next data set until the data becomes true in end determination process 46, that is, until the end of the operations is determined. Herein, end determination means the determination of whether computation for analysis is to be continued or not, for example, by judging whether the computation of all data of the actual equipment's data set 40 has been finished or whether a command to quit analysis has been entered by a user.
As the result of the repeated operations that continue until end determination is made, evaluation result information stored in the data storage process 44 is outputted in result output process 48. Specifically, data is outputted by the above-mentioned output means 11 (
Moreover, with regard to the output of results, the above-mentioned combination of result output processes 45 and 48 is an example. The difference between those two processes is the output timing. Aside from whether the amount of processed data is the whole data sets of the actual equipment or a portion of the data set, output data is the same, and the output timing and contents of data can be either one of them. For example, when all data sets of the actual equipment of a specific period are batch-processed repeatedly, the results are outputted at the timing of result output process 48. Furthermore, when the system is automatically activated periodically, for example, daily, and data sets created after previous activation are repeatedly processed, the results are outputted in the same manner. On the contrary, for example, when the system is connected online to the actual equipment measurement system and evaluation is conducted in real time, results, specifically an alarm, are outputted at the timing of result output process 45.
To verify the improved performance evaluation accuracy of the method of the present invention, a model adjustment method according to the present invention was applied to the performance evaluation of the actual equipment and accuracy was evaluated. Table 1 shows the model adjustment methods which were compared one another. As an example of the conventional method, a simple method (constant rate correction type) is shown wherein correction is made by multiplying a correction coefficient of the constant value. Furthermore, with regard to the method of the present invention, a method (simplified type) is also shown wherein uninfluential factors are excluded and correction is made by using only inlet air temperature. The method of the present invention wherein all of is called a full type to distinguish it from the simplified type. By using actual equipment's operation data, model adjustment was conducted by each method, and by using the results for the performance computation executed by the procedure shown in
At this time, the constant rate correction type model adjustment was executed by using the constant value for the correction coefficient γ of the above-mentioned equation (7) so that the error is minimized in the entire evaluation period. The simplified type model adjustment was executed by using an equation, which excludes the compressor pressure ratio and the fuel flow rate from the above-mentioned equation (8), as a model equation according to the same procedure as the full type model adjustment. The full type model adjustment was executed according to the procedure described in the explanation about
As stated above, it was found out that the simplified type can improve accuracy more than the conventional type, and the full type can further improve the accuracy. The difference between the simplified type and the full type appeared to be a difference not only in the degree of improvement of power output errors but also a difference in the degree of reduction in exhaust temperature errors. This may be the advantage of the full type because the full type uses all of the factors that were found to correct the deviation in energy balance.
Moreover, in the above example, the simplified type uses only inlet air temperature to correct fuel flow rate. However, if both inlet air temperature and compressor pressure ratio, or both inlet air temperature and fuel flow rate are to be used to correct the fuel flow rate, it is expected that the accuracy will be improved to become almost close to the accuracy obtained by the full type.
In the Table 1, symbol “◯” means that the corresponding parameter is considered in the model adjustment method.
(Comparison Between the Method of the Present Invention and the Alternative)
As stated above, the present invention proposes to correct the fuel flow rate in order to correct the deviation in energy balance between input and output energy. On the other hand, the main cause of the deviation in energy balance is considered to be the deviation between the measurement value of the exhaust temperature and the theoretical value. Therefore, there could be an approach to correct the exhaust temperature. Nevertheless, the reason why fuel flow rate is corrected is the practical advantages described below.
That is, if the conventional method shown in
On the other hand, in the fuel flow rate correction procedure according to the present invention, by inputting the corrected fuel flow rate into the performance computation module, not only the computed exhaust temperature but also power output changes. For this reason, by simply configuring a single adjustment module 6 (equations (7) and (8)) by using the adjustment function setting module 12, the model adjustment can be executed so that errors of both power output and exhaust temperature can be small. The method according to the present invention is created such that correction can be made by simply correcting only the fuel flow rate among various error factors such as deviation in energy balance between input and output energy and the deviation between the gas turbine performance characteristics and the actual measured values. Therefore, the alteration or adjustment of the main body of the performance computation module 7, which is the center of performance analysis, can be minimized and the adjustment and operation of the analysis system can be executed easily and efficiently.
According to the embodiment of the present invention, it is possible to adjust complicated factors related to the deviation between the actual measured value and the computed value by correcting only the fuel flow rate (adjustment module 6) using three input items (inlet air temperature, compressor pressure ratio, and fuel flow rate). Furthermore, the adjustment function setting module 12 to configure the adjustment module 6 can also be executed by using data of only five items which include power output and exhaust temperature in addition to the above three items. The reason why such complicated error factors can be adjusted by using the data of only a few items is that the entire gas turbine cycle was thermodynamically analyzed in a comprehensive manner and efforts were made to retrieve dominating factors by simplification (equations (1) to (6)).
| Number | Date | Country | Kind |
|---|---|---|---|
| 2007-214932 | Aug 2007 | JP | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 7020595 | Adibhatla et al. | Mar 2006 | B1 |
| 20080168016 | Sekiai | Jul 2008 | A1 |
| Number | Date | Country |
|---|---|---|
| 2001-174366 | Jun 2001 | JP |
| 2006-010229 | Jan 2006 | JP |
| 2007-231804 | Sep 2007 | JP |
| 2007-264796 | Oct 2007 | JP |
| 2007-265212 | Oct 2007 | JP |
| Number | Date | Country | |
|---|---|---|---|
| 20090055105 A1 | Feb 2009 | US |
