JP3392526B2 - Equipment maintenance management support equipment - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to high temperature equipment such as a gas turbine applied to a thermal power plant,
It relates to a maintenance management support device for equipment that monitors the need for repairs against deterioration or damage of components due to use, and especially optimizes repair costs or parts replacement costs based on theoretical prediction of deterioration and damage. Device maintenance management support device.

[0002]

2. Description of the Related Art In recent years, from the viewpoint of effective use of energy and protection of the global environment, in order to improve efficiency in power plants and the like, the temperature of motors such as gas turbines is rapidly increasing and the size of motors is increasing. Conditions are becoming increasingly severe. In particular, the main high-temperature members exposed to high-temperature fluid suffer damage such as deterioration, cracks, and deformation of the metallographic structure within a relatively short period of time, so it is critical to properly judge repair or replacement based on regular inspections. Indispensable for ensuring stable operation (Reference: Yoshioka, "Deterioration behavior of gas turbine materials" Japan Society of Mechanical Engineers No.920-6 Workshop Material: Structural strength design in high temperature crisis-From development of new material to life evaluation -1992.1 Tokyo).

Conventionally, the repair judgment of the structural member of the equipment is performed by setting a repair allowable value based on the design information of the parts and past experience information, and comparing this allowable value with the deterioration / damage measurement amount. Is common. Replacement of parts is judged when repair is impossible or after comparison of repair cost and new part replacement cost. Inspection intervals, repair standards, and parts replacement standards vary depending on the usage environment, fuel type, equipment start / stop patterns, etc. The same maintenance management measures are applied to.

[0004]

However, there are many forms of deterioration and damage that occur in each part, and the degree of progress is not uniform in many cases. In group management, repair costs including replacement parts costs are high. Can cause In addition, it is not preferable in terms of maintenance budget planning that the parts replacement times of the units that started operation at the same time coincide, so it is necessary to equalize the repair cost occurrence time as much as possible.

A technique for quantitatively predicting the deterioration / damage of a gas turbine member is important as a basic technique for providing a rational basis for repair or replacement of parts. So far, simulation technology, trend analysis technology, etc. Deterioration of parts
Technology has been developed according to the characteristics of damage (eg:
Japanese Patent Application No. 4-31327, Saito, "Non-destructive evaluation of long-term deterioration of high temperature equipment for thermal power generation and life diagnosis", Japan Society of Mechanical Engineers No.930
-30 Lecture Summary P1-10), its use for optimal maintenance is just the beginning.

The problem to be solved by the present invention is to reduce the repair / replacement costs of structural members such as prime movers whose operating conditions are becoming severer with the improvement of efficiency as much as possible, and to level the timing of occurrence. It is to establish a rational method.

That is, an object of the present invention is to provide a maintenance management method which optimizes an operation pattern and a repair / part replacement time while ensuring a predetermined operation time, and which is the most efficient and low cost, for each part, each device or It is to be provided on a plant basis.

[0008]

The invention according to claim 1 is a deterioration / damage measuring means for measuring the degree of deterioration or the amount of damage occurring to a structural member of a device, and operation, repair, deterioration of the device.
Operation history recording means for recording damage measurement and other history,
Data management means for managing data obtained from each of these means, operation condition setting means for setting future expected operation conditions of the equipment, data input from the data management means and operation condition setting means, and the deterioration・ Calculation based on the expected operating conditions with damage amount data as the initial value,
The structure and the deterioration and damage amount prediction means for predicting the progress of deterioration and damage of members, the repair time of the structural member on the basis of the predicted deterioration and damage amount thereby obtained, the repair mass and complement
Calculate the repair period and compare it with the repair amount and the repair period.
Calculation of repair cost including sum of parts
The present invention is characterized in that a repair cost and the like calculating means for judging the strike height and the repair cost occurrence time and a display means for outputting the calculation result are provided.

The invention of claim 2 is based on deterioration / damage measurement history data in place of or in addition to the deterioration / damage amount prediction means and repair cost calculation means in the equipment maintenance management support apparatus of claim 1. Degradation / damage amount tendency analysis means for analyzing the trend of deterioration / damage amount, Deterioration / damage amount tendency prediction means for predicting the trend of deterioration / damage amount transition of structural members based on future expected operating conditions, and its deterioration - amount of damage on the basis of transition tendency repair timing of the structural member, repair amount, predicts the transition tendency of the repair period and repair costs, complement
The present invention is characterized by comprising a trend predicting means for repair cost and the like for judging whether the repair cost is high or low and when the repair cost occurs .

According to a third aspect of the present invention, deterioration / damage amount prediction means, deterioration / damage amount tendency analysis means, repair cost calculation means, deterioration / damage tendency prediction in the equipment maintenance management support apparatus according to the first or second aspect. Deterioration / damage occurrence probability calculation that complements the change tendency of deterioration / damage occurrence probability based on discrete deterioration / damage data obtained for each inspection hand
Stage and, deterioration and damage occurrence probability prediction means for predicting the change trend of the future of probability based on the changing trend of deterioration and damage occurrence probability
And calculate the expected value of the repair cost,
Expected repair cost calculation to determine when repair costs occur
It is characterized in that it is provided with a fixing means .

The invention of claim 4 is an alternative to the deterioration / damage amount predicting means in the equipment maintenance management support device of claim 1.
E. The temperature / stress estimating means for inversely estimating the temperature and stress conditions of the structural member based on the comparison between the deterioration / damage amount history data and the predicted deterioration / damage estimated amount, and the current temperature / stress estimated value based on the temperature / stress estimated value A deterioration / damage amount predicting means for predicting the progress of deterioration or damage amount of the structural member corresponding to future operating conditions, using the deterioration / damage measurement data as an initial value, is provided.

According to a fifth aspect of the present invention, instead of the deterioration / damage amount predicting means and the repair cost calculating means in the equipment maintenance management support apparatus according to the first aspect, a combination of a future operation pattern and a repair time or period is used. Deterioration / damage simulation means for predicting deterioration / damage according to it, adaptive operation condition determination means for selecting an assumed operation pattern that matches the repair time or period in the deterioration / damage simulation prediction result, and repair cost for the selected operation pattern Perform other calculations to determine whether the repair cost is high or low and the repair cost is
It is characterized in that it is provided with a repair cost calculating means for judging the birth time .

According to a sixth aspect of the present invention, in the maintenance management support apparatus for the device according to the first aspect, the device is configured to have a plurality of parts having different types of deterioration or damage or different degrees of progress, and the operation history record The means and the data management means are capable of recording the history and managing the data for each of the parts, and based on the history data of the past operation, use, deterioration or damage obtained from the data management means, For the deterioration / damage prediction model setting means for determining the constants of the model necessary for the deterioration / damage prediction according to the parts and the deterioration / damage prediction model for each part set by the deterioration / damage prediction model setting means, The deterioration / damage state at the time of this inspection obtained from the data management means is set as the initial state, and from this initial state the operation condition setting means sets Using the pattern, the deterioration / damage amount prediction means that predicts the deterioration / damage state after the next inspection and judges repair or replacement at the time of inspection before reaching the allowable value, and this deterioration / damage amount prediction means If it is determined that a part will be repaired or replaced, the repair cost or replacement part cost will be calculated for each part unit, and the repair cost will be calculated.
It is characterized in that it is provided with a unit repair cost calculation means for judging whether the cost is high or low and the repair cost occurrence time .

According to a seventh aspect of the present invention, in the equipment maintenance management support apparatus according to the sixth aspect, the equipment is a plant for power generation or the like, and the deterioration / damage amount prediction means is all units constituting the plant or It has a history database of operation, use, deterioration / damage configured for all parts, and replaces the unit repair cost etc. calculation method, and repair cost according to the combination of operation patterns of each unit of the plant and its elapsed time Change is required, and repair costs are high and low.
And a plan repair cost calculating means for determining the leveling level of the repair cost generation time .

[0015]

According to the first aspect of the invention, for example, the repair cost is calculated by changing the operating condition within a predetermined condition, and the level of the repair cost and the time when the repair cost occurs are determined and displayed.

According to the second aspect of the present invention, for example, the repair cost in the case of repairing the existing part without replacement and the part cost and the repair cost of introducing the new part are calculated and compared,
Displays whether the repair cost is high or low and when the repair cost occurs.

According to the third aspect of the present invention, for example, the operating condition is changed under a predetermined constraint condition to calculate the expected repair cost value, and the level of the repair cost and the timing of occurrence of the repair cost are determined and displayed.

According to the fourth aspect of the present invention, for example, the operating condition is changed under a predetermined constraint condition to calculate the repair cost, and the level of the repair cost and the time when the repair cost is generated are determined and displayed.

According to the fifth aspect of the invention, for example, an assumed operation pattern that matches the repair timing / period in the deterioration / damage simulation prediction result is selected, the repair cost is calculated for the selected operation pattern, and the repair cost is low and the repair cost is low. The time of occurrence is determined and displayed.

According to the sixth aspect of the present invention, for example, deterioration / damage is predicted according to a combination of a future operation pattern and repair time / period, and the repair time / period in the deterioration / damage prediction result is estimated.
Select an expected operation pattern that matches the period, calculate the repair cost for the selected operation pattern, calculate the parts cost and repair cost associated with new product introduction, and repair the entire plant due to differences in operation patterns and parts replacement The high and low cost and the time when the repair cost occurs are judged and displayed.

According to the seventh aspect of the invention, for example, deterioration / damage is predicted according to a combination of a future operation pattern and a repair time / period, a repair cost is calculated for an operation pattern matching the repair time / period, and a new product The parts cost and the repair cost associated with the input are calculated, and the difference between the operation patterns, the level of the repair cost associated with the parts replacement, and the leveling level of the repair cost occurrence are determined and displayed.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a device maintenance management support apparatus according to the present invention will be described below with reference to the drawings.

Embodiment 1 (FIGS. 1 to 4) This embodiment corresponds to the invention of claim 1. Figure 1
Is a schematic configuration diagram, and FIGS. 2 to 4 are operation explanatory diagrams.

As shown in FIG. 1, the equipment maintenance management support apparatus according to the present embodiment has a deterioration / damage measuring means 1 for measuring the degree of deterioration or the amount of damage that occurs in the structural members of the equipment, and a calculation processing of measurement data. Data converting means 2 for converting data for use, operation history recording means 3 for recording the history of operation, repair, deterioration / damage measurement, etc. of equipment, and storage / management of data obtained from these means 1, 2, 3 Data management means 4 for
Data is input from the operating condition setting means 5 for setting the future expected operating conditions of the equipment, the data managing means 4 and the operating condition setting means 5, and calculation is performed based on the expected operating conditions with the deterioration / damage amount data as an initial value. Deterioration / damage amount predicting means 6 for predicting the progress of deterioration / damage amount of a structural member, and judging whether repair is possible or not based on the predicted deterioration / damage amount obtained thereby, the repair time of the structural member, the repair amount, It comprises a repair cost calculation means 7 for calculating a repair period, a repair cost, etc., and a display means for outputting the calculation result.

Then, as shown in FIG. 2, in the deterioration / damage measuring means 1, deterioration of the metal structure found in a moving blade material, a stationary blade material or a sintered equipment material as a structural member at the time of inspecting equipment such as a gas turbine. Damage such as cracks, deformation, voids, and peeling is measured by nondestructive measurement methods such as image input, replica observation, hardness measurement, dimension measurement, ultrasonic measurement, electromagnetic measurement, and polarization measurement.

Here, the deterioration of the metal structure is shown in FIG.
As shown in the schematic diagrams of microstructures in (a) to (c), in a gas turbine superalloy in which high-temperature strength is improved by complex precipitates, the precipitates become coarse during use at high temperatures. Is a phenomenon in which the properties relating to material life such as creep strength and toughness change from the state during manufacturing.

The microstructure is transferred by a replica,
By image-processing the microscope observation image, it is possible to quantify parameters such as area ratio, volume ratio, and number density. In addition, a method such as a polarization method can be used as a method of detecting the degree of deterioration from the dissolution current of precipitates by an electrochemical method (reference: Saito, “Nondestructive evaluation of aging deterioration of high temperature equipment for thermal power generation and life "Diagnosis" The Japan Society of Mechanical Engineers No.930-30 Lecture Summary P1-10).

Creep voids, which are damages caused by the growth of minute voids leading to cracks, can be measured by the same method. However, as a means for nondestructively detecting what has occurred inside a member, a method such as an ultrasonic method is used. Can be applied. Although cracks are mainly observed by visual observation, accurate input can be performed by using an image input means.

These pieces of information are converted into digital data which can be numerically calculated by the data conversion means 2 and transmitted to the data management means 4.

In the operation history recording means 3, for example, as shown in FIG.
As shown in (a), (b), and (c), transition data of load, rotation speed, or gas temperature is monitored on the control panel. And, the number of past start and stop, start and stop pattern,
It is transmitted to the data management means 4 as digital data regarding the steady operation time and the inspection period. Further, in addition to this, by manually inputting, history information such as the presence / absence of repair / replacement of parts and the transition of equipment in which parts are incorporated is also transmitted to the data management means 4.

The data management means 4 stores these data in the "operation history table", "usage history table",
Filed as "deterioration / damage table" and "reference information table" and searched by "equipment No.", "part No.", "part No.", "inspection No.", etc. to correlate each data Build a relational database that can take

Next, the deterioration / damage amount predicting means 6 is checked this time based on the past tendency of deterioration / damage of the structural member or the physical rule of deterioration / damage development and the operating conditions such as temperature / stress. Based on the deterioration / damage measurement amount obtained by, the deterioration / damage prediction corresponding to the future expected operating conditions is performed.

As a deterioration / damage prediction method, Japanese Patent Application No.
The methods such as the trend analysis method and the simulation analysis method proposed in -312327 are used. If it is predicted that the amount of deterioration / damage will reach a predetermined allowable value by the next inspection, it is determined that repair will be performed this time, and the state after the current repair will be used as the initial value to determine the amount of deterioration / damage until the next inspection. Predict.

The repair cost calculation means 7 calculates the repair amount R _i from the deterioration / damage amount D _i at the repair time i predicted by the deterioration / damage amount prediction means 6. That is,

R _i = f (D _i ) (Formula 1) Further, the repair period L _i is calculated from the repair amount. That is,

L _i = g (R _i ) (Equation 2) The repair cost is the sum of the fixed cost portion C _o , the sum of the repair amount R _i and the portion proportional to the repair period L _i , and the replacement part cost C _r . Calculated as

[0035]

[Number 3] _{C i = C o + αR i} + βL i + C r ( Equation 3) where, _{C o} is an exploded costs, transportation costs, inspection costs, is the sum of the re-assembly costs, α is, repair material costs, Cost coefficient related to repair work cost, β is loss cost coefficient due to operation stop.

(Equation 3) is the following equation for one group of parts.

[0037]

[Equation 4]

The above-mentioned repair cost calculation procedure is repeatedly calculated by changing the operating conditions in the operating condition setting means 5 within a range satisfying a predetermined constraint condition (for example, required power generation time). As the first of the future operating conditions, the average operating time and the number of start-stops that are easily calculated from the past operating history in the data management means 4 are used. As other operation patterns, those changed from this standard pattern or those artificially set are used.

The above calculation results are displayed on the display means 8 as a trend diagram of future repair costs so that the level of repair costs and the timing of occurrence of repair costs can be distinguished at a glance.

According to the configuration of the first embodiment described above, it is possible to accurately predict the maintenance management cost of parts and the timing of occurrence thereof according to the deterioration / damage state based on the deterioration / damage history data of the equipment members, and to whom it can be calculated. However, there is an effect that an accurate cost evaluation can be easily performed.

[0041] Example 2 (FIGS. 5-10) This example relates to the invention of claim 2, applied to the repair cost trend prediction of thermal fatigue cracking damage of the gas turbine stator vane. 5 is a schematic configuration diagram of the present embodiment, FIG. 6 is a diagram showing a specific configuration example, and FIGS. 8 to 10 are operation explanation diagrams.

As shown in FIG. 5, the equipment maintenance management apparatus of the present embodiment is replaced with the deterioration / damage amount prediction means 6 and the repair cost calculation means 7 in the maintenance management support apparatus of the first embodiment (or). In addition), analysis of the trend of deterioration / damage amount based on deterioration / damage measurement history data
Damage amount tendency analysis means 9 and deterioration of structural members based on future expected operating conditions
The damage amount tendency predicting means 10 and the repair cost etc. trend predicting means 11 for predicting the change tendency of the structural member repair period, the repair amount, the repair period and the repair cost based on the change tendency of the deterioration / damage amount are provided. It is configured.

In this embodiment, the deterioration / damage measuring means 1
Is used as the crack image scan input device 12, as shown in FIG. This crack image scan input device 1
2 has a measuring head 14 that moves along a component 13 to be inspected. As shown in FIG. 2B, the measuring head 14 has a position measuring roller 15 and a CCD image pickup device 16.

The data converting means 2 is composed of an AD converter 17 and an image processing device 18, as shown in FIG. 6 (a) .

The operation history recording means 3 is a computer and software (not shown) having a communication function, which is connected to a monitor device mounted on the central control panel of the plant by a data communication cable. The data management means 4 is composed of a magnetic disk, a control device therefor, and relational database software. Operating condition setting means 5, deterioration
Damage amount tendency analysis means 9, deterioration / damage amount tendency prediction means 10
The repair cost trend predicting means 11 is a computer and software. The display means 8 is a display device such as a CRT.

In the second embodiment having such a structure, first, the measurement head 14 of the crack image scan input device 12 shown in FIG. 6 as an input unit is moved along the shape of the target component 13 to crack. The position and the dimension of are input to the image processing device 18 through the AD converter 17.

FIG. 7 shows the processing operation of the input image. That is, after the noise a is erased from the input image A and the thinning process is performed, the existing regions A1, A2, A of the crack b are present.
3 is divided, and the cracks are numbered as 1, 2, 3, for example.
The length of each crack is represented by the length of a straight line connecting both ends of the crack, and the angle θ between this straight line and the reference line is defined as the crack angle,
Further, the circumscribed rectangles C1, C2, C3 of the crack are set and the area thereof is calculated. These calculated values are stored in a file in the form of a "crack table" in the data management means 4 of FIG. 5, as shown in FIG. 8 (c).

From the operation history recording means 3 of FIG.
The changes in rotation speed and gas temperature are transmitted as data,
In the data management means 4, as shown in FIG.
The file is saved as a "driving history table". Usage history data such as other parts repair / replacement and transition of the device in which the parts are incorporated are stored in a file as a “usage history table” shown in FIG. 8B.

Next, the operation of the deterioration / damage tendency analysis means 9 will be described with reference to FIG. The distribution of the multi-crack length l generated in the stationary blade is well approximated by the lognormal distribution. The average and dispersion of the crack lengths tend to increase as the number of times of starting and stopping increases as shown in the figure, and the number of cracks also increases at the same time.
These are expressed by the following relational expressions.

[0050]

(5) μ = A1N + A2 (Equation 5)

## EQU6 ## V = B1 N + B2 (Equation 6)

N = C1N (Equation 7) where μ: average of logarithm of crack length, V: variance of logarithm of crack length, n: number of cracks, N: number of start / stop, A1, A
2, B1, B2, C1: constants (constant values of Equation 5) to (Equation 7) are calculated by applying optimal approximation to the operation history data and the crack data from the data managing means 4.

From (Equation 5) to (Equation 7), the total crack length is calculated as the average number of true cracks × the number as a crack damage parameter. The prediction width can also be calculated using the variance value. Assuming that there are fluctuations in the number of start and stop until the next regular inspection, we predict the place with the highest probability of being combined with the distribution of total crack length. Contour line d in FIG. 9
Shows equi-probability lines of the number of start-stops and crack damage parameters at the next inspection. On the other hand, the number of start and stop is specified,
If fluctuations are not assumed, only the crack damage parameter distribution is considered.

Further, with reference to FIG. 10, the operation from the operating condition setting means 5 and the deterioration / damage amount tendency prediction means 10 to the display means 8 will be described.

The deterioration / damage amount tendency predicting means 10 predicts the deterioration / damage amount at the time of the next inspection by the above-mentioned tendency analysis method based on the measured deterioration / damage amount. If this predicted value does not exceed the predetermined allowable value, the prediction at the time of inspection is further performed one by one, and this calculation is repeated until the predicted value exceeds the allowable value. To determine. In the case of repair judgment, deterioration / damage is predicted with the state after repair as the initial value.

The repair cost / trend predicting means 11 determines the repair amount Ri at the repair time i as j having a length equal to or greater than a certain value calculated by the (expression 5) to (expression 7) in the deterioration / damage amount predicting means 6. The product of the area S _{j of the} rectangle surrounding the th crack and the plate thickness b _j is multiplied by a predetermined correction coefficient η to calculate the sum.

[0055]

Equation 8] calculates the _{R i = Σ η · S j} (N, t) · b j maintenance period _{L i} from equation (8) further repair amount. Repair period is welded man and crack removal steps in the repair weld, R _i
Proportional to. That is,

Equation 9] L _i = ζR _i (Equation 9) repair cost per component, the fixed costs portion _{C o,} repairing amount of the portion proportional to _{R i} and spare period _{L i} of the sum and replacement parts costs _{C r} The sum is calculated by (Equation 10).

[0056]

[Equation 10] For one group of parts assembled in one unit of gas turbine,

[Equation 11] On the other hand, assuming that a new part has been introduced at the time of regular inspection, the deterioration / damage trend predicting means 10 predicts the subsequent development of deterioration / damage to the same expected operating condition by using the above-mentioned method, and repairs the repair. Repair amount in the cost trend prediction means 11,
Calculate repair time and repair cost.

The above calculation results are displayed on the display means 8 so that whether the repair cost is high or low and the time when the repair cost is generated can be seen at a glance when the new part is not supplied and when the new part is supplied.

According to the second embodiment having the above configuration, the maintenance cost of parts and the timing of occurrence can be accurately predicted according to the deterioration / damage state based on the deterioration / damage history data of the equipment members, and the new Compared with the case of replacing with parts,
The effect is that anyone can easily perform accurate cost evaluation.

Embodiment 3 (FIGS. 11 to 18) In this embodiment, the invention of claim 3 is applied to thermal fatigue crack damage of a mixed air hole weld portion of a gas turbine combustor liner. FIG. 11 is a schematic configuration diagram of this embodiment, FIG. 12 is a diagram showing a specific configuration example, and FIGS. 13 to 18 are operation explanation diagrams.

In the present embodiment, the deterioration / damage amount predicting means 6, the deterioration / damage amount tendency analyzing means 9, the repair cost calculating means 7, and the deterioration / damage in the equipment maintenance management support apparatus described in the above-mentioned Embodiment 1 or 2 are used. Instead of (or in addition to) the tendency predicting means 10 and the tendency predicting means 11 such as repair cost, deterioration determined by complementing the change tendency of the occurrence probability of deterioration / damage based on the discrete deterioration / damage data obtained for each inspection.・ Damage occurrence probability calculation means 19 and deterioration that predicts the trend of future probability based on the changing tendency of the deterioration / damage occurrence probability
The structure is provided with a damage occurrence probability predicting means 20 and a repair cost expected value calculating means 21 for calculating an expected value of the repair cost.

In this embodiment, the deterioration / damage measuring means 1 is a visual observation of cracks or a penetrant inspection,
The data converting means 2 is a computer and software for converting the crack generation position and the dimension input into digital data. The operation history recording means 3 is the same as that of the second embodiment, and includes data management means 4, operating condition setting means 5, deterioration / damage occurrence probability calculation means 19, deterioration / damage occurrence probability prediction means 20, and repair cost expected value calculation. The means 21 is a computer and software. The display means 8 is a display device such as a CRT.

FIG. 12 shows a schematic view of the appearance of the combustor liner 22 and the state of crack generation. In the illustrated example, the crack 2
3 has occurred in the welded portion of the mixed air hole collar 24. Since one crack 23 corresponds to one mixed air hole 25,
The information indicating whether or not the crack 23 has occurred in the combustor liner 22 with respect to the specific mixed air hole 25 is damage information.

In the deterioration / damage measuring means 1, this crack 23
Is visually detected, and the crack generation position and the crack length are input.

The data conversion means 2 transmits this input as digital data to the data management means 4. The operation history data and the usage history data are the operation history recording means 3
Is transmitted to the data management means 4 in the same manner as in the second embodiment.

In the data management means 4, as shown in FIG.
As shown in (b) and (c), the "driving history table",
A "use history table" and a "crack generation table" are prepared.

Next, the operation of the deterioration / damage occurrence probability calculating means 19 will be described with reference to FIG. FIG. 14 shows the cumulative number of damages of the combustor liner 22 in which a crack 23 is generated at a predetermined portion (for example, a specific mixed air hole 25) based on the history database and the inspection information of this time, and the number of start and stop at the time of crack detection. It shows the relationship.

Here, the data (damage data) of the phenomenon in which the crack has already occurred at the time of the plotted inspection is
It is not possible to specify the exact time when the crack occurs only by giving the information that "a crack has occurred at some point between the previous inspection and the present inspection". Therefore, since this data is estimated to be larger than the actual number of crack initiation / stop times, the probability of crack initiation is underestimated.

Therefore, a statistical analysis is carried out together with the estimation result of the crack generation time based on the damage data and the data (health data) of the event that the crack has not yet occurred at the time of inspection.
The crack occurrence suppression is estimated by calculating the damage occurrence probability distribution.

FIG. 15 shows a method of allocating the crack generation data found in the same inspection. That is, if a crack is generated at a predetermined position of the n combustor liners 22 between the i-th inspection and the (i + 1) th inspection, the estimated value N _{j of the} number of crack generation start / stop times of the j-th liner is Calculated by the formula.

[0070]

[Equation 12] Here, the predicted value Fo (N) of the cumulative crack occurrence probability of the liner
Is assumed to be proportional to the number N of startups and shutdowns.

[0071]

[Formula 13] Fo (N) = Ao · N (Formula 13) However, Ao: constant From (Equation 13), (Equation 12) becomes the following equation.

[Equation 14]

FIG. 16 shows the relationship between the estimated number of times cracks are started and stopped and the cumulative number of times cracks are generated as described above.

On the other hand, the sound data includes information that "a crack has not occurred until a certain specific time", and when this is ignored, the crack occurrence probability is overestimated. Cumulative crack occurrence probability is assigned to the estimated crack initiation / shutdown count data and sound data based on the damage data by using the Johnson method proposed as a statistical processing method of random censoring data. In this method, 1) damage data and censored data (the total number of data is n) are arranged in ascending order and numbers 1 to n are assigned to the data. However, if the smallest data is the cutoff data, this is omitted. Let i be the assigned natural number. 2) Only the damaged data are arranged in ascending order and numbered as in 1). This number is hereinafter referred to as j. 3) Calculate the following values for damage data.

[Equation 15] 4) The calculated X _j is regarded as the rank number of the damage data, and the damage probability is assigned to the data as in the case of no censoring.

Here, the mean rank method was used to assign the damage probability F. That is,

F = X _j / (n + 1) (Equation 16) where F: cumulative damage probability

FIG. 17 shows the relationship between the number of times of starting and stopping obtained as described above and the cumulative crack occurrence probability.

By considering the damage probability distribution obtained here as the pre-estimated damage probability and repeating the above-mentioned means again, it is possible to calculate the probability distribution with higher accuracy. Also,
Even when the cumulative damage probability cannot be predicted in advance, (Equation 1
The damage probability distribution can be calculated by using the damage probability distribution obtained in the above procedure as the pre-predicted damage probability by using 3) as the first approximation of the pre-predicted damage probability.

Next, the operating condition setting means 5 will be described with reference to FIG.
The operations from the deterioration / damage occurrence probability prediction means 20, the repair cost expected value calculation means, and the display means 8 will be described.

The damage probability distribution f can be calculated from the cumulative damage probability F by the following equation.

F = dF / dN (Formula 17)

From the damage probability distribution, the probability that damage will be found in the component at the next inspection is calculated as follows.

The damage probability λ for one start / stop count at an arbitrary start / stop count N is expressed by the following equation.

Λ (N) = f / (1-F) (Equation 18) In this embodiment,

F (N) = aN (Formula 19) However, a: constant Therefore,

[Formula 20] λ (N) = a / (1-aN) (Equation 20) Furthermore, if the inspection is performed at the number of times of start / stop, and no crack is generated in this inspection and the start / stop is performed ΔN times until the next inspection, the probability FNo (ΔN) that a crack is generated at the next inspection is ,

[Equation 21] Becomes

The number of liners per gas turbine is m
Assuming that the number of liners is 50, the probability P that cracks will occur simultaneously in q liners in the next inspection can be obtained as follows. 1) Consider all combinations of taking q out of n liners. 2) Find the product of FNo (ΔN) for each combination. 3) Find P as the sum of the products of FNo (ΔN). Especially when m liners have the same operation history and repair history,

## EQU22 ## P (q) = mCqFNo ^q (ΔN) (Equation 22)

The repair cost expectation value calculating means 21 calculates the damage probability P calculated by the deterioration / damage occurrence probability predicting means 20.
The expected value E (C) of the liner repair cost C is calculated from (q) as follows.

The liner repair cost C is as follows:
The expected value E (C) of the repair cost of the liner is calculated by the following formula because it is represented by the sum of the fixed cost Cc for transportation and the like and the cost Cv · q proportional to the number of cracks.

[Equation 23]

Further, the repaired liner is continuously used together with the unrepaired liner, but the damage of the liner is completely removed by the repair, and the crack generation probability of the repaired liner becomes equal to that of the unused liner. The repair liner evaluates the next cost based on the difference in the operation history from the time of repair.

In the present embodiment, the expected value of the repair cost thus obtained is output to the display means 8 together with the timing of occurrence of the repair cost as the result of the expected evaluation value of the repair cost.

According to the third embodiment having the above configuration, the occurrence probability is predicted from the discrete inspection information regarding the deterioration / damage event, and the transition of the expected repair cost value is predicted based on this, and the repair cost is predicted. It became possible to accurately predict the level of the expected value and the timing of its occurrence. Further, in this embodiment, since the probability of occurrence of deterioration / damage is obtained from past inspection information, there is an effect that it is possible to accurately predict the difference in lots of component materials and the operation status.

Embodiment 4 (FIGS. 19 to 25) In this embodiment, the invention of claim 4 is applied to the calculation of repair cost for creep deformation of a transition piece of a gas turbine. FIG. 19 is a schematic configuration diagram of this embodiment, and FIG.
0 to 25 show the operation.

In this embodiment, as shown in FIG. 19, instead of the deterioration / damage amount predicting means 6 in the maintenance management support apparatus of the embodiment 1, deterioration / damage amount history data and predicted deterioration / damage estimated amount are used. The temperature / stress estimating means 26 for inversely estimating the temperature and stress conditions of the structural member based on the comparison, and the current deterioration / damage measurement data as initial values based on the temperature / stress estimated values are adapted to future operating conditions. And a deterioration / damage amount predicting unit 27 for predicting the progress of the deterioration or damage amount of the structural member.

In this embodiment, the deterioration / damage measuring means 1
Is an image input device, and the data conversion means 2 is an AD converter and an image processing device. The data management means 4 is the same as that in the first embodiment. The temperature / stress estimating means 26, the deterioration / damage amount predicting means 27, the operating condition setting means 5 and the repair cost calculating means 7 are a computer and software. The display means 8 is a display device such as a CRT.

FIG. 20 shows the deterioration / damage process of the transition piece 28. In FIG. 20,
Axial or circumferential deformation, which is one of the factors governing the life of the transition piece 28, is a time-dependent phenomenon due to high temperature creep, but at high temperatures, coarsening of strengthening precipitates and deterioration phases of the strengthening precipitates occur in the metallic structure of the material. As the generation progresses, the deformation resistance decreases and the creep accelerates, so that the deformation also tends to accelerate. Therefore, the repair interval tends to be shortened with long-term use, and the repair cost tends to increase.

Moreover, when the ceramic coating for heat shielding is peeled off due to deterioration, the metal temperature rises and creep deformation is further accelerated. Therefore, when the deterioration progresses, it is important to correct the design temperature and stress to accurately grasp the actual temperature and stress. Therefore, regarding the deformation of the transition piece 28, the accelerating tendency thereof is estimated together with the usage conditions of temperature and stress based on the observation of the metal structure, and the repair cost is predicted.

21 and 22 show how the deterioration / damage measuring means 1 is applied to the transition piece 28. The replica taken from the main body is observed with a microscope, the observed image is input as an image, and the data conversion means 2 performs noise reduction to further distinguish the images of the carbides present at the grain boundaries from the carbides present within the crystal grains. Extract.

From this image, the grain boundary coverage of the grain boundary precipitates (ratio of the precipitate length to the grain boundary length) ρ and the volume ratio V of the intragranular precipitates are calculated.

The peeling state of the coating is also input as an image, and the peeling position and area are calculated. This is because in the temperature / stress estimation described below, the heat shielding performance deteriorates at the coating peeling portion and Is also used as a basis for overheating.

Further, the axial deformation and the radial deformation of the transition piece 28 are measured by the dimension measuring device 29 and data is input. This size measuring device 29
Is a measurement bed that determines a reference position and a micrometer or a non-contact laser light distance measuring device that measures the distance, and the data is directly input as a signal or by an input device.

These data are stored in the data management means 4, as shown in FIGS. 23 (a), 23 (b), 23 (c) and 23 (d).
Filed as a "transformation table" and an "organization table" and made into a relational database.

[0097]

[Outer 1]

[0098]

[Equation 24]

On the other hand, from the structural data, the creep rate is estimated by the following formula (reference: Research Report Vol.34, No. 2, Committee No. 123, Heat-resistant Metallic Materials Vol. 34, P155-163).

[Equation 25]

[Outside 2]

[Equation 26]

[Outside 3]

[0100]

[Outside 4]

[Equation 27]

When it is determined that Δ exceeds the allowable value by the next inspection, the deformation correction is performed this time, and the deformation prediction is performed using the dimension after the deformation correction as the initial value.

The repair cost calculation unit 7 calculates that the deformation correction amount R _i is proportional to the deformation amount Δ _i at the repair time i predicted by the deterioration / damage prediction unit 5.

R _i = μΔ _i (Equation 28) where μ: constant

Further, the repair period Li is calculated using (Equation 9) which is proportional to the deformation correction amount Ri.

The repair cost is the sum of the fixed cost portion Co including the heat treatment cost after the repair and the portion proportional to the deformation correction amount R _i and the repair period L _i for one part (Equation 10) and one unit. The component group of one group incorporated in is calculated by (Equation 11). deterioration·
For damage prediction and repair cost calculation, the cost of the coating peeling repair amount and the effect of changes in temperature and stress due to the coating repair on the deformation rate are also taken into consideration when necessary.

The above procedure is repeated by changing the operating condition within the range of the predetermined target power generation amount and repeating the calculation to calculate the respective repair costs. The temporal transition of the calculation result is displayed on the display means 8 so that the level of the repair cost and the time when the repair cost occurs can be seen at a glance.

According to the embodiment having the above configuration, even when the development speed such as temperature changes with the development of deterioration / damage, deterioration / damage can be predicted in consideration of the acceleration tendency, and the repair cost can be improved. It is possible to accurately predict the height of the and the time of its occurrence.

Embodiment 5 (FIGS. 26 to 29) In this embodiment, the invention of claim 5 is applied to thermal fatigue crack damage of a gas turbine stationary blade. FIG. 26 shows the schematic configuration of this embodiment, and FIGS. 27 to 29 show the operation.

In this embodiment, as shown in FIG. 26, instead of the deterioration / damage amount predicting means 6 in the maintenance management support apparatus of the first embodiment, deterioration according to a combination of a future operation pattern and repair time or period is performed. A deterioration / damage simulation means 30 for predicting damage, and an adapted operating condition determination means 31 for selecting an assumed operation pattern to which the repair time or period in the deterioration / damage simulation prediction result is adapted.

In this embodiment, the deterioration / damage measuring means 1, the data converting means 2, and the data managing means 4 are the same as those in the first embodiment. The deterioration / damage simulation means 30, the compatible operation condition determination means 31, the operation condition setting means 5 and the repair cost etc. calculation means 7 are a computer and software. The display means 8 is a display device such as a CRT.

When a large change in the operation pattern is expected in the future, it is important in the deterioration / damage prediction to consider the difference in the order and combination of the start / stop patterns. Therefore, in the present embodiment, the operation of the repair evaluation based on the deterioration / damage prediction in consideration of the order effect of the start / stop patterns will be described.

First, the operations of the deterioration / damage measuring means 1, the data converting means 2, the operation history recording means 3 and the data managing means 4 are the same as those in the first embodiment.

FIG. 27 shows a processing procedure by the operating condition setting means 5, the deterioration / damage simulation means 30, and the adapted operating condition judging means 31.

First, the crack distribution state at the current inspection is reproduced based on the operation history, the usage history and the crack data from the data management file 4. The details of this procedure are shown in FIG. That is, the statistical distribution form F of the crack length a is determined from the crack table in the data management means 4. That is, the cumulative probability P is

P = F (a) (Formula 29)

Next, a random number r (0 ≦ r based on this distribution form
≦ 1) is generated by the following equation for the number n of cracks.

A = F ⁻¹ (r) (Equation 30)

Further, the distribution coordinate position x of the crack is expressed as a function of the crack length a, the temperature T and the stress σ.

X = x (a, T, σ) (Equation 31) This determines the initial state of crack distribution.

Next, the operating condition setting means 5 generates the assumed operation pattern P. This is done by giving constraint conditions such as "annual operating time" corresponding to a predetermined amount of power generation, "daily start count", "weekend stop count", "inspection / repair time" and its "duration", and the occurrence thereof. Generate some driving pattern models with different order.

Then, a multi-crack simulation of crack initiation and growth is performed from the above-mentioned initial state, corresponding to each operation pattern model. In this simulation, the material region of interest is represented as a set of discrete element points, and a random crack initiation resistance R is given to each element point.
i and the crack growth resistance Rg are given. Against the resistance, the development type of damage D corresponding to repeated stress at each point following
Give by formula .

[0118]

[Equation 32]

It is assumed that damage development is calculated from (Equation 32) at each point in the simulation region according to the operation pattern, and cracking or growth occurs at each point when D = 1. 29A, 29B, and 29C show examples of the simulation of the stationary blade surface executed in this manner.

In FIG. 27, the adaptive operating condition determining means 31 predicts when the crack damage amount (eg, total crack length) obtained as described above reaches an allowable value,
Compared with the operation pattern model that sets the repair time determined accordingly, when the two do not match, the operation pattern model is rejected, and when they match, the crack damage calculation amount in that operation pattern and the repair amount and repair period determined by it To calculate.

For the repair cost calculation, a circumscribing rectangle is set around the crack and the repair cost is calculated in the same manner as in the second embodiment. Display means 8
Then, the operation pattern candidates remaining after the above selection are displayed as shown in the second embodiment so that the level of the repair cost and the repair time can be seen at a glance.

According to the fifth embodiment configured as described above,
Even if future operation patterns change significantly,
By considering the order of the combination of patterns and selecting the optimum operation pattern under predetermined conditions, it is possible to accurately predict the level of repair cost and the timing of occurrence of repair cost.

Embodiment 6 (FIGS. 30 and 31) In this embodiment, the invention of claim 6 is applied to a gas turbine unit.

In this embodiment, a plurality of parts having different forms of deterioration or damage or different degrees of progress are targeted, and the operation history recording means 3 and the data management means 4 are capable of storing history and managing data for each part. Has been done. Further, a deterioration / damage prediction model setting for determining a constant of a model required for deterioration / damage prediction corresponding to each part based on each historical data of past operation, use, deterioration or damage obtained from the data management means 4 Means 32 and the deterioration / damage prediction model for each component set by the deterioration / damage prediction model setting means 32 are set to the deterioration / damage state at the time of this inspection obtained from the data management means 4 as an initial state. The deterioration / damage state after the next inspection is predicted by using the operation pattern set by the operation condition setting means 5 from this initial state, and the repair or replacement is determined at the inspection point before reaching the allowable value. Deterioration / damage amount prediction means 33 and repair cost or replacement part for each component unit when it is determined that the deterioration / damage prediction means 33 repairs or replaces parts Is a unit repair costs etc. calculating means 34 for calculating the cost of the configuration with.

In the present embodiment, the deterioration / damage measuring means 1
The data conversion means 2 and the data conversion means 2 are the same as those in the first embodiment.
The data management unit 4 has a table for each component group that constitutes one unit of the gas turbine as a database structure. The deterioration / damage prediction model setting means 32, the deterioration / damage amount prediction means 6, the operating condition setting means 5 and the unit repair cost calculation means 34 are a computer and software. The display means 8 is the same as that in the first embodiment.

Of the parts that make up one unit of the gas turbine, the parts that require frequent repairs are moving blades, stationary blades, combustor liners, transition pieces, etc. used under severe high temperature gas conditions. The deterioration / damage factors of these parts are different, and the dependence of the deterioration / damage on the operation pattern is also different. Therefore, the level of repair cost and spare parts cost and the timing of their occurrence are determined and displayed in consideration of the difference in the result / damage form of each part and the difference in the degree of deterioration / damage. This action will be described below.

First, the deterioration / damage measuring means 1 and the data converting means 2 are used in combination with means corresponding to the deterioration / damage form of each part. Since the vane damage is multiple crack damage as shown in the second and fifth embodiments, the image input / image processing devices 12 to 18 shown in FIG. 6 are used as the vane damage measuring means. Since the damage to the combustor liner is a crack in the air hole as shown in the third embodiment, the cracking position input device is used for the damage to the liner.

Since the damage to the moving blade and the transition piece is a creep accompanied by deterioration of the metal structure, Example 4
The apparatus for observing the metallographic structure and the image processing apparatus shown in FIG.
For the rotor blade, a non-destructive method such as an ultrasonic method is applied as necessary in order to determine the state inside the coating layer from the outside in a non-destructive manner, and a deformation measuring device is also used for the transition piece. The operations of the driving history recording means 3 and the data management means 4 are the same as in the second embodiment.

The deterioration / damage prediction model setting means 32 selects a deterioration / damage prediction method according to each part, and makes a prediction based on the past operation, use, deterioration / damage history data obtained from the data management means 4. Determine the model constants required for. For the stationary vane, the method of Example 5 or Example 2,
The method of Example 3 can be applied to the combustor liner, and the method of Example 4 can be applied to the transition piece. For rotor blades, it is the inverse estimation of temperature and stress and the prediction of creep damage based on the deterioration simulation of coating and the quantification of the metal structure of the base material, and the method similar to the transition piece can be applied.

FIG. 31 shows operating condition setting means 5, deterioration / damage amount prediction means 6 and unit repair cost calculation means 34.
Shows the processing procedure by.

First, the operating condition setting means 5 generates an assumed operation pattern. That is, some candidates for start / stop and operation patterns at 100% load and partial load that satisfy the power generation demand imposed on the target gas turbine unit in the future are set.

Next, in the deterioration / damage amount prediction means 6,
The deterioration / damage means at the time of this inspection obtained from the data management means 4 is set as an initial state with respect to the deterioration / damage prediction model selected for each of the moving blade, the stationary blade, the liner, and the transition piece. From this initial state, the operating pattern set in the operating condition setting means 5 is used to predict the deterioration / damage state after the next inspection, and it is determined that repair or replacement will be performed at the time of the inspection before reaching the allowable value.

When it is determined that the repair is to be performed, the repair amount of the part is calculated to calculate the repair cost, and when the part is replaced, the replacement part cost is calculated. Repair time i as a unit, repair cost C in operation pattern k
_U (i, k) is calculated by the following equation.

[Expression 33]

The repair cost for one group of moving blades is expressed by the following equation.

[Equation 34] The same evaluation formula is used for the stationary blade, liner, and transition piece.

The above-described repair cost calculation is executed for each assumed operation pattern, and the repair cost calculation result is displayed on the display means 8 as a graph of each part and unit sum so that the cost can be seen at a glance. indicate.

According to the sixth embodiment having the above configuration, anyone can easily and accurately determine the operating condition that minimizes the repair cost while securing a predetermined amount of power generation for a gas turbine unit consisting of a plurality of parts. be able to.

Embodiment 7 (FIGS. 32 and 33) In this embodiment, the invention of claim 7 is applied to a power plant composed of a plurality of gas turbines.

In the maintenance management support apparatus of this embodiment, deterioration /
The damage amount predicting means 6 is configured to operate, use, and deteriorate all units or all parts constituting the plant.
It has a history database of damages, and is provided with a plant repair cost calculating means 35 for obtaining the repair cost according to the combination of the operation patterns of each unit of the plant and its change over time.

In the present embodiment, the deterioration / damage measuring means 1
The data conversion means 2 and the data conversion means 2 are the same as those in FIG. The data management means 4 is the same as that described above, but a table is prepared for each part of each unit constituting the plant.
The deterioration / damage prediction model setting means 32, the deterioration / damage amount prediction means 6, the operating condition setting means 5, and the plant repair cost calculation means 35 are a computer and software. The display means 8 is the same as that described above.

When the plant is operated by a plurality of gas turbine units, the operation of the present invention for leveling the repair time and the parts replacement time will be described. It is assumed that the total power generation time and the start / stop pattern of the entire plant have been decided, and it is undecided which unit will share the operation pattern.

The operation of the deterioration / damage measuring means 1, the data converting means 2, the operation history recording means 3, the data managing means 4, the deterioration / damage prediction model setting means 32 and the deterioration / damage amount prediction means 6 in FIG. Although substantially the same as in Example 6, the history databases of operation, use, and deterioration / damage are configured for all units and all components that make up the plant.

Next, the operation of the operating condition setting means 5, the deterioration / damage amount prediction means 6, the plant repair cost calculation means 35 and the display means 8 will be described with reference to FIG.

The operating condition setting means 5 creates a combination of the operating time and the start / stop pattern of each gas turbine unit so as to satisfy the power generation plan of the plant, and sets it as an assumed operating pattern.

Next, in the deterioration / damage amount prediction means 6,
The deterioration / damage state at the time of this inspection obtained from the data managing means 4 is set as an initial state with respect to the deterioration / damage prediction model selected for each of the moving blade, the stationary blade, the liner, and the transition piece.

From this initial state, the operating pattern set in the operating condition setting means 5 is used to predict the deterioration / damage state after the next inspection, and it is determined that repair or replacement will be performed at the time of the inspection before reaching the allowable value. .

When it is determined that the repair is to be performed, the repair amount of the part is calculated to calculate the repair cost, and when the part is replaced, the replacement part cost is calculated.

The repair time i as a plant and the repair cost C _P (i, k) in the operation pattern k are calculated by the following equation.

[Equation 35]

The above-mentioned repair cost calculation is executed for each assumed operation pattern, and the repair cost calculation result is displayed on the display means 8 as a graph of each unit and the total sum of the plant so that the cost level can be seen at a glance. indicate.

According to the seventh embodiment having the above configuration, the repair cost and its change with time according to the combination of the operation patterns of the respective units of the plant are displayed, and the operation in which the level of the repair cost and the timing of occurrence of the repair cost are leveled The pattern can be easily determined. This system can be applied not only to one plant, but also to multiple plants, and can obtain a greater effect.

[0150]

As described above, according to the present invention, deterioration
For equipment consisting of multiple parts that differ in the state of damage and the timing of repair / replacement accompanying it, and for plants that consist of multiple equipment, reduce the repair cost as much as possible, and level the repair cost occurrence time. It is possible to perform optimal maintenance management while maintaining the required operating rate.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a first embodiment of the present invention.

FIG. 2 is a diagram showing an operation of the first embodiment.

3 (a), (b) and (c) are explanatory views showing a state of deterioration of a metal structure occurring in a gas turbine high temperature member in Example 1. FIG.

4A, 4B, and 4C are schematic diagrams showing an example of an operation pattern in the first embodiment.

FIG. 5 is a diagram showing a configuration of a second embodiment of the present invention.

6A and 6B are explanatory views showing the appearance and function of the crack image scan input device according to the second embodiment.

FIG. 7 is a diagram showing an image processing procedure of a crack input image according to the second embodiment.

8A, 8B, and 8C are diagrams showing the outline of a history data table in the second embodiment.

FIG. 9 is a diagram showing a procedure of deterioration / damage amount tendency analysis in the second embodiment.

FIG. 10 is a diagram showing a repair cost tendency prediction procedure based on deterioration / damage tendency prediction in the second embodiment.

FIG. 11 is a diagram showing a configuration of a third embodiment of the present invention.

FIG. 12 is a diagram showing cracks generated in a mixed air hole of a combustor liner in Example 3.

13A, 13B, and 13C are diagrams showing the outline of a history data table in the third embodiment.

FIG. 14 is a graph showing the relationship between the cumulative number of damaged liners in which a crack has occurred and the number of times of starting and stopping when a crack is found in Example 3;

FIG. 15 is a diagram showing a method of allocating crack generation data in Example 3;

FIG. 16 is a graph showing the relationship between the estimated number of times cracking is started and stopped and the number of cumulative cracks generated in Example 3.

FIG. 17 is a diagram showing a relationship between the number of times of starting and stopping and cumulative crack occurrence probability in the third embodiment.

FIG. 18 is a diagram showing a repair cost expected value prediction procedure based on damage occurrence probability prediction in the third embodiment.

FIG. 19 is a diagram showing a configuration of a fourth embodiment of the present invention.

FIG. 20 is an explanatory diagram showing a deterioration / damage process that occurs in the gas turbine transition piece according to the fourth embodiment.

FIG. 21 is a diagram showing a method for measuring deterioration and deformation of a transition piece according to the fourth embodiment.

FIG. 22 is a diagram showing a method for measuring deterioration and deformation of a transition piece according to the fourth embodiment.

23 (a), (b), (c), and (d) are diagrams showing an outline of a history data table in the fourth embodiment.

FIG. 24 is a diagram showing a method of estimating temperature / stress from the metallographic structure of the transition piece and the deformation history in Example 4;

FIG. 25 is a diagram showing a repair cost prediction procedure based on deformation prediction in Example 4;

FIG. 26 is a diagram showing the configuration of Example 5 of the present invention.

FIG. 27 is a diagram showing a procedure for determining a compatible operation pattern and repair cost based on a crack growth simulation in Example 5.

FIG. 28 is a diagram showing a method of reproducing the initial crack distribution in the simulation in Example 5;

29 (a), (b) and (c) are diagrams showing an example of crack growth simulation of a vane surface of a stationary blade in Example 5. FIG.

FIG. 30 is a diagram showing a configuration of a sixth embodiment of the present invention.

FIG. 31 is a diagram showing a procedure of prediction and repair cost calculation in the sixth embodiment.

FIG. 32 is a diagram showing a configuration of a seventh embodiment of the present invention.

FIG. 33 is a diagram showing a repair cost procedure in the seventh embodiment.

[Explanation of symbols] 1 Deterioration / damage measuring means 2 Data conversion means 3 Operation history recording means 4 Data management means 5 Operating condition setting means 6 Degradation / damage prediction method 7 Repair cost calculation method 8 display means 9 Deterioration / damage trend analysis means 10 Deterioration / damage amount tendency prediction means 11 Tendency prediction means for repair costs, etc. 12 Crack image scan input device 13 parts 14 Measuring head 15 Position measuring roller 16 CCD imaging device 17 AD converter 18 Image processing device 19 Deterioration / damage occurrence probability calculation means 20 Deterioration / damage occurrence probability prediction means 21 Expected repair cost calculation method 22 Combustor liner 23 cracks 24 Mixed air hole color 25 mixed air holes 26 Temperature / stress estimation means 27 Deterioration / damage prediction method 28 transition pieces 29 Dimension measuring device 30 Deterioration / damage simulation means 31 Applicable operating condition determination means 32 Deterioration / damage prediction model setting means 33 Degradation / damage prediction method 34 Unit repair cost calculation method

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Daizo Saito 4 2-2 Suehiro-cho, Tsurumi-ku, Yokohama-shi, Kanagawa Toshiba Corporation Keihin Business Office (72) Inventor Hiroaki Yoshioka 4-4, Suehiro-cho, Tsurumi-ku, Yokohama-shi, Kanagawa Stock company Toshiba Keihin office (56) Reference JP-A-6-180281 (JP, A) JP-A-4-370741 (JP, A) JP-A-4-265425 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) G05B 23/00-23/02 G01N 17/00

Claims (7)

(57) [Claims] 1. A deterioration / damage measuring means for measuring a degree of deterioration or an amount of damage occurring to a structural member of a device, and an operation history recording means for recording a history of operation, repair, deterioration / damage measurement and the like of the device, Data management means for managing data obtained from each of these means, operation condition setting means for setting future expected operation conditions of the equipment, data input from the data management means and operation condition setting means, and the deterioration・ Deterioration / damage amount prediction means for predicting the progress of deterioration / damage amount of the structural member by performing calculation based on the predicted operating conditions with the damage amount data as an initial value, and the predicted deterioration / damage amount obtained thereby. Calculate the repair time, repair amount and repair period of the structural member based on
Of the repair cost including the sum of the part proportional to the repair amount and the repair period
High and low calculation and repair costs and when repair costs occur
A maintenance management support apparatus for equipment, comprising: a repair cost calculation means for making a determination and a display means for outputting the calculation result. 2. The deterioration / damage amount based on the deterioration / damage measurement history data in place of or in addition to the deterioration / damage amount prediction means and the repair cost calculation means in the maintenance management support apparatus for equipment according to claim 1. Deterioration / damage amount trend analysis means for analyzing the trend, deterioration / damage amount tendency prediction means for predicting the trend of deterioration / damage amount of structural members based on future expected operating conditions, and transition of the deterioration / damage amount Predict the trend of repair timing, repair amount, repair period, and repair cost of the structural member based on the trend, and determine whether the repair cost is high or low.
A maintenance management support device for equipment, comprising: a trend cost predicting means for determining a repair cost for determining the time of occurrence of the repair cost . 3. A deterioration / damage amount prediction means, a deterioration / damage amount tendency analysis means, a repair cost etc. calculation means, a deterioration / damage tendency prediction means and a repair cost etc. in the equipment maintenance management support apparatus according to claim 1 or 2. In place of or in addition to the trend predicting means, a deterioration / damage occurrence probability calculating means that complements the change tendency of the occurrence probability of deterioration / damage based on discrete deterioration / damage data obtained for each inspection , and Deterioration / damage occurrence probability prediction means for predicting future trends of probability based on the change tendency of damage occurrence probability , and repair cost period
Calculates the waiting price and determines whether the repair cost is high or low and when the repair cost occurs
A maintenance management support device for equipment, comprising: a repair cost expected value calculating means for determining the above . 4. The temperature of a structural member based on a comparison between deterioration / damage amount history data and a predicted deterioration / damage estimated amount in place of the deterioration / damage amount predicting means in the equipment maintenance management support apparatus according to claim 1. A temperature / stress estimating means for inversely estimating the stress condition, and the progress of deterioration or damage of the structural member corresponding to future operating conditions with the current deterioration / damage measurement data as an initial value based on the temperature / stress estimated value. A device maintenance management support device, comprising: a deterioration / damage amount prediction means for predicting 5. A deterioration / damage according to a combination of a future operation pattern and a repair time or period in place of the deterioration / damage amount predicting means and the repair cost calculating means in the equipment maintenance management support apparatus according to claim 1. rows and deterioration and damage simulation means and adapted operating condition determination means for selecting a repair time or with fits assumed driving pattern in its deterioration and damage simulation prediction results, the repair cost and other calculated for driving pattern selected to make predictions
It determines whether the repair cost is high or low and when the repair cost will occur.
Maintenance support system apparatus being characterized in that a Nau repair costs etc. calculating means. 6. The equipment maintenance management support apparatus according to claim 1, wherein the equipment has a plurality of parts having different forms or degrees of deterioration or damage,
The operation history recording means and the data management means are capable of recording a history and data management for each of the parts, and the past operation and use obtained from the data management means,
Deterioration / damage prediction model setting means for determining the constants of the model required for deterioration / damage prediction corresponding to each of the parts based on each historical data of deterioration or damage, and each set by this deterioration / damage prediction model setting means For the deterioration / damage prediction model for parts, set the deterioration / damage state at the time of this inspection obtained from the data management means as the initial state, and from this initial state, set the operation pattern set by the operation condition setting means. Using the deterioration / damage amount prediction means that predicts the deterioration / damage state after the next inspection and determines repair or replacement at the time of inspection before reaching the allowable value, and this deterioration / damage amount prediction means When it is determined that repair or replacement will be performed, the repair cost or replacement part cost will be calculated for each component unit, and the repair cost will be higher or lower and the repair cost will be generated.
An apparatus maintenance management support device comprising a unit repair cost calculation means for determining the birth time . 7. The equipment maintenance management support apparatus according to claim 6, wherein the equipment is another plant for power generation, and the deterioration / damage amount prediction means targets all units or all parts constituting the plant. It has a history database of configured operation, use, deterioration / damage, and replaces the unit repair cost etc. calculation method, and calculates the repair cost and its change over time according to the combination of the operation patterns of each unit of the plant , and repairs High and low cost and repair cost
A maintenance management support device for equipment, comprising a plant repair cost calculation means for judging the leveling of the occurrence time .