JP4699344B2 - Equipment maintenance planning method for plant equipment - Google Patents

Description

  The present invention relates to an equipment maintenance planning method for plant equipment, and in particular, treats an occurrence of an accident due to damage to plant equipment as a probabilistic event that models inspection performance and individual variations in an inspection method, instead of a single definite value. The present invention relates to an equipment maintenance planning method for plant equipment that can provide a maintenance plan that is optimal in terms of reliability and cost by performing a simulation that matches actual operation.

  Various large-scale products such as nuclear power plants, thermal power plants or petrochemical plants, and some large-scale products such as aircraft have a large influence on the surroundings at the time of failure, and appropriate maintenance is indispensable. For this reason, it is common to perform periodic maintenance and inspections during the depreciation period, and to perform inspections, repairs, replacements, etc. while considering the life and fatigue level of the parts and units that make up the plant and product. Has been done.

  In particular, equipment such as gas turbines used in plant facilities generates fatigue cracks, cracks caused by heat, stress corrosion cracking, etc. over the years of use, but such cracks require force and temperature. If left untreated, it can lead to equipment damage. Therefore, if these flaws can be found and repaired by inspection, the life of the device can be extended.

  However, if no flaws are found even after the inspection, it leads to an accident, requiring a large cost for repairs, and the equipment becomes unusable, which causes a serious problem from the viewpoint of credit. Therefore, there is a method of shortening the inspection interval and increasing the number of times, but since the inspection itself requires a large cost, the inspection is performed at regular intervals, and repairs are also performed according to the size of the flaw. It is preferable to

  Further, in such plant equipment, the possibility of occurrence of flaws leading to failure usually exists in a plurality of parts. For example, FIG. 17 shows, as an example, (A) a schematic plan view of a crack generated in a cooling hole of a tail cylinder in a gas turbine for a generator, and (B) a schematic cross-sectional view of a hole in which the crack is generated. .

  That is, the cooling hole of the transition piece in the generator gas turbine is provided in a part of the grooves 411, 412, 413,... 416, such as 401, 402, 403,. As described above, as shown in FIG. 17B, a cross-sectional schematic view of a part of the crack 412 is shown as cracks 420, 421 caused by fatigue cracks, heat, stress corrosion cracking, etc. 422 may occur, but such cracks may become large due to the application of force or temperature.

  Moreover, when such cracks occur in directions facing each other, such as the cracks indicated by 420 and 421, for example, even if each crack is small, the cracks are developed and connected by heat and stress applied thereto, It may become a big flaw in a short time and may lead to breakage. In other words, in plant equipment, not only the occurrence of cracks in each part, but if the relationship with the surroundings is not taken into account, even small cracks can become large scratches in a short time, leading to an accident. There is a reason.

  In addition, although such cracks are likely to occur, cracks do not occur uniformly in such areas, but occur with some variation, and accidents occur at different times. It is normal.

  In addition, as shown in the schematic diagram 18 showing the process up to the occurrence of the accident, cracks 420 and 421 occur in 402 and 403 among the plurality of portions (cooling holes) 402, 403,. The inspection 430 was carried out, and each defect was detected as indicated by 431 and 432. For example, the crack of 421 was repaired / replaced as indicated by 442. If repair is not performed as in 441 based on the view that the inspection is okay, there are cases where the crack progresses to 450 and leads to an accident as in 451.

  For example, although the crack 420 was detected as 431 at the stage of the inspection 430, there was a difference between the size of the flaw shown in the inspection result and the actual size of the flaw, and it was found that it was underestimated. This was because the flaw that was made became the view that it was okay until the next inspection. That is, the accident in this case can be said to be an accident caused by the low performance related to the accuracy of the inspection method.

  Regarding the maintenance of such plant equipment, for example, Patent Document 1 assumes that a maintenance management plan for a gas turbine is used, and a damage assessment master curve is used for the deterioration / damage state of the gas turbine during periodic inspection of the gas turbine. A maintenance management method is shown that is evaluated by a remaining life diagnosis device and determines whether or not repairs are made and whether or not they are discarded.

  In Patent Document 2, the display status of the plant parts, the event tree development for the failure event and countermeasure of each part, and the maintenance management information of the parts are displayed on the display device. A maintenance management plan support method is disclosed in which an optimum maintenance management plan can be drawn up from the viewpoint of economy by allowing changes.

  Further, in Patent Document 3, in order to perform safety management of plant facilities according to the failure probability of each device constituting the plant device, the relationship between the shared years and the failure probability is determined for each device according to the design data of the plant device. An apparatus for managing a risk by determining a replacement probability function, generating a corrected failure probability function based on at least one of an operation history and an inspection relating to the device, and determining replacement of the device is shown.

  Further, in Patent Document 4, in order to generate a maintenance plan for a plant that has the largest reduction in risk cost and satisfies the constraints, information including maintenance methods and costs for each part of the plant, and maintenance are performed. The maintenance conditions are stored, and the maintenance reduction plan is calculated for each maintenance part of the plant by calculating the amount of risk reduction when maintenance is performed based on the combination of the maintenance timing and the maintenance method to be used. A maintenance plan generator is shown.

  Further, Patent Document 5 discloses a unit that configures a plant or product at the end of each short-term maintenance period so that a maintenance plan over the entire depreciation period of the plant or product can be quantitatively planned in consideration of cost. In addition, a method for determining an optimum maintenance plan is described in which the state related to the parts and the parts are calculated from the damage probability, and the calculation of the inspection cost and the maintenance cost is repeated until the depreciation year.

  Also, in Patent Document 6, general material life characteristics such as creep and fatigue and their probability distribution are stored in advance in a statistical material database, and the life against creep and fatigue is calculated using the material life characteristics and the probability distribution. An operation support apparatus for plant equipment that evaluates and selects an optimum operation method is shown.

Japanese Patent Laid-Open No. 10-293049 JP 2004-234536 A JP 2004-191359 A JP 2005-148955 A JP 2003-99119 A JP 200494663 A

  However, the maintenance management planning method disclosed in Patent Document 1 estimates the remaining life using a single damage evaluation master curve, and determines whether or not there is repair and whether or not it is discarded. In plant equipment, although there are parts where cracks are likely to occur, cracks do not occur uniformly in such parts, but they occur with some variation and accidents occur at different times. Because of this, it is not possible to create a cost-effective maintenance plan with these methods.

  In addition, in the maintenance management plan support method disclosed in Patent Document 2, it is only necessary to specify whether or not to repair a device, and an inspection for determining whether to perform the repair is not incorporated. It is thought that it does not have a function.

  Furthermore, in the risk management device disclosed in Patent Document 3, the inspection / repair target is limited to one part, and for example, inspection / repair is performed on a plurality of parts such as a cooling hole for a gas turbine. It is not possible to target equipment that requires

  In the maintenance plan generation apparatus disclosed in Patent Document 4, the probability of damage for each part is used in the risk cost evaluation formula, but there is a method of handling inspection performance and repair methods that match the actual operation of maintenance. It is not equipped and it is hard to say that the system is suitable for practical use.

  Further, in the optimum maintenance plan determination method shown in Patent Document 5, the damage probability uses a single value, and cracks do not occur uniformly even in parts where cracks are likely to occur as described above. Plant equipment that occurs with some variation cannot handle damage and repairs that can occur when multiple parts are combined.

  Further, Patent Document 6 obtains the failure probability based on statistical data of general-purpose materials, and more realistic damage states based on operation conditions set by operation methods unique to plant equipment such as combustion methods. It is not possible to provide a maintenance plan that takes into account Further, it does not show the influence of the detection performance of a measuring instrument that measures material damage, the influence of power sales revenue as a power plant facility, the influence on the environment, or the like.

  Therefore, in the present invention, the occurrence of an accident or the like due to breakage in a plurality of parts in plant equipment is handled as a random event modeled on the inspection performance and individual variations in the inspection method, instead of a single definite value. Considering the operating conditions according to the operation method unique to the facility, the power sales revenue, and the impact on the environment, etc., it is possible to provide an optimal maintenance plan in terms of reliability and cost by performing simulations that match actual operation It is a problem to provide an equipment maintenance planning method for plant facilities.

In order to solve the above problems, the plant maintenance equipment maintenance planning method according to the present invention,
A step of presetting a specific part where the occurrence of a crack of each part of the plant equipment and the number of the specific parts is set in advance, and detecting the occurrence of the crack for each specific part set in advance; A step of calculating each progress based on a parameter designated in advance when it occurs, the presence / absence and occurrence direction of a crack in a part adjacent to the specific part and the length of the generated crack A crack calculation step comprising: a connection crack determination step for determining whether or not the calculated crack is a connection crack in which the progress of the crack progresses quickly depending on whether or not the crack distance between the parts is equal to or less than a threshold value. And the presence or absence of cracks for each specific part, the crack size of the generated part, and the position until the number of specific parts set previously in the crack calculation step for each specific part is reached. , Generating crack occurrence data including the presence or absence of occurrence of connected cracks, and based on the obtained crack occurrence data, a specific site and its neighboring neighboring sites (hereinafter referred to as “multiple sites that affect crack progress”). ) Predicting the degree of progress of cracks, and determining the probability of component damage and the occurrence of an accident corresponding to the “plurality of parts that affect crack progress” based on the predicted value of the degree of crack progress. Features.

  In this way, by simulating the equipment maintenance plan for plant equipment as a collection of multiple parts that take into account the mutual correlation of multiple events based on the operating conditions of the plant equipment, the mutual effects of parts that could not be handled in the past have been achieved. The simulation can be performed, the risk prediction accuracy can be improved, and an optimal maintenance plan can be provided in terms of reliability and cost.

Cracks and the database, for each "multiple sites affects the crack growth from each other" in the plant equipment, and inspection records containing test costs, the occurrence probability data for cracks based on the design data, that occurred in the past Prepare a database that stores data including repair costs related to repairs and costs required for recovery in the event of an accident due to cracking ,
Based on the predicted value of the degree of progress of cracks of `` multiple parts that affect crack progress with respect to each other '', the determination of the probability of occurrence of part damage and accident occurrence corresponding to the plurality of parts is based on the predicted value. The determination is made while increasing the probability of occurrence of an accident as it approaches the reference size of an accidental crack, and based on the determination result, the cracks of a plurality of parts that affect the progress of the cracks that occurred in the past from the database. A maintenance plan that reflects the accident risk is created by comparing the repair costs related to the accident and the costs required for recovery in the event of an accident due to crack progress .

Equipment maintenance plan Thus plant equipment, de - database in correspondence with the actual inspection records for each of the plurality of sites in an equipment of the stored plant equipment, the probability of occurrence of cracks based on the design data for each site, crisp By simulating the degree of progress over time and the necessity and method of repair as a collection of multiple parts that take into account the correlation between them, improvement of risk prediction accuracy, reliability and cost It is possible to make an optimal maintenance plan from

Then, the simulation for each inspection technique intends row by adding the test performance is determined by the difference in size of the actual crack the size of the crack test results showed.

When using multiple inspection methods, these inspection methods generally have their pros and cons, and the test results will differ depending on whether the person performing the inspection is an expert or a beginner. There is. Thus, in this manner for each inspection method, the inspection result by adding the test performance is determined by the actual difference in size of the crack and the size of the crack shown, not only the cost, reliability From the aspect, it is possible to create an optimal maintenance plan that is suitable for actual operation.

In addition, when predicting repairs based on the predicted value, when repairing only parts where cracks occur, repairing including peripheral parts, repairing all of the parts, and replacing the equipment with a new one The risk of damage to equipment and the cost of each are calculated and reflected in the maintenance plan.

Thus, when predicting repairs based on the predicted value, even if cracks are predicted in one part of multiple parts, repair only the part where the cracks occur, including the surrounding parts. Effective maintenance including repair methods in line with actual operation by simulating cost effectiveness including subsequent failure risks, assuming various cases such as repairing everything, replacing with new ones, etc. A plan can be created.

Further, the periodic inspection means for preventing the occurrence of cracks is stored in the database, and the risk and cost of an accident that occurs when the periodic inspection means for preventing the occurrence of the cracks is implemented in the simulation, It is characterized by being reflected in the maintenance plan.

Simulation results for predicting the progress degree of cracking "Multiple sites affecting crack growth from each other", even if it is not cracks occurred or a minor for example generated prediction, the stochastically future cracks so there is likely to occur, in which case this way, by calculating the risk and cost of accidents caused when implemented generation preventive maintenance means crack, preventive maintenance further in line with actual operation A maintenance plan can be created.

Further, the simulation calculates a risk while increasing the probability of occurrence of an accident as it approaches the reference size of a crack that becomes an accident with respect to the predicted size of the crack , and reflects it in the maintenance plan. And

For example, in the case of a crack generated in the cooling hole of the tail cylinder in the above-described generator gas turbine, an accident usually occurs when the crack progresses and a leak occurs. Therefore, it is possible to determine the occurrence of an accident based on the presence or absence of a leak based on the result of each progress, but in this case, the accident is not determined until a leak occurs. In contrast, as the present application, with respect to the size of the expected crack, we calculate the risk while increasing the accident probability approaches the reference size of the crack as the accident immediately to more actual operation Thus, a maintenance plan reflecting the occurrence of an accident as a risk can be created.

Further, the data of "a plurality of regions that affect crack growth from each other" includes the positional relationship of the mutual position information for each specific site relates cracks predicted to one site, and cracks occurring in the surrounding site A simulation is carried out by incorporating a model in which the crack growth speed is accelerated depending on the interval and direction of the cracks , and reflected in the maintenance plan.

  As described with reference to FIG. 17, when the cracks occur in directions facing each other, such as cracks 420 and 421, even if each crack is small, the cracks are developed and connected in a short time due to heat and stress applied thereto. May lead to an accident. However, in this way, a model that includes the positional information for each part and the mutual positional relationship and incorporates a model that accelerates the speed of flaws depending on the distance and direction of the flaws that occur in the peripheral part with respect to flaws that are predicted in one part. Thus, it is possible to create a maintenance plan that reflects, as a risk, a damaged state in which a plurality of parts affect each other.

It is to be noted that the simulation is carried out using the Monte Carlo method in a preferred embodiment of the present invention, the equipment of the plant equipment is a gas turbine tail cylinder, and the crack is attached to the gas turbine tail cylinder. By applying to a plurality of provided cooling holes, a greater effect can be obtained.

Further, according to the present invention, there is provided a plant facility equipment maintenance plan method for generating a plant equipment maintenance plan, comprising a database for each of “a plurality of parts that affect crack progress in the plant equipment ”. In the database, the operation method specific to the plant equipment is stored, and based on the operation method, the thermal and mechanical operation conditions are set in the “multiple parts that influence the crack progress with respect to each other” to generate cracks. The presence or absence of cracks and the size of cracks are obtained and the probability of breakage is simulated and reflected in the maintenance plan for the plant equipment.
Thus, by setting the thermal and mechanical operating conditions in said plurality of portions based on the operating condition according to the plant equipment specific operation method, damage to seek the magnitude of the presence or absence of cracks and cracks Since the probability is simulated, an optimum maintenance plan specific to the target plant can be accurately performed as compared with the method of simulating based on general-purpose general data.

Further, the plant equipment is a gas turbine, the operation method specific to the plant equipment is an operation by diffusion combustion of the gas turbine and an operation by mixed combustion, and the thermal and mechanical operating conditions are the diffusion combustion and mixed combustion. It is characterized by the temperature and stress conditions in each.
Thus, for example, the gas turbine inlet temperature and the stress due to vibration differ depending on whether the operation method of the gas turbine is operation by diffusion combustion or operation by mixed combustion. Use operating conditions of temperature and stress according to the method. Further, since also performed to change the operation method after periodic inspection, temperature tailored to change the operation method, using a stress condition, simulating the failure probability seeking magnitude of the presence or absence of cracks and cracks Therefore, the optimum maintenance plan specific to the target plant can be performed with higher accuracy than the method of performing simulation based on general-purpose general data.

In addition, when the difference between the failure probability obtained based on the inspection result and the failure probability obtained by the simulation is equal to or greater than a threshold value under a predetermined inspection condition, the failure probability is made closer to the failure probability obtained from the inspection result. The input data of the thermal and mechanical operating conditions of “a plurality of sites that influence the progress of cracks” in the database is adjusted.
The size of the crack as the actual value detected in the inspection includes an error with respect to the size as the true value, but the inspection conditions are within a certain range, that is, the inspection method or operation with high inspection performance. Since inspections with a lot of time are considered to have large cracks and high detection accuracy, by adjusting the input data in the database so as to approach the failure probability obtained from the inspection results, a highly accurate maintenance plan for each target plant Can be performed.

In addition, the plant equipment is a gas turbine for power generation, and the database stores the operation method of the gas turbine, the fuel used, and the amount of power generation. The data is sold by applying these data to a predetermined profit model. It is characterized by simulating revenue from electricity.
As described above, the power generation amount by the operation method of the gas turbine, the amount of fuel used by the operation method, and the like are applied to a predetermined revenue model, and the maintenance plan can be performed in consideration of the profit from the power sale by obtaining the revenue. .

The plant equipment is a gas turbine, and the database stores harmful exhaust gas emissions such as NOx and CO in the operation method of the gas turbine, and simulates the environmental impact based on these data. It is characterized by.
Thus, by simulating the environmental impact based on the data of harmful exhaust gas emissions such as NOx and CO by the operation method of the gas turbine, it is possible to perform a maintenance plan considering the environmental impact for each plant. .

The database stores a period required for repair, a storage location of the repair device, and a cost required for the repair, and the repair period and the repair cost are simulated based on these data.
In this way, when there are several places where equipment necessary for repair is stored, it can be determined by simulation whether the repair period is given priority or the repair cost is given priority. A maintenance plan that considers the repair period and repair location can be performed.

In addition, the database stores data in which a recycling rate is set for each device, and the degree of influence on the environment when the device is replaced with a new one by the treatment after the accident by the simulation or by the disposal by the inspection result It is characterized by simulating.
In this way, a maintenance plan that takes into account the impact on the environment when the equipment required for repair is discarded by simulating the degree of impact on the environment when the equipment is discarded by setting the recycling rate for each equipment. Can be performed.

As described above, according to the present invention, the occurrence of an accident or the like due to damage at a plurality of parts in the plant facility equipment is simulated by including the mutual influence of “a plurality of parts that influence the progress of cracks in the plant equipment ”. Can be handled as a probabilistic event that models the inspection performance and individual variations in the inspection method, instead of a single deterministic value. Considering the impact, etc., it is possible to provide an optimal maintenance plan in terms of reliability and cost by simulation that matches the actual operation.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. However, the dimensions, materials, shapes, relative arrangements, etc. of the components described in this embodiment are not intended to limit the scope of the present invention unless otherwise specified, but are merely illustrative examples. Only.

  FIG. 1 is a schematic flow diagram of an equipment maintenance planning method for plant equipment according to the present invention. In the following description, the present invention will be described by taking as an example the case of a crack generated in the cooling hole of the tail cylinder in the above-described generator gas turbine. However, the present invention is not limited to such a case, for example, It is clear that it can be applied to maintenance plans for related but different parts such as the relationship with blades and other rotating bodies in gas turbines, and the settling and wear of spring coils.

  In the present invention, as shown in FIG. 1, first, the plant operating condition 1 including the plant operating time, the number of start / stop times, the load fluctuation, the number of input parts, the field performance 3 and the design data of the newly designed parts are used. From the analysis prediction value 4 including the crack occurrence probability data based on the crack, for example, the crack generation data including the average size, variation, direction, etc. of the crack, the length of the crack with respect to the number of times of starting the gas turbine, etc. Part life data 2 consisting of crack progress data, crack length and limit length that need repair, maintenance plan 5 consisting of inspection plan, inspection interval, inspection method and inspection accuracy, etc. Cost data 6 consisting of costs, inspection costs, repair costs, costs required for recovery in the event of an accident, etc. is used as a database in advance. To.

And the contents stored in these databases as data, multiple sites 7 ₁ in an equipment of a plant facility in block _7, 7 2, _{7 3,} each ...... 7 _n, and for example, the Monte Carlo method while including respective correlation Perform the simulation used, calculate the result 8 consisting of the number of repairs, the number of scraps, the number of accidents, etc., and the repair rate, discard rate, accident occurrence etc., and finally check, repair, new The life cycle cost consisting of input, accident recovery cost, etc. is calculated as a report item 9 for the equipment maintenance plan of the plant facility.

  In the actual simulation 7, first, the gas turbine is mounted in step S10, and in step S11, the operation condition 1 of the plant including the operation time, the number of start / stop times, the load fluctuation, the number of input parts, etc. stored in the database is read out. Further, the component life data 2, the maintenance plan 5, the cost data 6, etc. are read out and the operation start mode is entered.

  When a crack occurs in step S12 and further progresses in step S13, it is determined in step S14 whether or not an accident occurs. If an accident does not occur, the process proceeds to step S15, where an inspection is performed to determine whether or not an accident has been detected in step S16. If no crack is detected, the process proceeds to step S17 to determine whether or not preventive maintenance is to be performed. If not, the process returns to step S10, and if so, preventive maintenance is performed in step S18.

  If it is detected in step S16, the process proceeds to step S19 to determine whether or not to repair. If not, the process proceeds to the left side of the figure or proceeds to the right side and returns to step S10 as it is. Although not shown, there is also a method for proceeding to preventive maintenance in step S16 and performing the preventive maintenance described above. On the other hand, when repairing, it progresses to step S20, repairs, and returns to step S10.

  If it is determined in step S14 that an accident has occurred, a recovery is made in step S22, a new article is inserted in step S23, and the process returns to step S10.

Then, the above-described routine is executed for each of the plurality of parts 7 ₁ , 7 ₂ , 7 ₃ ,... 7 _n including the respective correlations, and the result 8 including the number of repairs, the discard rate, the number of accidents, etc. , Calculate the repair rate, disposal rate, accident occurrence, etc., and finally report the life cycle cost consisting of inspection, repair, new product introduction, accident recovery cost, etc. to the equipment maintenance plan for plant equipment 9 Calculate as

  In this way, the equipment maintenance plan for plant equipment is made to correspond to the actual inspection records for each of the parts in the equipment of the plant equipment, and the probability of occurrence of cracks based on the design data for each part, and the passage of time when cracks occur By simulating the degree of progress, the necessity and method of repair, etc. as a collection of multiple parts taking into account the mutual correlation, it is possible to simulate the influence of parts that could not be handled in the past, and the risk prediction accuracy It is possible to improve and provide an optimal maintenance plan in terms of reliability and cost.

  The above is a schematic flow diagram of the equipment maintenance planning method for plant equipment according to the present invention. Next, the first embodiment of the present invention will be described in detail with reference to the flowcharts shown in FIGS. To do.

(First embodiment)
2 is a detailed flowchart of the equipment maintenance planning method for plant equipment according to the first embodiment of the present invention, FIG. 3 is a detailed flowchart of crack generation calculation, and FIG. 4 is a detailed flow of accident generation calculation. 5 is a detailed flow diagram of preventive maintenance case A, FIG. 6 is a detailed flow diagram of preventive maintenance case B, FIG. 7 is a detailed flow diagram of repair case A, and FIG. 8 is a detailed flow of repair case B. FIGS. 9 and 9 are detailed flowcharts of repair cases C and D. FIG. In these flowcharts, the same steps as those in FIG. 1 are given the same step numbers.

  First, the data used in this simulation, as explained in FIG. 1 above, is the plant operating condition 1, the part life data 2 which is crack generation / progress parameter data, the maintenance plan (maintenance plan) data 5, The cost data 6 and the number of Monte Carlo calculation cases are given as the calculation parameter data 10.

  In step S30, the simulation is started. In step S31, the component life data 2, maintenance plan (maintenance plan) data 5, and cost data 6 are read, and the number of Monte Carlo calculation cases given as the calculation parameter data 10 in step S32. Is set, and the operation years input according to the operation condition 1 are set in step S33.

  In the crack calculation step S11, the plant operating condition 1 including the operation time, the number of start / stop times, the load fluctuation, the number of input parts, etc. stored in the database, the part life data 2, the maintenance plan 5, and the cost data 6 are stored. Etc. are read out and the generation calculation is performed in step S12.

  In step S12A, it is determined whether or not a crack has occurred. If not, the process proceeds to step S14, and if it has occurred, the crack is further calculated in step S13.

  FIG. 3 shows a detailed flow of the crack calculation step S11. In the flow of FIG. 3, calculation is started in step S110, and parameters for the calculation are shown in step S111. In other words, this parameter is the size of the crack consisting of the number of parts to be calculated for cracks, part data consisting of position (coordinate value) data, stress data generated for each part, load data such as load fluctuation, depth, length, etc. Initial value, probability distribution type such as crack direction, average, crack parameter consisting of variation, type of probability distribution of parameter coefficient used for crack propagation calculation, average, crack propagation parameter consisting of variation, stress Operating parameters such as load fluctuation and temperature.

  When the crack calculation parameter is given in this way, the number of sites to be calculated is given in the next step S112, and the crack generation / progress parameter for each site is set in step S113 (setting is performed only once). This is set by a random number according to a specified probability distribution, as described on the right side of step S113.

  Then, it is determined whether or not an occurrence has already occurred in the next step S114, and if it has not occurred, the process proceeds to step S115 to compare the occurrence time with the current operation time and determine whether or not it has occurred. If it is determined that the occurrence has occurred, the process proceeds to step S116 for progress calculation from the next time, but in the present case, the process proceeds to step S117 as it is.

  On the other hand, if it is determined that the occurrence has occurred in step S114, the process proceeds to step S116, and the size and position (including the tip position) of the crack are calculated over time by progress calculation. This calculation uses a calculation formula selected according to the type of cracks such as fatigue, stress corrosion cracking, and creep, as described on the left side of step S116.

  Then, in the next step S117, it is determined whether or not the number of parts to be calculated for cracking is 0. If not, the process returns to step S112 and the number of parts set here is decremented. Repeated.

  If it is determined in step S117 that the calculation has reached the previously set number of parts (that is, the number of parts has become 0), the process proceeds to the next step S118, where the number of parts is also set. The

  In the next step S119, it is determined whether or not cracks have already occurred in the adjacent parts, such as cracks 420 and 421 shown in FIG. If it does not occur, the process proceeds to step S123. If it occurs, the distance between each part is calculated from each tip position in step S120.

  Then, in the next step S121, it is determined whether or not the connection is broken depending on whether or not the crack distance between the parts is equal to or less than the threshold value. If it is not the connection crack, the process proceeds to step S123. Two progress calculation parameter values are set for a connected crack that progresses faster than a normal crack.

  Then, in the next step S123, it is determined whether or not the number of parts to be cracked is 0. If not, the process returns to step S118 and the number of parts set individually is decremented. Repeated.

  If it is determined in step S123 that the calculation has reached the previously set number of parts (that is, the number of parts has become 0), the process proceeds to the next step S124, where the size of each part is large. The position, presence / absence of connected cracks, progress parameters, etc. are output, and the process returns to the flow of FIG. 2 in step S125.

  By inspecting the connected cracks in this way, they are generated in the directions facing each other like the cracks 420 and 421 described in FIG. 17, and even if each crack is small, the heat applied to the cracks is reduced. Even when the stress develops in a short time and leads to an accident, it is possible to create a maintenance plan that reflects as a risk a damaged state in which a plurality of parts affect each other.

  When the clear calculation is completed in this way, the process returns to step S14 in the flow of FIG. 2, and it is determined whether or not an accident occurs this time. This accident occurrence determination flow is shown in FIG. 4. Before that, the case where an accident occurs and the process proceeds to step S22 will be described. In step S22, the process at the time of occurrence of the accident is performed, and further step S23 is executed. In step S35, the device having the accident is replaced with a new one.

  Next, the detailed flow of the accident occurrence step 14 in FIG. 4 will be described. Also in the flowchart shown in FIG. 4, parameters for accident occurrence calculation are shown in step S141. That is, this parameter includes the number of parts to be calculated for cracks, part data consisting of position (coordinate value) data, (1) type, average, and variation of probability distribution of occurrence probability with respect to the crack size for each part, (2) Probability distribution of probability of occurrence with respect to the size of cracks taking into account the influence of each part, average, variation, (3) probability of occurrence of accidents, including the probability of occurrence for the number of sites whose crack size exceeds the threshold .

  When the accident occurrence parameters are given in this way, the number of parts for which the accident occurrence is calculated is given in the next step S142, and the accident probability distribution for each part is set in step S143. In this case, as shown on the right side of step S143, data according to a specified probability distribution is set.

  Then, in the next step S144, it is determined whether or not an accident has occurred. This determination is made by first using the type, average, and variation of the probability distribution of the probability of occurrence for each part shown in the accident probability (1) described in step S141, and the probability of occurrence for the current crack size. A is obtained from the probability distribution. Next, the value a is determined by a uniformly distributed random number from 0 to 1, and it is assumed that an accident occurs when a ≧ A. If it has occurred, the process proceeds to step S153. If it has not occurred, the process proceeds to step S145 to determine whether or not the magnitude is equal to or greater than a threshold value.

  If it is determined in step S145 that the magnitude is less than or equal to the threshold value, the process proceeds to step S147. If it is determined that the magnitude is greater than or equal to the threshold value, in step S146, the number of parts having a high possibility of accident occurrence (A) is incremented. Go to step S147.

  Then, in the next step S147, it is determined whether or not the number of parts has become zero. If not, the process returns to step S142, the number of parts set here is decremented, and the above description is repeated.

  If it is determined in step S147 that the previously set number of parts has been reached (that is, the number of parts has become 0), the process proceeds to the next step S148, and the accident occurrence probability data (3) is cleared. It is determined whether or not an accident due to the number of parts (A) having a high possibility of occurrence of an accident occurs using the probability of occurrence for the number of parts whose size exceeds the threshold. If an accident occurs, the process goes to step S153. If not, the number of parts is set again in step S149.

  Then, in the next step S150, the size of the crack in the adjacent part is calculated, and based on the size, it is determined whether or not an accident occurs in step S151. This determination is made based on the type, average, variation, etc. of the probability distribution of the occurrence probability with respect to the crack size in consideration of the mutual effect of the accident occurrence probability (2). If it is determined that an accident has occurred, the process proceeds to step S153. If not, the process proceeds to step S152 to determine whether the number of parts has become zero. If not, the process returns to step S149 and is set here. The number of sites thus decremented is decremented and the above description is repeated.

  In this way, by calculating the risk while increasing the probability of occurrence of an accident as it approaches the standard size of an accidental crack, the occurrence of an accident can be further improved in accordance with actual operation. A maintenance plan reflected as a risk can be created.

  If it is determined in step S152 that the previously set number of parts has been reached (that is, the number of parts has become 0), the next step S153 reports whether or not an accident has occurred. In step S154, the flow of FIG. Return.

  When the accident occurrence process is completed, the process returns to step S15 in the flow of FIG. 2 to perform an inspection, thereby determining whether or not an error has been detected in step S16. If no crack is detected, the process proceeds to step S17 to determine whether preventive maintenance is to be performed. If a crack is detected, the process proceeds to step S19 to determine whether repair is to be performed. If this repair is not performed, the process proceeds to step S17, where it is determined whether to perform preventive maintenance, and the process proceeds to step S34, where it is discarded. If repair is performed, the process proceeds to step S19A.

  Further, when no failure is detected in step S16 and no repair is performed in step S19, it is determined whether or not preventive maintenance is performed in step S17. If not, the process goes to step S35 to perform preventive maintenance. In that case, preventive maintenance is performed in step S18. The detailed flow of this preventive maintenance step S18 is shown in FIGS.

  First, in the flowchart shown in FIG. 5, first, parameters for preventive maintenance are shown in step S301. In other words, this parameter includes preventive maintenance consisting of the number of parts to be calculated for cracks, position (coordinate values), part data consisting of the number of parts detected for cracks, occurrence of cracks after preventive maintenance, progress parameters, and preventive maintenance costs. Preventive maintenance data comprising a method and, for example, case A in which preventive maintenance is performed on all sites if the number of sites in which cracks are detected is m or more, and case B in which only a specific range including sites in which cracks are detected is performed It is.

  The flow chart of FIG. 5 is a flow in the case where preventive maintenance is performed on all parts of the parameters when the number of parts in which the crack of case A is detected is m or more, and FIG. It is a flow in the case of implementing only the specific range including the site | part which detected one. As described in step S301, in case A, the case where m = 0 is not detected is performed. In addition, Case B can be less expensive than the full range implementation, and the implementation time can be shortened.

  When the parameters for preventive maintenance are given in this way, in FIG. 5, preventive maintenance is performed by the method specified in the next step S302, and the preventive maintenance cost for the entire range is calculated in step S303. In step S304, crack generation parameters after maintenance are set, in step S305, the presence / absence of preventive maintenance, maintenance costs, and part data including crack generation parameters are reported. In step S306, step S35 of the flow of FIG. Return to.

  On the other hand, the flowchart of FIG. 6 shows a case B in which only a specific range including a part where a crack is detected is implemented, and the number of parts in the preventive maintenance target region is set in step S310. Then, preventive maintenance is performed by the method specified in the next step S311, the crack generation parameter after the maintenance is performed is set in step S312, and whether or not the number of parts in the preventive maintenance target area has become zero in step S313. If it is determined that the number is not 0, the process returns to step S310, the number of parts set here is decremented, and the above description is repeated.

  If it is determined in step S313 that the number of parts set previously (that is, the number of parts has reached 0) is determined, the preventive maintenance cost is calculated in the next step S314, and the flow of FIG. The process returns to step S35.

  As a result of the simulation, even if the cracks that occur are minor or are not predicted, there is a possibility that cracks may occur in the future. By calculating the risk and cost of an accident that occurs when the preventive maintenance measures are implemented, a maintenance plan that includes preventive maintenance that is more suited to actual operation can be created.

  For this preventive maintenance, for example, water jet peening can be used. This is because high pressure jet water containing bubbles is sprayed on the metal surface to cause plastic deformation, thereby changing the residual stress on the surface from tension to compression, thereby causing stress corrosion cracking (SCC) of the weld. This is a method of significantly reducing the tensile residual stress in the vicinity of the weld, which is the cause.

  When the preventive maintenance process is completed in this way, the process returns to step S35 in the flow of FIG. 2. Before that, in step S19 of FIG. 2, the process proceeds to the elimination of step S34 and the repair step S19A. The repair flow will be described.

  First, in step S19, the process proceeds to the determination of disposal in step S34 without performing repair. If not discarded here, the process proceeds to step S35 as it is. And proceeds to step S35.

  In this case, the repair is performed in step S19. In this case, the process proceeds to repair in step S19A. Details of the repair step S19A are shown in the flowcharts of FIGS.

  First, in the flowchart shown in FIG. 7, first, parameters for repair are shown in step S191. In other words, this parameter is the repair method consisting of the number of parts for calculating cracks, the position (coordinate values), the part data consisting of the number of parts data for detecting cracks, the occurrence of cracks after repair, the progress parameter, and the repair cost, For example, in case A where only the part determined to be repaired is repaired, in case B where the part adjacent to the part determined to be repaired is repaired, and if there are n or more parts determined to be repaired, repair all This is repair case data including case C, case D that is replaced with a new one if there are n or more parts determined to be repaired.

  The flow chart of FIG. 7 shows the case where only the part determined to be repaired in case A is repaired, and FIG. 8 shows the part adjacent to the part determined to be repaired in case B. In the case of repair, FIG. 9 shows a case where all repairs are required if there are n or more parts determined to be repaired in cases C and D, and a case where there are n or more parts that are determined to be repaired, which are replaced with new ones. It is.

  When the repair parameters are given in this way, in FIG. 7, the number of parts is given in the next step S192. In step S193, it is determined whether repair is necessary for each part. If there is, the repair is performed by the repair method specified in step S194.

  Then, in the next step S195, the repair count is incremented, and in the next step S196, the repair cost is added to the repair cost. The progress parameters are set by updating the values so that progress is faster. However, depending on the repair method, the same value as the new product may be set.

  Then, in the next step S198, it is determined whether or not the number of parts has become zero. If it is not zero, the process returns to step S192 and the number of parts set here is decremented, and the above description is repeated.

  If it is determined in step S198 that the previously set number of parts has been reached (that is, the number of parts has become zero), the process proceeds to the next step S199, where the repair count number, repair cost, crack size, occurrence Then, the site data including the progress parameter is reported, and the process returns to the flow of FIG. 2 in step S200.

  Next, in the case where the part adjacent to the part determined to be repaired in case B shown in FIG. 8 is repaired, first, in step S192, the same number of parts as in FIG. 7 is given, and in step S201. It is determined whether repair has been performed for each part. If the repair has been completed, the process proceeds to step S198. If the repair has not been performed, it is determined in step S202 whether the repair is necessary depending on whether the crack size is larger than the repair reference value.

  If the crack size is smaller than the repair reference value, it is determined that repair is not necessary, and the process proceeds to step S208. If it is larger, it is determined that repair is necessary, and an adjacent part is selected in step S203. However, parts that have already been repaired are not selected. In the next step S204, the number of adjacent parts is set, the part is repaired in step S204A, repair completed data is set in the corresponding part in step S205, and the repair cost is added to the repair cost in step S206.

  Then, in step S207, as in the case of FIG. 7, the crack generation and progress parameters are set by updating the values so that the generation parameters are more likely to be generated and the progress parameters are faster in the direction in which the progress is faster than in the case of a new product. . However, depending on the repair method, the same value as the new product may be set. Then, in the next step S208, it is determined whether or not the number of parts has become zero. If not, the process returns to step S204 to decrement the number of parts set here, and the above description is repeated.

  If it is determined in step S208 that the previously set number of parts has been reached (that is, the number of parts has become 0), the process proceeds to the next step S198, and it is determined again whether the number of parts has become 0, If it is not 0, the process returns to step S192 and the number of parts set here is decremented, and the above description is repeated.

  When it is finally determined in step S198 that the previously set number of parts has been reached (that is, the number of parts has become 0), the process proceeds to the next step S199, and the repair count is the same as in step S199 in FIG. The site data including the number, repair cost, crack size, occurrence, and progress parameters is reported, and the process returns to the flow of FIG. 2 in step S200.

  Finally, FIG. 9 shows the flow of cases C and D in the case where all the parts determined to be repaired are n or more and repaired or replaced with new parts. In these cases, first, when the number of parts is given in step S192, the number of parts that need repair is obtained in the next steps S215, S216, and S217, and depending on whether the number of parts is n or more in step S210. It is judged whether to repair all or replace with a new one.

  If it is determined in step S210 that the number of parts does not reach n and repair (new article replacement) is unnecessary, the process returns to the flow of FIG. 2 in step S200. If it is determined in step S210 that the number of parts is n or more and a repair (new article replacement) is necessary, the process proceeds to step S211 and all parts are repaired or replaced with new parts. In step S212, the repair cost is added to the repair cost. In addition, after repairing in step S213, the crack generation and progress parameters are updated and set in the direction in which the generation parameters are likely to be generated and the progress parameter is in a direction in which the progress is faster. Returning to the flow of FIG. However, depending on the repair method, the same value as the new product may be set.

  Thus, in the simulation of repair, even if cracks are expected in one part of multiple parts, only the part where cracks occur, including the peripheral parts, repair everything, replace with a new one, etc. In this case, an effective maintenance plan including a repair method adapted to actual operation can be created by simulating cost effectiveness including subsequent failure risks.

  Referring to FIG. 2 again, when the repair in step S19A is performed in this way, in step S35, the cost, repair rate, execution time, etc. are added by adding inspection, repair, new article introduction, preventive maintenance cost, etc. Is calculated, and it is determined in step S36 whether or not the number of years of operation has reached the number of years set in step S33. If not, the process returns to step S33 and the above description is repeated.

  If it is determined that the number of years set in step S36 has been reached, it is determined whether or not the number of Monte Carlo calculation cases given as calculation parameter data 10 has been reached in the next step S37. In this case, the process returns to step S32 and the above description is repeated, and if it is determined that it has been reached, the process proceeds to step S38.

  In this step S38, all of the cost, the repair rate, the execution time, etc. are totaled, the result is output in step S39, and the process ends in step S40.

  The above is the first embodiment of the equipment maintenance planning method for plant equipment according to the present invention. As can be seen from the above description, according to the first embodiment, the equipment maintenance plan for plant equipment is Corresponding to the actual inspection records for each part in the equipment of, the probability of occurrence of cracks based on the design data for each part, the degree of progress over time when cracks occur, the necessity and method of repair, etc. By simulating a collection of multiple parts taking into account the mutual correlation, it is possible to simulate the influence of parts that could not be handled in the past, improving the risk prediction accuracy and achieving optimal maintenance in terms of reliability and cost A plan can be provided.

(Second Embodiment)
Next, a second embodiment of the present invention will be described.
FIG. 10 is a detailed flowchart of the equipment maintenance planning method for plant equipment according to the second embodiment of the present invention, FIG. 11 is a detailed flowchart of crack generation calculation, and FIG. 12A is input data update. Detailed flow chart of calculation, FIG. 12 (B) is an explanatory diagram showing the change in the probability of damage with respect to the number of years of operation, FIG. 13 is a detailed flow chart of calculation of the environmental impact degree, and FIG. A detailed flowchart, FIG. 15 is a detailed flowchart of repair, and FIG. 16 is a detailed flowchart of the degree of influence on environment due to disposal. In these flowcharts, steps similar to those in FIG. 1 or the first embodiment are denoted by the same step numbers.

Compared to the first embodiment, the second embodiment adds data on an operation method 500 unique to a plant facility to a database as shown in FIG. 10, and based on the data, a plurality of parts of a gas turbine. The crack calculation step S510 for setting the thermal and mechanical operating conditions in the above and determining the occurrence of cracks and the size of the cracks is provided instead of step S11 (see FIG. 2) of the first embodiment. It is.
Hereinafter, the same reference numerals are given to the configurations described in the first embodiment, and description thereof is omitted.

  In FIG. 11 showing the detailed flow of the crack calculation step S510, calculation is started in step S501. Parameters for calculation are shown in step S503. In other words, this parameter includes the number of start / stop times, operating conditions including load fluctuations, an operation method unique to the gas turbine, a calculation model used for crack generation, and a progress calculation model used for crack progress calculation.

  This gas turbine specific operation method is data set for the difference in gas turbine operation method, that is, operation by diffusion combustion and operation by mixed combustion. Is different. For example, when operated by diffusion combustion, the gas turbine inlet temperature is high, the combustion stability is large, the vibration is small, and the NOx emission amount of exhaust gas is large. Is low, combustion stability is small, vibration is large, and NOx emission amount of exhaust gas is small.

  Therefore, for example, when setting the operation method to the operation by the diffusion combustion at the beginning and switching to the operation by the mixed combustion after the periodic inspection, and conversely the operation by the mixed combustion at the beginning, the operation by the diffusion combustion after the periodic inspection is performed. In the case of switching to, setting can be made such that only the operation by diffusion combustion for the entire period is performed from the beginning, or only the operation by mixed combustion for the entire period is performed from the beginning. Furthermore, the ratio of the operation by diffusion combustion and the operation by mixed combustion can also be set. In this way, a unique operation method is set in advance for the target gas turbine and reflected in the simulation.

When the parameters for crack calculation are given as described above, in the next step S505, thermal and mechanical operating conditions are set as operating state data from the operating method set as input data. That is, the temperature and stress conditions of the target part are calculated and obtained.
Then, based on the operating state data set in step S505, the occurrence of the occurrence is calculated in step S507, the occurrence of the occurrence is determined in step S507A, and the progress is calculated using the model formula in step S508. Ask. The presence or absence of cracks is the same as in step S12 of the first embodiment, and the plant operating condition 1 is composed of the operating time and the number of start / stop times, load fluctuation, number of input parts, etc. stored in the database, The part life data 2, the maintenance plan 5, the cost data 6 and the like are read out and the occurrence calculation is performed. In addition, the presence or absence of a crack in step S507A is the same as step S12A in the first embodiment.

Regarding the crack progress in step S508, the crack progress amount ΔL during the crack progress is generally calculated from the stress intensity factor K and the stress σ as shown in the equation (1).
ΔL = C × (K (σ)) ^M (1)
The stress intensity factor K in progress is not a constant value, but is expressed as in equation (2) using the stress σ. The stress value σ depends on the crack shape (size).
K = F × σ (2)
In the equations (1) and (2), F, C, and M are calculation parameters.

  Therefore, for example, when the operation method is set to operation method A (diffusion combustion operation) for the first five years and operation method B (mixed combustion operation) for the next five years, operation method B is performed for the first five years, and thereafter. In the case of the operation method A, the stress set as the operation state data in step S505 is different in each operation method, so that the final crack sizes in both cases are different.

In the next step S509, the presence / absence of cracks and the size of the cracks calculated in steps S507A and S508 are output, and the flow returns to the flow of FIG. 10 in step S511.
As described above, the damage probability can be calculated accurately by reflecting the operation method specific to the target gas turbine and the timing of the operation method in the simulation, and the optimum maintenance plan for each target gas turbine can be accurately performed. be able to.

Next, the input data update step S519 for improving the crack size and progress accuracy calculated in the crack calculation step S510 will be described.
This input data update step S519 is based on the difference between the failure probability of the simulation for the size of the crack obtained in the crack calculation step S510 and the failure probability obtained from the size of the crack detected by the inspection and the detected number. The input data stored in the database is updated. A detailed flow is shown in FIG.

  In FIG. 12A, calculation is started in step S520, and parameters for calculation are shown in step S521. This parameter includes calculation parameters such as the number of cases, number of start / stop operations, load fluctuation operation data, gas turbine combustion method, gas turbine specific operation method such as combustion system change timing, calculation model used for crack generation, and clearance This is a progress calculation model used for one progress calculation, inspection time, inspection method, inspection information consisting of the number of inspection target parts, etc., and a probability probability data at the time of inspection.

  Thus, when the parameters for the input data update step S519 are given, the number of calculation sets is set in step S522, and the number of years until the periodic inspection of the gas turbine is set in step S523. Then, the crack size is calculated as already described in the crack calculation step S510, which includes steps S505, S507, S507A, and S508.

  Next, the process proceeds to step S524, where an inspection is performed, thereby determining whether or not an error has been detected in step S525, and if detected, the number of damaged data is incremented in step S527. If it is determined in step S529 whether or not the number of years until the examination set in step S523 has been reached, the process returns to step S523 and the above description is repeated.

  If it is determined that the number of years set in step S523 has been reached, it is determined whether or not the number of Monte Carlo calculation cases given as calculation parameter data S521 has been reached in the next step S531, and has not been reached. In this case, the process returns to step S522 and the above description is repeated, and if it is determined that it has been reached, the process proceeds to step S533.

  In step S533, the failure probability is calculated from the size of the crack calculated in the crack calculation step S510. In the next step S535, the difference between the damage probability obtained in step S533 and the damage record based on the number of damaged data detected by the inspection is calculated. In the next step S537, the inspection method and the inspection execution time when the inspection is performed. Then, it is determined whether or not the number of objects to be inspected satisfies the condition for data adjustment.

Since the damage record is a result of inspection, the record value includes an error with respect to the true value. This error varies depending on the detection performance of the inspection. Therefore, it is necessary to adjust the input data based on the degree of coincidence between the damage record obtained from the record value and the damage probability of the simulation result obtained in step S533 in consideration of the detection performance of the inspection.
That is, in the inspection method with high detection performance and the result of the inspection with a long operation time, it is considered that the detection accuracy is large and the detection accuracy is high. Therefore, in such a condition, the input data in the database is adjusted. On the other hand, no adjustment is made in cases where it is difficult to detect, and even if the result is the same inspection method, for example, due to the inspection schedule, adjustment should not be performed when the number of inspection targets is small. To.

  If the inspection satisfies the adjustable condition in step S537, that is, if the inspection method with high detection performance, the inspection with a lot of operation time, or the number of inspection objects is large, the failure probability determined in step S535 and the inspection In step S539, it is determined whether or not the difference from the damage record based on the number of damaged data detected by the above is equal to or greater than a threshold value. If the threshold value is greater than or equal to the threshold value, the input data of the parameter data 2 in the crack generation / progress parameter data 2 in the database is reduced so that the difference from the actual value is reduced in step S541, that is, close to the actual value. Update.

  This adjustment method is performed by performing a simulation in which the data to be adjusted is changed as a parameter and selecting a combination that provides a result close to the actual value by a known optimization method.

In step S543, the updated data is output, and in step S545, the process returns to the flow of FIG.
A characteristic diagram showing how the breakage probability changes with respect to the operation years is shown in FIG. As shown in FIG. 12B, the difference H between the failure probability calculated based on the actual value and the failure probability calculated by the simulation tends to increase as the operation years elapse. A dotted line S indicates a simulation, a thin line T indicates a true value, and a thick line R indicates a result. There is a difference H between the simulation S and the performance R at the time of inspection.
As described above, the input data update step S519 adjusts the input data in the database to reflect the inspection result data, so that a maintenance plan with higher accuracy can be performed for each target gas turbine.

Next, step S600 for calculating the degree of influence on the environment using the data of the operation method 500 unique to the plant equipment set in the database will be described.
FIG. 13 shows a detailed flow of this environmental impact calculation. In the flow of FIG. 13, calculation of the influence on the environment is started in step S601, and parameters for calculating the influence on the environment are shown in step S602. In other words, this parameter includes operating conditions consisting of operating time, number of start / stops, number of load fluctuations, load fluctuation stress, etc., operating method consisting of NOx emissions (Yj) during mixed combustion and diffusion combustion of the gas turbine, This is an evaluation formula of the environmental impact level.

When the parameters are given in this way, in the next step S604, the environmental impact A due to driving is calculated by the environmental impact evaluation formula. Since the degree of influence is relatively evaluated in the simulation, a calculation formula or the like is defined by input.
Specifically, the total emission amount A = g (Yj) is obtained by the following equation (3), and the degree of influence by the total emission amount is defined as shown in Table 1 below, and Table 1 is calculated from the calculated total emission amount A. Use to determine the impact.
Total discharge A = discharge per unit time by operation method x operation time (3)

In step S606, the environmental impact A is output, and the process returns to the flow of FIG. 10 in step S608.
As described above, by comparing the environmental impact level A obtained in step S600 for calculating the environmental impact level, it is possible to perform an optimum maintenance plan from the viewpoint of the environment of each gas turbine and plant equipment. In addition, when the discharge amount is determined in advance, the operation method is limited, and the optimum operation and maintenance plan can be made from the viewpoint of reliability and cost.

Next, step S610 for calculating profits by selling energy using the data of the operation method 500 unique to the plant equipment set in the database will be described.
FIG. 14A shows a detailed flow of the calculation of profits from this energy sale. In the flow of FIG. 14A, profit calculation by energy sale is started in step S611, and parameters of profit calculation by the energy sale are shown in step S612.
In other words, this parameter depends on the gas turbine combustion method, the operation method unique to the gas turbine, such as the change timing of the combustion method, the fuel used, the amount of power generation, and the power sales model, timing, fuel unit price, This is a revenue calculation parameter value such as a power selling price.

When the parameters are given in this way, the profit is calculated by the power sale model (revenue calculation formula) defined by the input in the next step S614. A specific profit model for power sales is defined by input, and parameter values necessary for calculation are set by input.
As a specific example, calculation is performed by the following equation (4).
Revenue = Electricity unit price forecast value x Electricity generated by operation method
-Estimated fuel unit price x Amount of fuel used by operation method (4)
As shown in FIG. 14B, the current unit price or the predicted unit price is obtained based on the change state with respect to the time (year and month). Further, as shown in FIG. 14C, the current unit price or the predicted unit price is obtained for the fuel unit price based on the change state with respect to the time (year and month).

And the profit value by the power sale calculated by step S614 is output by step S616, and it returns to the flow of FIG. 10 by step S618.
As described above, by calculating and outputting the profit by selling the energy in step S610, it is possible to make an operation and maintenance plan in consideration of the profit from the power sale.

Next, with respect to the repair step S19A of the first embodiment, a repair step S700 for performing a simulation in consideration of the repair period will be described.
FIG. 15 shows a detailed flow of the repair step S700. In the flow of FIG. 15, the process starts from step S701, and the repair parameters are shown in step S702.
This parameter includes a repair method indicating the type of repair, designation of whether to prioritize cost or repair period, the latest information on the equipment storage site regarding equipment, days, cost, and the like.

  Then, when such repair parameters are given, in the next step S704, a repair method corresponding to the size of the crack is selected, and in step S706, the equipment necessary for the selected repair method is stored. In step S708, it is determined whether there is a designation of cost priority or period priority for repair. If there is a designation, the process proceeds to step S710, and if the priority is given to the cost, the site with the lowest cost is selected from the database, and if the priority is given to the period, the optimum site is selected from the database. Select. If there is no designation, the process proceeds to step S712, and a site is randomly selected from the sites where the devices are stored.

  In order to cope with priority designation of cost, period, etc., for example, as shown in Table 2 at site a and as shown in Table 3 at site b, the latest number of days required for equipment and repairs and repair costs Information is stored in the database as input parameters.

  Then, by using the databases shown in Tables 2 and 3, priority is given to the period when there are cases where necessary equipment is stored only at a relatively remote site and costs are high but the repair days are short. If specified as a simulation condition, it can be repaired by equipment from the site where the equipment is stored, and it is possible to respond to designation with priority on the repair period. Similarly, in the case of cost priority, by specifying the cost as a simulation condition, it can be repaired with equipment from a site that stores a low-cost device that takes a long time, but responds to the specification of cost priority. it can. If there is no priority designation, the site where the device is stored is selected with the same probability that each site is selected.

In step S714, the cost and the number of repair days are calculated from the repair method and the selected site, and the process proceeds to step S716. In step S716, the repair method data, crack generation after repair, and progress calculation parameters are updated and set.
In step S718, the repair method, the repair period, the repair cost, the crack generation calculation parameter after repair, and the progress calculation parameter are output, and the process returns to the flow of FIG. 10 in step S720.
As described above, the repairing step S700 makes it possible to perform an operation and maintenance plan in consideration of the repair cost, the repair period, and the repair location.

Furthermore, in the second embodiment, there is provided a step S800 of calculating the degree of environmental impact due to disposal, which calculates the degree of environmental impact due to the disposal of parts generated by equipment replacement.
FIG. 16 shows a detailed flow of the environmental impact calculation step S800 due to the discard. In the flow of FIG. 16, the process starts from step S801. In step S802, parameters for calculating the degree of influence on the environment due to the disposal are shown.
This parameter includes the disposal device type Zi, the number of types of disposal device n, the disposal device data indicating the disposal amount for each device Zi, the evaluation formula of the environmental impact degree by disposal, and the like.

When the parameters are given in this way, the influence degree B is calculated in the next step S804 by the evaluation expression of the environmental influence degree due to the disposal. Since the degree of influence B is relatively evaluated in the simulation, a calculation formula or the like is defined by input.
For example, the environmental impact degree B for each repair site is obtained by the following equation (5). If the number of repair parts is integrated, the environmental impact B for each gas turbine and each plant equipment can be obtained.
Note that the recycling rate of the device Zi is defined in advance by input as shown in Table 4 below.
Environmental impact B = Swaste amount of Σ device Zi × (1.0−recycling rate of device Zi) (5)
i = 1, n n: number of types of repair equipment

In step S806, the environmental impact degree B due to disposal is output, and in step S808, the process returns to the flow of FIG.
In this way, by comparing the environmental impact B obtained in the environmental impact degree calculation step S800 due to the disposal for each repair site or for each gas turbine, it is determined by the disposal for each repair site or for each gas turbine or plant equipment. It is possible to carry out an optimal maintenance plan considering the environmental impact. In addition, when the amount of disposal is determined in advance, the repair method can be limited, and among them, an optimal maintenance plan can be determined in terms of reliability and cost.

  According to the present invention, an optimal maintenance plan can be provided from the viewpoint of reliability and cost, and can greatly contribute to equipment maintenance of plant facilities.

It is a schematic flowchart of the equipment maintenance planning method of the plant equipment which becomes this invention. It is a detailed flowchart of the equipment maintenance planning method for the plant equipment of the first embodiment. It is a detailed flowchart of the crack generation calculation in 1st Embodiment. It is a detailed flowchart of the accident occurrence calculation in 1st Embodiment. It is a detailed flowchart of case A of preventive maintenance in the first embodiment. It is a detailed flowchart of case B of preventive maintenance in the first embodiment. It is a detailed flowchart of case A of the repair in 1st Embodiment. It is a detailed flow figure of case B of repair in a 1st embodiment. It is a detailed flowchart of repair cases C and D in the first embodiment. It is a detailed flowchart of the equipment maintenance planning method for plant equipment of the second embodiment. It is a detailed flowchart of the crack generation calculation in 2nd Embodiment. (A) is a detailed flowchart of the input data update calculation in 2nd Embodiment, (B) is explanatory drawing which shows the change of the failure probability with respect to the years of operation. It is a detailed flowchart of the environmental impact degree calculation in 2nd Embodiment. It is a detailed flowchart of the profit calculation by the energy sale in 2nd Embodiment. It is a detailed flowchart of the repair in 2nd Embodiment. It is a detailed flowchart of the influence degree to the environment by discard in 2nd Embodiment. (A) is a schematic plan view of a crack generated in a cooling hole of a tail cylinder in a gas turbine for a generator, and (B) is a schematic cross-sectional view of a hole in which the crack is generated. It is the schematic diagram which showed the process until the accident occurred.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Plant operating condition 2 Parts life data 3 Field performance 4 Analysis prediction value 5 Maintenance plan 6 Cost data 7 Simulation block 7 ₁ , 7 ₂ , 7 ₃ ,... 7 _n Multiple parts in plant equipment 8 Simulation result 9 Report Item 10 Calculation parameter data S11, S510 Crack calculation step S14 Accident occurrence step S18 Preventive maintenance step S19, S700 Repair step S519 Input data update step S600 Environmental impact calculation step S610 Revenue calculation step by energy sale S800 To environment by disposal Impact calculation step

Claims (15)

A step of presetting a specific part where the occurrence of a crack of each part of the plant equipment and the number of the specific parts is set in advance, and detecting the occurrence of the crack for each specific part set in advance; A step of calculating each progress based on a parameter designated in advance when it occurs, the presence / absence and occurrence direction of a crack in a part adjacent to the specific part and the length of the generated crack A crack calculation step comprising: a connection crack determination step for determining whether or not the calculated crack is a connection crack in which the progress of the crack progresses quickly depending on whether or not the crack distance between the parts is equal to or less than a threshold value. And the presence or absence of cracks for each specific part, the crack size of the generated part, and the position until the number of specific parts set previously in the crack calculation step for each specific part is reached. , Generating crack occurrence data including the presence or absence of occurrence of connected cracks, and based on the obtained crack occurrence data, a specific site and its neighboring neighboring sites (hereinafter referred to as “multiple sites that affect crack progress”). ) Predicting the degree of progress of cracks, and determining the probability of component damage and the occurrence of an accident corresponding to the “plurality of parts that affect crack progress” based on the predicted value of the degree of crack progress. An equipment maintenance planning method for a plant facility. Including the previous Symbol plant equipment each "multiple sites affect the crack growth each other" in the equipment, and inspection records, including inspection costs, and the occurrence probability data of crack based on the design data, the repair costs related to crack that occurred in the past Prepare a database that stores data and the cost of recovery in the event of an accident due to cracking ,
Based on the predicted value of the degree of progress of cracks of `` multiple parts that affect crack progress with respect to each other '', the determination of the probability of occurrence of part damage and accident occurrence corresponding to the plurality of parts is based on the predicted value. Performing the determination while increasing the probability of occurrence of an accident as it approaches the reference size of a crack that becomes an accident, and based on the determination result, “a plurality of sites that affect the progress of cracks with each other” that occurred in the past from the database. 2. The plant according to claim 1, wherein a maintenance plan reflecting accident risk is created by comparing repair costs related to cracks and costs required for recovery in the event of an accident due to crack progress. Equipment maintenance planning method for facilities. In predicting repairs based on the predicted value, repairing only the part where cracks occur, repairing including peripheral parts, repairing all of the parts, and replacing the equipment with a new one The equipment maintenance planning method for a plant facility according to claim 1 or 2 , wherein the risk of equipment damage and cost are calculated and reflected in the maintenance plan. The maintenance plan is stored in the database by storing periodic inspection means for preventing the occurrence of cracks, and causing the simulation to calculate the risk and cost of accidents that occur when the periodic inspection means for preventing the occurrence of cracks is performed. The apparatus maintenance planning method for a plant facility according to any one of claims 1 to 3 , wherein the apparatus maintenance plan method is reflected in the method. In the simulation for predicting the degree of progress of cracks in the above-mentioned “multiple sites that affect crack progress,” the probability of occurrence of an accident increases as the magnitude of the predicted crack becomes closer to the reference size of the crack. The equipment maintenance planning method for a plant facility according to any one of claims 1 to 3 , wherein a risk is calculated while increasing and reflected in the maintenance plan. The data of "multiple sites affecting crack growth from each other", the interval including the positional relationship of the mutual position information for each specific site relates cracks predicted to one site, the cracks occurring in the surrounding site , by the direction, a simulation incorporating a model growth rate of cracks is accelerated, equipment maintenance planning process plant equipment as claimed in any one of claims 1 to 5, characterized in that is reflected in the maintenance plan. 6. The plant maintenance equipment maintenance planning method according to claim 5 , wherein the simulation is performed using a Monte Carlo method. The plant according to any one of claims 1 to 7 , wherein the equipment of the plant facility is a transition piece of a gas turbine, and the crack is a plurality of cooling holes provided in the transition piece of the gas turbine. Equipment maintenance planning method for facilities. A plant equipment equipment maintenance plan method according to claim 1, wherein generating a maintenance plan of the plant equipment, has a database of each "multiple sites affects the crack growth from each other" in the plant equipment, the In the database, the operation method specific to the plant equipment is stored, and based on the operation method, the thermal and mechanical operation conditions in the “multiple parts that affect the crack progress with respect to each other” are set, and whether cracks are generated or not. A device maintenance planning method for plant equipment, wherein the size of cracks is obtained and the probability of breakage is simulated and reflected in the maintenance plan for the plant equipment. The plant equipment is a gas turbine, the operation method specific to the plant equipment is an operation by diffusion combustion and an operation by mixed combustion of the gas turbine, and the thermal and mechanical operating conditions are respectively in the diffusion combustion and mixed combustion. The equipment maintenance planning method for a plant facility according to claim 9 , wherein temperature and stress conditions are satisfied. When the difference between the failure probability obtained based on the inspection result and the failure probability obtained by the simulation is equal to or greater than a threshold value under a predetermined inspection condition, the database is brought closer to the failure probability obtained from the inspection result. The apparatus maintenance planning method for plant equipment according to claim 9 , wherein input data of thermal and mechanical operating conditions of “a plurality of sites that influence the progress of cracks” is adjusted. The plant facility is a gas turbine for power generation, and the database stores the operation method of the gas turbine, the fuel used, and the amount of power generation. By applying these data to a predetermined revenue model, 10. The equipment maintenance planning method for a plant facility according to claim 1 or 9 , wherein revenue is simulated. The plant equipment is a gas turbine, and the database stores harmful exhaust gas emissions such as NOx and CO in the operation method of the gas turbine, and simulates the environmental impact based on these data. An equipment maintenance planning method for a plant facility according to claim 1 or 9 . In the database, the time required for repair, the repair storage location and cost are stored required for repair of equipment, to claim 1 or 9, characterized in that to simulate the repair period and repair costs on the basis of these data Equipment maintenance planning method for the described plant equipment. The database stores data that sets the recycling rate for each device, and simulates the degree of impact on the environment when the device is replaced with a new one by the measures taken after the accident by the simulation or by the disposal by the inspection result An apparatus maintenance planning method for a plant facility according to claim 1 or 9 , characterized in that:

Images