JP2004191359A - Risk management device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform safety management of a plant facility according to the failure probability of each equipment constituting the plant facility. <P>SOLUTION: A design-stage failure probability function calculating part 11a obtains, for each equipment constituting a plant facility, a failure probability function which represents the relationship between the use years and the failure probability based on design data on the plant equipments at the design stage. A probability function changing par 11b modifies the failure probability function to prepare a modified failure probability function based on at least one of production data on production of the equipments, operation histories of the equipments and inspection on the equipments. A equipment exchange determining part 11c compares a modified failure probability obtained from the modified failure probability function depending on the use years with a preset value for determination to determine that the equipments should be exchanged when the modified failure probability exceeds the preset value, and manages the risk of the plant facility according to the modified failure probability. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

  INDUSTRIAL APPLICABILITY The present invention can be applied to various plants such as a nuclear power plant, a thermal power plant, and a petrochemical plant while ensuring safety and reliability in consideration of the risk information thereof (for example, economic expenditure, social loss, The present invention relates to a risk management device capable of reducing loss.

  Generally, at various plants such as nuclear power plants, thermal power plants, and petrochemical plants, periodic inspections and voluntary inspections are conducted to check for failures and damages of each device. Repair or replacement. The inspection content and inspection cycle for each device are determined in consideration of plant safety and the importance of each device.

  On the other hand, some devices with high importance have a low probability of failure (failure probability: for example, a failure probability function with the service life taken on the horizontal axis and the probability of failure taken on the vertical axis). If the inspection content and the inspection cycle are determined in accordance with the above, if the importance is low, the inspection content will be simplified and the inspection cycle will be long if the probability of failure is high.

  Conventionally, there is a system in which maintenance management of plant equipment is performed in consideration of a failure probability. In this case, a failure rate, an evaluation time, a failure occurrence probability evaluation formula corresponding to the equipment / failure mode, a Birnbaum (Birnbaum) Using the importance and the failure occurrence probability value of the entire plant or system, approximately calculate the failure occurrence probability and the values of various importance evaluation formulas, quantitatively judge important equipment, and perform recommended maintenance work. Guidance is output (see Patent Document 1).

  Further, in order to predict the deterioration of the metal component based on the usage environment, material characteristics, the number of times of starting and stopping, etc. of the metal component, and create an optimal maintenance plan of the power plant, data from the three-dimensional data and maintenance personnel are used. Based on the instruction, select the parts to be repaired, search the relevant information for the selected parts, extract the relevant parts related information for that part, and input the relevant parts related information and remaining life prediction input data. The remaining life of the part is predicted in accordance with the above, remaining life information is generated, and what kind of maintenance (repair, replacement, etc.) is performed on the part based on the remaining life information and the LCC minimization calculation input data. There is one that calculates whether it is possible to minimize the life cycle cost of a power plant, and creates an optimal maintenance plan (see Patent Document 2).

  In addition, when determining the maintenance method for individual equipment and devices that make up a large-scale plant, quantitative evaluation is performed to support plant design. It stores the evaluation data, calculates the expected loss value in the event of a major accident based on the safety evaluation data, stores the operating rate evaluation data that is the probability data of the event leading to the plant shutdown, and stores it in the operating rate evaluation data. The expected value of the loss when the plant is stopped is calculated based on the calculated value.

  Then, the two types of expected loss values are added, at least one maintenance method is selected, the cost required when each maintenance method is performed is calculated, and the added value of the expected loss value and the selected maintenance value are calculated. There is a method in which the amount of change in cost is compared, and the maintenance cost is less than or equal to the sum of expected loss values and a minimum maintenance method is selected, and this is set as an optimal maintenance method (Patent Document 3). reference).

  In addition, during the maintenance management of plant equipment, the quality of the parts to be maintained is quantitatively grasped and the optimal design and maintenance are carried out. Is analyzed by advanced technology and quantitatively evaluated, and the failure of the moving equipment is quantitatively evaluated by the functional failure mode effect analysis. There is a method in which the frequency of occurrence of a failure and the degree of influence due to the occurrence of the failure are set using an evaluation method, and (frequency × influence degree) is evaluated as a risk (see Patent Document 4).

JP-A-2000-2785 (paragraphs (0010) to (0013), FIG. 4) JP-A-2002-297710 (paragraphs (0021) to (0053), FIG. 1, FIG. 2, FIG. 4, FIG. 6, and FIG. 7) JP-A-8-77211 (paragraphs (0009) to (0032), FIGS. 1 to 3) JP-A-2002-123314 (paragraphs (0021) to (0037), FIG. 1 and FIG. 4)

  In conventional reliability evaluation or maintenance risk management of plant equipment (hereinafter simply referred to as conventional risk management), important equipment is determined in consideration of the failure probability of each equipment, and recommended maintenance work is performed for each important equipment. Although guidance is selected, the failure probability of each device varies depending on the design, manufacture, operation, and inspection of the plant. Therefore, in a plant where the failure probability is regarded as important (for example, a nuclear power plant), it is difficult to determine what the failure probability is when it is better to replace the equipment, and to secure the reliability of the plant. There is a problem that it is difficult to perform (that is, to ensure safety management). In other words, there is a problem that it is difficult to appropriately manage the risks related to the plant.

  Furthermore, in conventional risk management, for example, comparing the cost incurred in the event of an accident (cost at the time of an accident: potential cost) with the cost incurred by replacement and the cost of inspection, it is necessary to replace equipment in any state. However, it is difficult to determine whether the cost is the lowest, and as a result, there is a problem that it is difficult to reduce the cost related to maintenance.

  In addition, in conventional risk management, risk management has not been performed by evaluating social impacts, and the failure probability curve (breakage probability curve) has been corrected by reflecting plant characteristics such as operating conditions. There is a problem that optimal risk assessment cannot be performed.

  In addition, in conventional risk management, it is not only difficult to formulate a repair plan by quantitatively evaluating the inspection value, but also it is not possible to select the number of inspections, the inspection time interval, and the inspection method according to the inspection results. There is also a problem that it is not possible to evaluate the allowable service life according to the inspection result and to select a plant maintenance method according to the evaluation result. In addition, the conventional risk management also has a problem that it is not possible to evaluate the investment effect on plant equipment in consideration of the influence on the probability of breakage due to technology development.

  An object of the present invention is to provide a risk management device capable of reliably performing plant safety management according to a failure probability.

  Another object of the present invention is to provide a risk management device capable of reliably performing plant safety management while reducing maintenance costs.

  Still another object of the present invention is to provide a risk management apparatus and a risk evaluation method capable of solving various problems described above and performing optimal risk management in plant equipment.

  According to the present invention, there is provided a risk management apparatus for managing the risk of plant equipment, wherein each of the equipment constituting the plant equipment has been in service for each of the equipment in a design stage according to design data relating to the equipment. And a failure probability function calculating means for determining a failure probability function indicating a relationship between the failure probability and the failure probability function based on at least one of manufacturing data relating to the manufacture of the device, an operation history of the device, and an inspection regarding the device. A failure probability function changing unit that corrects the failure probability function to be a modified failure probability function, and compares a modified failure probability obtained from the modified failure probability function according to a service life with a predetermined determination value. Judging means for judging replacement of the equipment when the corrected damage probability exceeds the judgment value, wherein the plant setting is performed in accordance with the corrected damage probability. You have to manage the risk risk management system is obtained characterized by. For example, the failure probability function changing means corrects the failure probability function according to the inspection accuracy in the inspection.

  Further, in the present invention, when the device is replaced in accordance with the replacement determination by the determining unit, the service life of the device replaced once by temporarily resetting the corrected damage probability function corresponding to the replaced device. And reset reset means for moving the modified failure probability function so that the start point becomes the starting point.

  According to the present invention, there is provided a risk management apparatus for managing the risk of plant equipment, wherein the cost and the service life when an accident has occurred with respect to the equipment at the design stage for each equipment constituting the plant equipment. A cost function calculating means for obtaining a cost function indicating the relationship, and correcting the cost function based on at least one of a manufacturing cost related to manufacturing the device, an inspection related to the device, and an operation history of the device. A cost function changing means for setting the corrected cost function as a corrected cost function, and comparing the corrected cost obtained from the corrected cost function with a predetermined judgment value according to the service life. Determination means for determining the replacement of the plant equipment, and manages the risk of the plant equipment according to the correction cost. Risk management device that can be obtained. For example, the cost function changing means corrects the cost function according to the inspection accuracy in the inspection.

  Further, according to the present invention, when the device is replaced in accordance with the replacement determination by the determination unit, the service life for which the device was replaced by once resetting the corrected cost function corresponding to the replaced device is reset. There is reset changing means for moving the correction cost function so as to be a starting point.

  Further, in the present invention, the cost function calculating means, the design cost required in the design stage, the manufacturing cost, and the service life of the device according to the inspection cost required for the inspection and the design cost, the manufacturing cost, the manufacturing cost, An operation cost function indicating a relationship with an inspection cost is obtained, the cost function change unit corrects the operation cost with an operation / monitoring cost required for operating and monitoring the device, and the reset change unit includes the determination unit. When the replacement of the device is performed according to the replacement determination by the device, the operation cost function is corrected according to the replacement cost required for replacing the device.

  According to the present invention, there is provided a risk management apparatus for managing the risk of plant equipment, wherein each of the equipment constituting the plant equipment has been in service for each of the equipment in a design stage according to design data relating to the equipment. A function for calculating a cost function indicating the relationship between the cost and the service life when an accident has occurred with respect to the device, and a function for calculating a damage probability function indicating the relationship between the device and the probability of damage. Manufacturing data related to the manufacturing of the device, the manufacturing data, the operation history of the device, and correcting the failure probability function based on at least one of the inspections for the device as a corrected failure probability function and the cost function. Modifying the cost function based on at least one of an inspection of the device and an operation history of the device, and A function changing means for setting the function and a degree of importance indicating the importance of each device as a correction coefficient, and correcting the corrected damage probability obtained from the corrected damage probability function according to the service life with the correction coefficient. Correction means for the probability of damage, and a value obtained from the selection function according to the service life is preset by using the correction cost function and the damage probability function selectively as a selection function according to the corrected damage probability. A determining means for determining the replacement of the equipment when the determined value is exceeded, and managing the risk of the plant equipment in accordance with the corrected failure probability function and the corrected cost function. A management device is obtained.

  Further, the present invention is a risk management device for evaluating and managing secondary damage as a secondary risk in consideration of the social impact of plant equipment, and performs a past similar case analysis on plant equipment. Assumption setting means for setting assumptions based on the analysis result, evaluation means for evaluating at least economic loss and social loss according to the assumption setting to obtain an evaluation result, and using the evaluation result as a loss element And a risk estimating means for estimating a risk cost.

  In the present invention, at least human damage, property damage, in-house equipment damage, operation loss, business compensation, inspection compensation, and increase in power generation cost are evaluated as the economic loss, and at least traffic obstruction and evacuation are considered as the social loss. And third party business losses should be assessed.

  In the present invention, when managing the risk of plant equipment, a risk management device that performs the risk management of the plant equipment according to the characteristics of the plant equipment, the service life of each device constituting the plant equipment and A correction means for obtaining a corrected failure probability function by correcting the original failure probability function according to the characteristics of the plant equipment, using a failure probability function indicating a relationship with the failure probability as an original failure probability function, and And a risk management means for managing the risk of the plant equipment.

  In the present invention, the correction means includes a first means for obtaining a damage occurrence time expression represented by a reciprocal of the damage sensitivity of the equipment material, a second means for identifying a correction value according to the damage occurrence time expression, A third means for obtaining a temperature and stress distribution for the device according to the distribution, and a fourth means for obtaining a damage occurrence time distribution by a Monte Carlo method according to the damage occurrence time formula, the temperature and stress distribution, and the correction value. And a fifth means for adapting the failure time distribution to a predetermined probability distribution to obtain the corrected failure occurrence probability curve.

  According to the present invention, there is provided a risk management apparatus that manages the risk of the plant equipment by quantitatively evaluating the value related to the inspection of the plant equipment when managing the risk of the plant equipment. A first optimizing means for optimizing a repair plan using the current failure probability function with a current failure probability function indicating a relationship between a service life and a failure probability for each device as a current failure probability function; When an inspection is performed using a plurality of inspection methods different from each other using the current failure probability function, an inspection result prediction unit that calculates a plurality of expected inspection results for each inspection method, and an inspection update is performed on each of the expected inspection results. A failure probability function updating means for obtaining a later failure probability curve as an updated failure probability function; and a repair plan for each of the updated failure probability functions using the updated failure probability function. A second optimizing means for optimizing to obtain a plurality of optimizing measures, a risk valuation amount for an expected inspection result for the optimizing measures, a risk valuation amount at the time of performing the inspection, and a first optimizing means A risk management apparatus characterized by having evaluation means for evaluating the value of the test by comparing the risk amount of the optimization measure obtained by the means.

  In the present invention, when managing the risk of plant equipment, according to the number of inspections for each plant equipment, the inspection interval, and the inspection method, for each device constituting the plant equipment, the relationship between the service life and the probability of breakage A risk management device for performing risk management of the plant facility by updating the failure probability function shown in the table, wherein the probability that the inspection result will be obtained based on the inspection result according to the type of the failure probability for each device constituting the plant facility. Means as a first probability, second means for obtaining the probability of each type of failure probability as a second probability for the inspection result using Bayes' theorem, and the second means by updating based on Bayes theory. A third means for setting the probability of the third failure probability to the third probability of the failure probability type, and a fourth means for determining the failure probability function as an average of the third failure probability. Risk management apparatus characterized by having a means is obtained.

  According to the present invention, there is provided a risk management apparatus for selecting a maintenance method in accordance with an inspection result for each of the plant equipment and managing the risk of the plant equipment in accordance with the selected maintenance method when managing the risk of the plant equipment. The first means for selecting the inspection period for each plant equipment, the second means for selecting the inspection method for each plant equipment, and the inspection method selected by the second means are used. And a third means for obtaining an expected inspection result at the time of updating, and updating and updating an original failure probability curve showing a relationship between a service life and a failure probability for each device constituting the plant equipment based on the expected inspection result. A fourth means for obtaining a damage probability curve, and a fifth means for obtaining an optimal measure corresponding to the updated damage probability curve and calculating an estimated cost amount, Risk Management device is obtained which is characterized in that so as to select a maintenance method according to the virtual cost.

  In the present invention, a risk management device that manages the risk of the plant equipment by evaluating the maintenance effect of the plant equipment by technology development when managing the risk of the plant equipment, and for each device constituting the plant equipment A first means for obtaining a damage probability distribution indicating a relationship between the service life and the damage probability, a second means for correcting the damage probability distribution by introducing the technical development to obtain a corrected damage probability function, A risk management device characterized by having a third means for calculating the amount of risk reduction when the corrected failure probability distribution is used, and further according to the input time of the technical development and the application period after the input. It is desirable to have a fourth means for calculating an investment amount for the technology development based on the accumulated risk reduction amount.

  As described above, in the present invention, a failure probability function indicating the relationship between the service life and the failure probability for each device according to the design data related to the device at the design stage for each device constituting the plant facility is determined, The damage probability function is corrected based on at least one of the manufacturing data relating to the manufacture of the device, the operation history of the device, and the inspection of the device as a corrected damage probability function, and is obtained from the corrected damage probability function according to the service life. When the corrected damage probability exceeds the determination value by comparing the corrected damage probability with a predetermined determination value, the replacement of the device is determined, and the risk of the plant equipment is managed according to the corrected damage probability. In addition, there is an effect that the safety management of the plant can be reliably performed according to the failure probability.

  According to the present invention, a cost function indicating the relationship between the cost and the service life when an accident occurs in the equipment at the design stage is determined for each equipment constituting the plant equipment, and the cost function is used for manufacturing the equipment. A cost, an inspection related to the device, and a correction cost function that is corrected based on at least one of the operation history of the device and a correction cost obtained from the correction cost function according to the service life and a predetermined determination value. If the repair cost exceeds the judgment value, the replacement of the equipment is determined, and the risk of the plant equipment is managed according to the repair cost. There is an effect that it can be performed reliably.

  Further, according to the present invention, an operation cost function indicating the relationship between the service life of the device and the design cost, the manufacturing cost, and the inspection cost according to the design cost required in the design stage, the manufacturing cost, and the inspection cost required for the inspection. The operating cost is corrected according to the operation and monitoring costs required for the operation and monitoring of the equipment, and when the equipment is replaced according to the replacement determination, the operating cost function is corrected according to the replacement cost required for the replacement of the equipment. Thus, there is an effect that the cost for operating the device can be easily grasped in accordance with the years of service.

  According to the present invention, the degree of importance indicating the importance of each device is used as a correction coefficient, and the corrected damage probability obtained from the corrected damage probability function according to the service life is corrected with the correction coefficient to obtain the corrected damage probability. If the value obtained from the selection function according to the number of years of use exceeds a predetermined determination value, the replacement of the device is determined selectively using the modified cost function and the failure probability function as the selection function according to the probability. In addition, since the risk of the plant equipment is managed in accordance with the repair failure probability function and the repair cost function, not only the cost can be reduced, but also the safety management of the plant can be reliably performed in accordance with the failure probability. There is.

  According to the present invention, similar past case analysis relating to plant equipment is performed, assumptions are set in accordance with the analysis results, and economic and social losses are evaluated in accordance with the assumptions to obtain evaluation results. Since the risk cost is estimated using the result as a loss factor, there is an effect that the cost when an accident or the like occurs in the plant equipment can be calculated in advance.

  In the present invention, the original failure probability function is corrected according to the characteristics of the plant equipment to obtain a corrected failure probability function, and the risk management of the plant equipment is performed based on the corrected failure probability function. This has the effect that risk management can be performed accurately.

  In the present invention, the repair plan is optimized using the current failure probability function, and further, when inspection is performed using a plurality of inspection methods different from each other using the current failure probability function, a plurality of inspection methods are provided for each inspection method. Calculate the expected inspection result, calculate the damage probability curve after the update for each of the expected inspection results as an updated failure probability function, and then optimize the repair plan for each updated failure probability function using the updated failure probability function Of the optimization measures, the risk evaluation amount for the expected inspection results for these optimization measures and the risk evaluation amount when the inspection is performed, and the risk amount of the optimization measures obtained by the first optimization means. Are compared to evaluate the value of the test, so that it is easy to determine whether the test is worthwhile.

  According to the present invention, the probability of the inspection result is determined as the first probability based on the inspection result according to the type of the failure probability for each component constituting the plant equipment, and each failure probability is determined for the inspection result using Bayes' theorem. The probability of the type is determined as a second probability, and further, the second probability is determined as the third probability of failure of the failure probability type by updating based on Bayes theory, and the failure probability function is determined as an average of the third probability of failure. Therefore, there is an effect that the breakage time of each device can be easily determined according to the breakage probability function.

  In the present invention, the inspection period is selected for each plant facility, the inspection method is selected for each plant facility, an expected inspection result when the selected inspection method is used is determined, and the original failure probability curve is calculated. After updating based on the expected inspection result and obtaining the update failure probability curve, determine the optimal measure according to the update damage probability curve, calculate the expected cost amount, and optimize the countermeasure and the maintenance method according to the expected cost Is selected, there is an effect that the maintenance method can be accurately selected for each plant facility.

  In the present invention, after correcting the failure probability distribution by introducing the technology development, a corrected failure probability function is obtained, and a risk reduction amount when the corrected failure probability distribution is used is calculated. Since the amount of investment in technology development is calculated based on the cumulative risk reduction according to the application period of technology, there is an effect that the amount of investment due to technology development and the technology development introduction can be estimated.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention thereto, unless otherwise specified. It is only an example.

  First, referring to FIG. 1, the illustrated risk management device 10 includes a risk calculation processing device 11, a display device 12 such as a display, an input device 13 such as a keyboard, and an output device 14 such as a printer. The arithmetic processing unit 11 is connected to a design information database 21, a manufacturing information database 22, an inspection method accuracy database 23, and an inspection result database 24, and is provided with monitoring information described later.

  Referring to FIG. 2, in a plant such as a nuclear power plant, first, various devices constituting the plant are designed and the whole plant is designed. In this design, for example, in consideration of the skill of the designer for each device, the probability of breakage of the device according to the service life of the plant (year of disposal) is obtained, and the design time shown in FIG. It is stored in the design information database 21 as a failure probability function.

  For example, when a designer's skill level and service years classified by rank are input from the input device 13, the design-time damage probability function calculation unit 11a outputs a damage probability function in which a past damage probability function for similar plant equipment is stored. With reference to a case database (not shown), a design failure probability function of the device is determined (design: step S1).

  When the above-described design stage is completed, plant equipment is manufactured. In this manufacturing process, the design may be changed as necessary, and the equipment may be manufactured according to the skill level of the worker. Changes the probability of failure. Then, when an error at the time of manufacturing (error between the design value and the manufacturing value) and the skill of the worker are input from the input device 13, the failure probability function changing unit 11b outputs the manufacturing error and the skill of the worker for each plant device. The failure probability function at the time of design is updated in accordance with, and a failure probability function at the time of manufacture is generated (manufacturing: step S2).

  Then, the manufacturing probability function is stored in the manufacturing information database 22. For example, in the above-mentioned damage probability function case database, the correction rate is stored in accordance with the manufacturing error and the worker skill, and the break probability function changing unit 11b stores the manufacturing error and the worker skill for each. The design failure probability function is updated according to the correction rate to generate a manufacturing failure probability function (see FIG. 3B).

  When the above-described manufacturing stage is completed, the plant is assembled and installed, and a trial operation is performed, for example, a pre-service inspection is performed (in some cases, a pre-service inspection is not performed: Step S3). In this way, when the plant is commissioned, whether the pre-service inspection is to be performed is input from the input device 13 and the inspection method and its accuracy are input. If the probability function changing unit 11b determines that there is no pre-service inspection (step S4), the probability function function is changed to the device damage probability function without changing the manufacturing-time damage probability function.

  On the other hand, when the pre-service inspection is performed, the probability function changing unit 11b reads a correction rate corresponding to the inspection method and accuracy input from the input device 13 from the inspection method accuracy database 23, and according to the correction rate. The manufacturing breakage probability function is updated to be a device breakage probability function.

  If it is determined that there is no pre-service inspection, the probability function changing unit 11b outputs a device failure probability function shown in FIG. 3C without inspection and stores it in the inspection result database 24. On the other hand, if it is determined that the inspection is a low-precision inspection, the probability function changing unit 11b corrects the manufacturing failure probability function at a correction rate corresponding to the low-precision inspection (low-precision inspection: step S5), and FIG. ), An equipment failure probability function indicated by the low-precision inspection is output and stored in the inspection result database 24.

  Further, when it is determined that the inspection is a high-precision inspection, the probability function changing unit 11b corrects the manufacturing failure probability function at a correction rate corresponding to the high-precision inspection (high-accuracy inspection: step S6), and FIG. ), A device failure probability function indicated by the high-precision inspection is output and stored in the inspection result database 24.

  When the pre-service inspection is completed, the plant will be put into operation. For example, during plant operation, the status of each device of the plant may be monitored (monitored) to obtain monitoring information. When each device of the plant is monitored, monitoring information indicating that the device is monitored is given to the risk calculation control device 11 for the monitored device. Then, the probability function changing unit 11b changes the device failure probability function at a preset correction rate for the device corresponding to the monitoring information (operation: step S7).

  For example, as for the device corresponding to the monitoring information, as shown in FIG. 3D, the device failure probability function is moved downward based on a preset correction rate to reduce the failure probability. Then, the changed device failure probability function is stored in the inspection result database 24, and the inspection result database 24 is updated.

  During the operation of the plant, the equipment is inspected as necessary (in-service inspection: step S8). When performing the in-service inspection, whether or not to perform the in-service inspection is input from the input device 13 and the inspection method and its accuracy are input. When determining that there is no in-service inspection (step S9), the probability function changing unit 11b does not change the device damage probability for the device.

  On the other hand, when the in-service inspection is performed, the probability function changing unit 11b reads the correction rate corresponding to the inspection method and the accuracy input from the input device 13 from the inspection method accuracy database 23, and according to the correction rate. The inspection result database 24 is updated by changing the equipment failure probability function.

  As shown in FIG. 3E, when it is determined that there is no inspection in service, the inspection result database 24 stores an equipment failure probability function shown in FIG. If it is determined that the inspection is a test, the probability function changing unit 11b corrects the device failure probability function at a correction rate corresponding to the low-precision inspection (low-precision inspection: step S10), and FIG. The inspection result database 24 is updated as a device failure probability function indicated by.

  Further, when it is determined that the inspection is a high-precision inspection, the probability function changing unit 11b corrects the device failure probability function at a correction rate corresponding to the high-precision inspection (high-precision inspection: step S11), and FIG. Then, the inspection result database 24 is updated as a device failure probability function indicated by the high-precision inspection.

  In this way, after the in-service inspection is performed, when the determination command is given to the risk calculation control device 11 from the input device 13, the device replacement determination unit 11c uses the inspection result database 24 to check the device damage for each device. The probability function is read, and a failure probability corresponding to the service life is obtained from the equipment failure probability function. Then, a comparison is made between the predetermined judgment value (criterion) and the probability of damage, and it is checked whether or not the condition ≧ the probability of damage (step S12). At this time, the criterion is compared with the probability of breakage when the device is used up to the service life until the next inspection.

  If Criteria ≧ damage probability, the device replacement determination section 11c displays on the display device 12 that device replacement is unnecessary. On the other hand, when the condition of “criterion <damage probability” is satisfied, the device replacement determination unit 11c displays on the display device 12 that the device replacement is required. In this way, after determining whether or not each device has been replaced based on the device failure probability function, the device replacement determination unit 11c prints out the replacement device list to the output device 14.

  After the corresponding device is replaced based on the replacement device list, device replacement end information is input from the input device 13 (information indicating the replacement device is added to the device replacement end information). ), The failure probability resetting unit 11d resets the failure probability function corresponding to the replacement device (see FIG. 3 (f)), and uses the service life corresponding to the intersection of the criterion and the failure probability as the origin, and the device failure probability function Is updated (step S13) and stored in the inspection result database 24 (see FIG. 3 (g)).

  In this way, the failure probability function for each device set at the design stage is updated according to the manufacturing stage, the pre-service inspection, operation, and the in-service inspection, and according to the updated failure probability function and the criteria. Since it is determined whether or not to replace equipment, it is possible to make detailed replacement determination according to the risk of the plant equipment, that is, the probability of breakage, and as a result, safety management of the plant is performed according to the probability of failure. It can be done reliably.

  Next, another example of the risk management apparatus according to the present invention will be described with reference to FIG. In the illustrated risk management device 30, the same components as those of the risk management device shown in FIG. 1 are denoted by the same reference numerals. The risk management device 30 has a risk calculation processing device 31. The risk calculation processing device 31 includes a design information database 41, a manufacturing information database 42, an inspection method / accuracy / inspection cost database 43, and an inspection result database 44. In addition to being connected, an accident cost / design cost database 45, a manufacturing cost database 46, an operation / monitoring cost database 47, and a replacement cost database 48 are connected.

  Referring to FIG. 2 as well, the accident cost / design cost database 45 stores accident cost information required when an accident (failure) occurs for each device, and stores design cost information for each device. Is stored. The cost at the time of an accident increases as the years of service elapse.

  In the risk management device shown in FIG. 4, when device information (information for specifying a device: device ID or the like) is given from the input device 13 in step S1 shown in FIG. The accident cost and the design cost corresponding to the ID are read from the accident cost / design cost database 45, and the potential cost of the equipment according to the service life of the plant (year of disposal) is determined based on the accident cost. 5 (a) is stored in the design information database 21 as a potential cost function at design (potential cost = cost at accident).

  Next, in step S2 shown in FIG. 2, the cost function changing unit 31b reads the manufacturing cost related to the device from the manufacturing cost database 46 according to the device ID, and modifies the potential cost according to the manufacturing cost. Generate a manufacturing cost function. Then, the manufacturing cost function is stored in the manufacturing information database 42 (see FIG. 5B).

  In step S4 shown in FIG. 2, when the cost function changing unit 31b determines that there is no pre-service inspection, the inspection result database is used as an equipment cost function without changing the manufacturing cost function (see FIG. 5C). 44.

  Further, the cost function changing unit 31b reads the corresponding inspection cost without inspection from the inspection method / accuracy / inspection cost database 43 and, together with the above-described design cost and manufacturing cost, the inspection cost when there is no inspection indicating the secular change of the inspection cost. A design / manufacturing / inspection cost function (operation cost function) related to the design / manufacturing / inspection cost is generated (see FIG. 5D) and stored in the inspection result database 44.

  Since the design cost and the manufacturing cost are costs generated at the time of design and at the time of manufacture, respectively, the design / manufacturing / inspection cost function changes depending on the inspection cost.

  On the other hand, when the pre-service inspection is performed, the cost function changing unit 31b reads the correction rate corresponding to the inspection method and accuracy input from the input device 13 from the inspection method / accuracy / inspection cost database 43, and performs the correction. The manufacturing cost function is updated according to the rate, and stored in the inspection result database 44 as an equipment cost function. For example, when the pre-service inspection is determined to be a low-precision inspection, the cost function changing unit 31b corrects the manufacturing cost function at a correction rate corresponding to the low-precision inspection in step S5 shown in FIG. The device cost function indicated by the low-precision inspection is output to (c) and stored in the inspection result database 44.

  At this time, the cost function changing unit 31b reads the inspection cost corresponding to the low-precision inspection from the inspection method / accuracy / inspection cost database 43, and, together with the above-described design cost and manufacturing cost, a low-accuracy indicating the aging of the inspection cost. A design / manufacturing / inspection cost function related to the design / manufacturing / inspection cost at the time of inspection is generated (see FIG. 5D) and stored in the inspection result database 44.

  Further, when it is determined that the inspection is the high-precision inspection, the cost function changing unit 31b corrects the manufacturing cost function at a correction rate corresponding to the high-precision inspection in step S6 shown in FIG. The equipment cost function indicated by the high-precision inspection is output and stored in the inspection result database 24. At this time, a design / manufacturing / inspection cost function related to the design / manufacturing / inspection cost at the time of high-precision inspection is generated (FIG. 5D) and stored in the inspection result database 44.

  In step S7 shown in FIG. 2, the cost function changing unit 31b reads the operation / monitoring cost from the operation / monitoring cost database 47 for the device corresponding to the monitoring information, and reads the device cost function and the design / manufacture for the device. / Modify the operation inspection cost function. As shown in FIG. 5E, for equipment corresponding to the monitoring information, the equipment failure probability function is moved downward based on the operation / monitoring cost to reduce the potential cost.

  On the other hand, as shown in FIG. 5F, the design / manufacturing / inspection cost function is moved upward based on the operation / monitoring cost to increase the design / manufacturing / inspection cost. That is, the operation / monitoring cost is added to the design / manufacturing / inspection cost, and the function becomes a design / manufacturing / inspection / operation cost function. Then, the changed equipment cost function and design / manufacturing / operation / inspection cost function are stored in the inspection result database 44, and the inspection result database 44 is updated.

  Further, in step S9, when the cost function changing unit 31b determines that there is no in-service inspection, the cost function changing unit 31b does not change the device cost function for the device (FIG. 5 (g)). On the other hand, when the in-service inspection is performed, the cost function changing unit 31b reads a correction rate corresponding to the inspection method and accuracy input from the input device 13 from the inspection method / accuracy / inspection cost database 43, and corrects the correction. The inspection result database 44 is updated by correcting the equipment cost function according to the rate.

  For example, if the in-service inspection is determined to be a low-precision inspection, the cost function changing unit 31b corrects the device cost function at a correction rate corresponding to the low-precision inspection in step S10 shown in FIG. In g), the inspection result database 44 is updated with the equipment cost function indicated by the low-precision inspection. At this time, the cost function changing unit 31b reads the inspection cost corresponding to the low-precision inspection from the inspection method / accuracy / inspection cost database 43 and corrects the design / manufacturing / operation / inspection cost (FIG. 5 (h)). ), And updates the inspection result database 44.

  Similarly, if it is determined that the inspection is a high-precision inspection, the cost function changing unit 31b corrects the equipment cost function at a correction rate corresponding to the high-precision inspection in step S11 shown in FIG. 2 (see FIG. 5G). ), The inspection result database 24 is updated. At this time, the design / manufacturing / operation / inspection cost function is corrected according to the inspection cost corresponding to the high-precision inspection (see FIG. 5H), and the inspection result database 44 is updated.

  In step S12, in the risk calculation control device 31, the device replacement determination unit 31c reads the device cost function and the design / manufacturing / operation / inspection cost function for each device from the inspection result database 24, and outputs the device cost function. From the design / manufacturing / operation / inspection cost function, the potential cost and the design / manufacturing / operation / inspection cost corresponding to the service life are obtained. Then, a comparison is made between a predetermined determination value (criterion) and the potential cost to check whether or not the condition is equal to or greater than the potential cost.

  When the condition ≧ the potential cost, the device replacement determination unit 31c displays on the display device 12 that the device replacement is unnecessary. On the other hand, when the condition of “criterion <potential cost” is satisfied, the device replacement determination unit 31c displays on the display device 12 that the device replacement is required. In this way, after determining whether or not each device has been replaced based on the device failure probability function, the device replacement determination unit 31c prints out the replacement device list to the output device 14.

  In step S13 shown in FIG. 2, the cost reset / correction unit 31d resets the equipment cost function corresponding to the replacement equipment, and updates the equipment cost function with the service life corresponding to the intersection between the criteria and the potential cost as the origin. Then, the replacement cost is stored in the inspection result database 24 (see FIG. 5 (i)), and the replacement cost for the equipment replacement is read from the replacement cost database 48, and the replacement cost for the design / manufacturing / operation / inspection cost is calculated. In addition, the design / manufacturing / operation / inspection cost function is updated and stored in the inspection result database 24 (see FIG. 5 (j)).

  In this way, the potential cost function for each device set at the design stage is updated according to the manufacturing stage, pre-service inspection, operation, and in-service inspection, and according to the updated potential cost function and criteria. Since it is determined whether or not to replace the equipment, the cost related to the plant equipment can be reduced and the safety management of the plant can be reliably performed.

  By the way, the functions described in the risk management device 10 shown in FIG. 1 and the risk management device 30 shown in FIG. 4 may be combined to determine the device replacement in consideration of both the failure probability and the potential cost. In this case, as described with reference to FIG. 2, when determining device replacement according to the device damage probability function, the importance set in advance for each device is used as a correction coefficient, and the device damage probability function Is corrected (correction means). Then, a failure probability corresponding to the number of years of service is obtained from the corrected device failure probability function, and it is determined according to the value of the failure probability whether to prioritize the potential cost or the failure probability.

  For example, if the damage probability obtained from the corrected device failure probability function is 80% or more, the device replacement is determined by giving priority to the damage probability (before correction) (determination means). On the other hand, if the failure probability obtained from the corrected device failure probability function is 50% or less, replacement of the device is determined with priority given to potential costs. Then, when the failure probability obtained from the corrected device failure probability function is more than 50% and less than 80%, the replacement of the device is determined in consideration of both the failure probability (before correction) and the potential cost. .

  In this case, when the damage probability exceeds the damage determination value (breakage criterion) and the potential cost exceeds the cost determination value (cost criterion), it is determined that the device is replaced.

  With this configuration, it is possible to determine the replacement of the device in consideration of both the probability of breakage and the potential cost. As a result, not only can the cost be reduced, but also the safety management of the plant can be reliably performed according to the failure probability. .

  Here, a risk evaluation method in consideration of the social impact on the plant equipment will be described. Referring to FIG. 6, when the damage factor is evaluated for the nuclear power plant, the evaluation is performed by classifying the damage factor into two types, a pre-evaluation and a post-evaluation. The ex-ante evaluation is to evaluate the impact of an accident / trouble that has not actually occurred, and makes assumptions about the details of the accident / trouble and the effects (scenario) of the accident / trouble. On the other hand, in the ex-post evaluation, accidents / troubles have actually occurred, and it is necessary to organize and specifically define the effects that have actually occurred.

  Therefore, in the prior evaluation, it is necessary to analyze past similar cases, set assumptions based on the analysis result, and extract knowledge of experts in the field and reflect the assumptions.

  In the case analysis setting 50a, in performing past similar case analysis, many similar cases are read from the similar case database 50d, and these similar cases are analyzed to obtain an analysis result. Then, assumptions are set based on these analysis results. Therefore, not only accidents / troubles related to the field of the plant equipment but also accidents / troubles having similar characteristics regarding social impact are analyzed. For example, aircraft accidents and accidents related to the gas business can be cited as those having characteristics similar to accidents in the nuclear field.

  When setting assumptions, past cases will be used as described above, but if no cases exist or records are not recorded, as described above, experts in the field and Knowledge is extracted using the damage evaluation result, and assumptions are set. Then, the loss evaluation 50b evaluates at least the economic loss and the social loss according to the assumption setting to obtain the evaluation result, and the risk estimation 50c estimates the risk cost using the evaluation result as a loss element.

  Economic losses include human damage, property damage, damage to company facilities, operating loss, business compensation, inspection compensation, and increased power generation costs. The number of victims is set by multiplying the number of houses or the affected area existing in the affected area by the population density, and the number of victims is multiplied by the average reparation to obtain human damage. As the average compensation, for example, the average amount paid for automobile insurance in a car accident described in the White Paper on Road Safety is used.

  Property damage is damage caused by fires and explosions caused by accidents and troubles, and is calculated by calculating the average amount of damage to the number of damaged buildings. The number of damaged buildings is set based on the number of buildings existing in the area affected by the fire or explosion, and the average amount of damage is referred to, for example, the amount of damage by building structure published in a white paper on firefighting.

  As the damage to the company equipment, for example, the repair / replacement cost of the company equipment damaged by the accident or trouble is calculated. That is, the repair / replacement cost of the in-house equipment = Σ (repair / replacement required equipment × repair / replacement basic unit price) is obtained. For the equipment requiring repair / replacement, a list is made from the set accidents / troubles, and the repair / replacement cost of each equipment is separately estimated, and a product obtained by multiplying these is calculated.

  The operating loss is calculated based on the contract terms and the effect on the business revenue of the suspension of supply for a certain range and for a certain period of time or more. For example, in the telecommunications business, there is a contract condition such as a refund of a basic fee corresponding to a period of time in which a call cannot be made for a certain period of time, and an operation loss is calculated based on such a case. That is, the operation loss is obtained as the number of supply interruptions × the supply interruption time rate × the basic monthly charge. As for the number of supply stops, the supply possible range is assumed according to the accident / trouble assumption, and the number in the supply stop range is calculated. As for the supply stop time rate, a time during which supply is disabled is assumed based on the assumption of an accident or trouble.

  For business compensation, business damage compensation = (profit amount during the same period of the accident-affected period)-(profit amount after accident / trouble)-business compensation must be provided for damages on the part of the person making the compensation. It is the sum of the claimants who must be paid, and the amount of business damages = {(profit amount during the same period of the accident-affected period)-(profit amount after accident / trouble)}. Regarding inspection compensation, in the case of a nuclear power plant, there are a health checkup fee for humans and a radiological inspection fee for the region, and the health checkup fee for humans = the number of health checkup patients x the unit price of health checkups.

  The number of medical check-ups is set based on the assumption of accidents and troubles in areas where radioactivity is likely to be affected and the number of residents in those areas. The unit price for health checkup shall be the one set by the Ministry of Labor. On the other hand, the radioactivity inspection cost for the area = the area (area) required for radioactivity inspection × (per area) the inspection unit price. The area required for radiological inspection is assumed based on the assumption of accidents and troubles, and the unit price for inspection is set based on the response from the inspection company.

  The increase in power generation costs is calculated based on the increase in power generation costs = increase in alternative power generation costs by thermal power plants + purchase costs from others. It is calculated by the amount of increase in alternative power generation cost by thermal power plant = alternative power generation amount x difference in power generation cost between nuclear power and thermal power. For example, the difference in power generation cost between nuclear power and thermal power is 3.50 yen per 1 kWh. . Further, the purchase cost from another company = the purchase power generation amount × the purchase unit price.

  Social loss includes traffic impairment, evacuation, loss of third-party business, etc.For traffic impairment, the amount of traffic impairment is calculated by the following formula: traffic impairment guarantee = troubled section x average section fare x troubled time x number of users per hour. For example, for the troubled section, the section is set according to the range of influence assumed based on the assumption of the accident / trouble, and the average fare is determined by the means of transportation. Further, the trouble time is set based on the influence time assumed from the accident / trouble assumption. The hourly user is set using actual data published by the transportation agency based on the assumption of the time of occurrence of the accident / trouble.

  In the case of evacuation, the economic loss is required, but the economic loss due to evacuation is the amount of money lost as a result of workers being unable to engage in production activities due to evacuation. Economic loss = Economic loss unit price (yen / person / hour) x number of evacuees (persons) x evacuation time (hours). Here, the unit cost of economic loss is a labor cost that cannot be obtained because one worker cannot work for one hour due to evacuation. For example, in the Japan Statistical Yearbook, “Company size, industry, It is calculated by dividing the average monthly labor cost of regular workers by breakdown of labor costs by the average total actual hours worked per month by regular workers by industry. The number of evacuees is affected by the assumption of accidents and troubles, and the number of workers working there is estimated, assuming a range where evacuation is required. Evacuation time is assumed to be the time when evacuation is required due to the effects of accidents and troubles.

  Third-party business loss is compensation for business loss to third parties that rely on power supply for factories and other businesses, and is calculated by the following formula: business loss = the number of units that are out of supply x the ratio of compensation targets x supply outage time x average business profit. . The average business profit is estimated based on, for example, commercial statistics and the like, and the number of supply interruptions and the supply interruption time are calculated assuming the range of influence and time according to the assumption of accidents and troubles.

  In addition, as a loss of corporate value, public relations measures (costs for apology interviews, apology through the media, PR for safety measures, etc.), administrative measures (responses to local governments and regulatory authorities, and costs required for special inspections and collections) , Loss of demand (loss due to change from electricity to gas, etc. triggered by an accident), hindrance to business expansion (decrease in sales due to postponement of new business plan due to accident), and loss of brand value (decrease in corporate value, brand value) Sales). Then, as described above, the calculated loss factors are added together to calculate the risk cost in consideration of the social impact.

  Here, a method will be described in which a basic failure probability curve is corrected in consideration of characteristics of each plant facility such as operating conditions, and a corrected failure probability curve unique to the plant facility is obtained. Referring to FIG. 7, first, in step P1, an SCC occurrence time Tscc basic expression is selected. Here, the SCC susceptibility of the alloy is expressed by Equation 1 as depending on the microstructure, temperature, and stress of the material.

(Equation 1)
SCC sensitivity = A · σ · n · exp (-Q / (RT))
Here, A: correction value by material microstructure, σ: stress, T: temperature (K), Q: activation energy 40,000 (cal / mol), R: gas constant (1,986 (cal / mol · k)), n: Exponential coefficient of stress (for example, n = 4).
Then, assuming that this sensitivity is inversely proportional to the SCC occurrence time Tscc, Tscc = 1 / SCC sensitivity is set.

  Next, in Step P2, the correction value A is identified. Here, the correction value A is identified from Equation 1 using the T, σ, and Tscc data of the constant load SCC test result for the alloy (the average value Am and the standard deviation As of A are calculated).

  Further, in Step P3, the distribution of stress and temperature is calculated. Here, the average and the standard deviation are designated for each of the temperature and the stress for each plant facility, and the variation is calculated by the normal distribution. In step P4, the distribution of Tscc is calculated. Here, the average value Am, the standard deviation As, and the variation distribution of temperature and stress are given to Tscc = 1 / SCC sensitivity, and the distribution of Tscc is calculated by the Monte Carlo method.

  Next, in Step P5, an SCC occurrence probability curve is calculated from the Tscc distribution, and this is set as a corrected failure probability curve. Here, the Tscc distribution is fitted (adapted) to a probability distribution such as a Weibull distribution to obtain an SCC occurrence probability curve.

  Next, a method for quantitatively evaluating the value of the inspection and evaluating the repair plan will be described. Here, the repair plan is evaluated by quantitatively evaluating the value of the inspection based on the option theory. Referring to FIG. 8, first, a repair plan is performed using damage probability curve 50 at the present time (optimization measure 51 is performed (see Example 1 or 2)). Further, using the failure probability curve 50, expected inspection results 52-1 to 52-N (N is an integer of 2 or more) when an inspection is performed using the inspection method 52 are calculated.

  Here, the expected test results 1 to N are expected under different input conditions. Similarly, although not shown, a plurality of expected inspection results are calculated using an inspection method 53 different from the inspection method 52. Further, a plurality of expected test results may be obtained in a similar manner using another test method. In any case, a plurality of expected test results are obtained using a plurality of test methods.

  Then, for each of the expected inspection results 52-1 to 52-N, a failure probability curve 54-1 to 54-N after the inspection is updated is obtained, and an optimization measure (using these failure probability curves 54-1 to 54-N) ( Repair plan) 55-1 to 55-N is performed. It should be noted that, with respect to the expected inspection result obtained by the inspection method 53 or the like, an optimization measure is similarly performed by obtaining a failure probability curve after the inspection update.

  With respect to the optimization measures 55-1 to 55-N obtained as described above, the risk for the expected test result 52-1 (that is, when the test for obtaining the expected test result 52-1 is performed) is defined as the risk probability (risk). In the same manner, the risk establishment for the expected test results 52-2 to 52-N is evaluated as risk evaluation amounts 56-2 to 56-N, respectively. 1 to 56-N are added to obtain a risk evaluation amount (risk amount with inspection) 57 when the inspection is performed. Then, the risk evaluation value in the optimization measure 51 is set as the risk evaluation value 58 without the inspection, the risk amount 57 with the inspection is compared with the risk evaluation value 58 without the inspection, and the difference is used as the value 59 of the inspection. To be evaluated.

  Further, a method of performing up-to-date by inspection will be described. Here, up-to-date means that when reviewing the failure probability curve based on the inspection result, the failure probability curve is updated to the latest one in consideration of the number of inspections, the inspection interval, the inspection method, and the like. . When performing Up-to-date, so-called Bayes estimation is used.

Now, using the value of the parameter A, the state Hi of the target object is defined as H1: parameter A = average−σ, H2: parameter A = average, and H3: parameter A = average + σ. Also, the cumulative probability of crack occurrence in the state Hi is denoted by F _Hi (t). For the prior probability P (Hi), to calculate the posterior probability _P X (Hi) based on the fact that crack wanted to be confirmed as test results at a certain time t X is, up-to-date by the inspection It is.

Then, since the fact X is a definition that "cracking was not confirmed at a certain time t", the Bayesian estimation equation (Equation 2) was used as P _Hi (X) = 1-F _Hi (t). Up-to-date.

(Equation 2)
P _X (Hi) = (P _Hi (X) · P (Hi)) / P (X)
P _X (Hi) = (P _Hi (X) · P (Hi)) / (ΣP (Hi) · P _Hi (X))
Here, P _X (Hi): probability of occurrence of event Hi when event X is satisfied, P _Hi (X): probability of occurrence of event X when event Hi is satisfied, P (X): Probability of occurrence of event X, P (Hi): Probability of occurrence of event Hi.

Referring to FIG. 9, if the inspection result X is obtained (step R1), first, the type of the damage probability is set to Hi (i is the total number N of types), and the damage probability of Hi is set to f (Hi, t) ( t is time), and the probability of being Hi is Pi (step R2). Then, the current failure probability p (t) is obtained by an average of the respective types of failure probability p (t) = 1 / (N · Σ _i Pi · f (Hi, t)) (step R3). Further, based on the inspection result X, the probability p (X | Hi) that becomes the inspection result X when each of the failure probability types Hi is assumed is obtained for all i (step R4).

Then, using the Bayes' theorem, the probability p (Hi | X) of each failure probability type Hi when the inspection result is X is determined (step R5). Then, p (Hi | X) is newly set as the probability Pi of the damage probability type (state) Hi by updating based on Bayes theory (step R6). Thereafter, the current failure probability p (t) is determined by the average of the newly updated failure probabilities p (t) = 1 / (N · Σ _i Pi · f (Hi, t)). (Step R7). The results of performing the above-described up-to-date are shown in FIGS. FIG. 10 shows an example of the up-to-date in the inspection, and FIG. 11 shows the relationship between the cumulative probability and the time according to the evaluation result shown in FIG.

  Further, here, a method of selecting a maintenance strategy based on the inspection result will be described. Here, the allowable service life is evaluated according to the inspection result, and a maintenance method is selected and an appropriate maintenance selection is performed according to the result.

  Referring to FIG. 12, first, the source (original) failure probability curve is read out from database 60 relating to the target part (step U1), and the inspection period is fixed (selected) (step U2). The inspection method is fixed (selected) by referring to the data (step U3). Thereafter, using the selected inspection technique, the input conditions are made different, and an inspection result assumed (estimated) is obtained for each input condition (step U4).

  For example, the inspection results 62-1 to 63-M (M is an integer of 2 or more) are obtained as the inspection results, and the inspection result 62-1 is the life consumption rate of 10%, and the inspection result 62-2 is the life consumption rate of 20%. Assume that the inspection result 62-M has a life consumption rate of 90%. Then, for the inspection result 62-1, the probability of becoming the inspection result 62-1 is obtained, and the original failure probability curve is up-to-date with the inspection result 62-1 to obtain the first failure probability curve (step U5). ).

  Further, for the inspection result 62-2, the probability of becoming the inspection result 62-2 is obtained, and the original failure probability curve is up-to-date with the inspection result 62-2 to obtain the second failure probability curve (step U6). In the same manner, for the inspection result 62-M, the probability of becoming the inspection result 62-M is obtained, and the original failure probability curve is up-to-date with the inspection result 62-M, and the M-th failure probability curve is obtained. (Step U7).

  Then, by using the method described in the third embodiment, an optimal countermeasure according to the first failure probability curve is obtained, and an expected cost is calculated (step U8). Similarly, in the same manner, an optimal measure corresponding to the second failure probability curve is calculated, an expected cost is calculated (step U9), and an optimal measure corresponding to the M-th failure probability curve is determined, and the expected cost is calculated. Calculation is performed (step U10).

  Then, after summing up these estimated cost amounts (total calculation of expected costs: step U11), it is determined whether or not the inspection method is optimal (step U12). If it is determined that the inspection method is optimal, the inspection method is selected. It is determined whether the inspection time is optimal (step U13). When it is determined that the inspection time is optimal, the optimal inspection time, the inspection method, and the estimated cost are output (step U14). If it is determined in step U12 that the inspection method is not optimal, the process returns to step U3, and if it is determined in step U13 that the inspection time is not optimal. Returning to step U2, the process is performed again.

  By the way, the cumulative failure probability curve described in the sixth embodiment is a gentle curve because various possibilities (factors) are mixed. When the inspection is performed, the approximate life due to various possibilities can be estimated, so that the cumulative failure probability curve becomes somewhat sharp, and the manner of cutting depends on the inspection method and the inspection time.

  Now, assuming that the original failure probability curve is the curve shown in FIG. 13, a low-precision inspection is selected as the inspection method, and the original failure probability curve is up-to-date as described above. A failure probability curve as shown in (a) and (b) is obtained. As shown in the figure, when the inspection method is a low-precision inspection, the failure probability curve becomes gentle, and the possibility of damage increases, and it is necessary to set the inspection interval closely. On the other hand, if the inspection method is a high-precision inspection, as shown in FIGS. 15A to 15C, a failure probability curve after up-to-date is steep, and the time of failure can be narrowed down. By optimizing the repair time, the inspection interval can be extended.

  Here, a method for quantitatively evaluating the effect of the technology development will be described. Here, the effect on the probability of breakage due to technological development will be evaluated, and the investment effect will be evaluated. Referring to FIG. 16, it is assumed that a high-precision inspection technology for inspecting minute scratches with high accuracy has been developed as an inspection method (for example, currently, it is possible to inspect a 1-mm scratch for an inspection accuracy of 3 mm. However, this high-precision inspection technology can perform the inspection easily in a short time and with a small amount of exposure.

  In addition, as a repair method, a repair method has been developed in which the life after repair is extended, repair can be performed easily in a short time, and the amount of exposure is reduced, and processing is easy and operation (operating) conditions are reduced. Suppose a long-life material was developed.

  First, the current probability of damage distribution is calculated (Step W1), and the damage probability distribution is corrected using the technique described in the fourth embodiment by introducing the above-described technology development (Step W2). For example, by improving the accuracy of the inspection method, the probability of detection of damage is improved, so that the distribution of the time of occurrence of flaws is reviewed and the accuracy of estimating the progress of flaws is improved, so that maintenance can be performed at an appropriate time. With the technical progress of the repair method, the probability distribution after the repair is corrected, and the inspection and replacement time is lengthened. As a result, maintenance costs can be reduced by reviewing maintenance.

  Further, by simplifying the inspection and / or repair, the frequency of up-to-date due to the improvement of the frequency of inspection and repair is improved, and the maintenance plan can be more finely reviewed. And, by extending the life of the material, the review of the probability distribution of the occurrence / development of breakage, the inspection and replacement cycle become longer, and the maintenance cost can be reduced by reviewing the maintenance.

  Subsequently, calculation of the risk reduction by correcting the failure probability distribution is performed (step W3). Here, the optimal maintenance plan from the current time to the update of the target plant is determined for each of the conventional technology and the high-precision technology (new technology), and the plant life is calculated for each of the conventional technology and the new technology. Calculate the cumulative risk over the remaining period of Then, the difference between the cumulative risk of the conventional technical method and the new technical method (the sum of the target plants when there are a plurality of target plants) is determined as the amount of risk reduction. That is, the amount of risk reduction ΔR (t) = the cumulative risk in the conventional technique—the cumulative risk in the new technique is calculated as the risk reduction.

  Then, the investment amount for the development period is calculated (step W4). Here, the amount of cumulative risk reduction is calculated by the method of step W3 according to the introduction time of the new technology method and the application period after the introduction in the entire life cycle of the target plant, and the investment amount for the development period is calculated. (If the development is early, the investment will increase, and if the development is prolonged, the investment will decrease).

  At the design stage for each component of the plant equipment, a failure probability function indicating the relationship between the service life of the component and the probability of failure is determined for each component according to the design data related to the component, and the failure probability function is manufactured for the production of the component. Corrected damage probability function is corrected based on at least one of the data, the operation history of the device, and the inspection related to the device, and a corrected damage probability obtained from the corrected damage probability function according to the service life and a predetermined judgment. If the corrected failure probability exceeds the judgment value, the replacement of the equipment is determined and the risk of the plant equipment is managed in accordance with the corrected damage probability. It can be applied to risk management in various plants such as plants.

It is a block diagram showing an example of a risk management device by the present invention. 3 is a flowchart for explaining the operation of the risk management device shown in FIG. FIG. 2 is a diagram showing a failure probability function generated by the risk management device shown in FIG. 1, (a) is a diagram showing a design-time damage probability function, (b) is a diagram showing a failure probability function corrected in a manufacturing stage, (C) is a diagram showing the failure probability function corrected by the presence or absence of the pre-service inspection, (d) is a diagram showing the failure probability function corrected by the operation history (monitoring), and (e) is a graph showing the presence or absence of the in-service inspection. FIG. 7F is a diagram illustrating a modified failure probability function, FIG. 7F is a diagram illustrating resetting of the failure probability function due to device replacement time, and FIG. 7G is a diagram illustrating a change in the failure probability function due to device replacement. It is a block diagram showing other examples of a risk management device by the present invention. 5A and 5B are diagrams illustrating a cost function generated by the risk management device illustrated in FIG. 4, wherein FIG. 5A illustrates a design-time cost function, FIG. 5B illustrates a cost function corrected in a manufacturing stage, and FIG. Is a diagram showing a cost function corrected by the presence / absence of a pre-service inspection, (d) is a diagram showing an operation cost function by design / manufacturing / pre-service inspection, and (e) is a cost function corrected by an operation history (monitoring). , (F) is a diagram showing the operation cost function corrected by the operation history (monitoring), (g) is a diagram showing the cost function corrected by the presence or absence of the in-service inspection, and (h) is the in-service inspection FIG. 7A is a diagram showing an operation cost function modified according to the presence or absence of (a), (i) is a diagram showing a change in the cost function due to device replacement, and (j) is a diagram showing a change in the operation cost function due to device replacement. It is a block diagram for explaining the risk evaluation method which considered the social influence in plant equipment in the risk management apparatus by this invention. 5 is a flowchart for explaining a method of obtaining a plant equipment-specific corrected failure probability curve according to characteristics of each plant equipment such as operating conditions in the risk management device according to the present invention. 6 is a flowchart for explaining a method of evaluating a repair plan by quantitatively evaluating the value of plant equipment inspection in the risk management device according to the present invention. 5 is a flowchart illustrating a method of updating a failure probability function by plant equipment inspection in the risk management device according to the present invention. It is a figure showing an example of an evaluation result in plant equipment inspection. FIG. 11 is a diagram illustrating a relationship between a cumulative probability and time according to the evaluation result illustrated in FIG. 10. 9 is a flowchart illustrating an example in which the risk management apparatus according to the present invention evaluates an allowable useful life in accordance with a plant equipment inspection result and selects a maintenance method in accordance with the result. It is a figure showing an example of an original failure probability curve. (A) And (b) is a figure which shows the failure probability curve obtained as a result of updating the original failure probability curve according to the low precision inspection. (A)-(c) is a figure which shows the failure probability curve obtained as a result of updating the original failure probability curve according to the high precision inspection. 5 is a flowchart for explaining a method for quantitatively evaluating the effect of technology development in the risk management device according to the present invention.

Explanation of reference numerals

10, 30 Risk management device 11, 31 Risk calculation processing device 12 Display device 13 Input device 14 Output device 21, 41 Design information database 22, 42 Manufacturing information database 23 Inspection method accuracy database 24, 44 Inspection result database 43 Inspection method / accuracy・ Inspection cost database 45 Accident cost / design cost database 46 Manufacturing cost database 47 Operation / monitoring cost database 48 Replacement cost database

Claims (18)

A risk management device for managing plant equipment risks,
Damage probability function calculation means for determining a damage probability function indicating the relationship between the service life and the damage probability for each of the devices in the design stage for each device constituting the plant equipment according to the design data for the device,
Damage probability function changing means for correcting the damage probability function based on at least one of the manufacturing data related to the manufacture of the device, the operation history of the device, and the inspection of the device, so as to obtain a corrected damage probability function,
A determining means for comparing the corrected damage probability obtained from the corrected damage probability function according to the service life with a predetermined determination value, and determining the replacement of the device when the corrected damage probability exceeds the determination value. Have
A risk management device, wherein the risk of the plant equipment is managed according to the correction failure probability.   When the replacement of the device is performed according to the replacement determination by the determination unit, the modified damage probability function corresponding to the replaced device is reset once so that the service life of the replaced device becomes the starting point. 2. The risk management apparatus according to claim 1, further comprising a reset changing unit that moves the corrected failure probability function.   The risk management apparatus according to claim 1, wherein the failure probability function changing unit corrects the failure probability function according to inspection accuracy in the inspection. A risk management device for managing plant equipment risks,
Cost function calculation means for obtaining a cost function indicating the relationship between the cost and the service life when an accident occurs in the equipment at the design stage for each equipment constituting the plant equipment,
A cost function changing unit that corrects the cost function based on at least one of a manufacturing cost related to the manufacture of the device, an inspection related to the device, and an operation history of the device, and the corrected cost function,
Determining means for comparing the correction cost obtained from the correction cost function according to the service life and a predetermined determination value to determine replacement of the device when the correction cost exceeds the determination value,
A risk management device, wherein the risk of the plant equipment is managed according to the correction cost.   When the replacement of the device is performed according to the replacement determination by the determination unit, the modified cost function corresponding to the replaced device is reset once so that the service life of the replaced device becomes the starting point. The risk management apparatus according to claim 4, further comprising a reset changing unit that moves the corrected cost function.   The risk management device according to claim 4, wherein the cost function changing unit is configured to correct the cost function according to inspection accuracy in the inspection. The cost function calculating means calculates the relationship between the service life of the device and the design cost, the manufacturing cost, and the inspection cost according to the design cost required in the design stage, the manufacturing cost, and the inspection cost required for the inspection. The operation cost function
6. The risk management apparatus according to claim 5, wherein the cost function changing unit corrects the operation cost with an operation / monitoring cost required for operating and monitoring the device.   The reset change unit is characterized in that, when the device is replaced according to the replacement determination by the determination unit, the operation cost function is modified according to a replacement cost required for replacing the device. The risk management device according to claim 7. A risk management device for managing plant equipment risks,
When a failure probability function indicating the relationship between the service life and the probability of failure is determined for each device in the design stage for each device constituting the plant equipment according to the design data related to the device, and when an accident occurs for the device Function calculating means for obtaining a cost function indicating the relationship between the cost of the device and the number of years of service,
The damage probability function is corrected data based on at least one of the manufacturing data related to the manufacture of the device, the operation history of the device, and at least one of the inspections on the device. Function changing means for correcting the cost function based on at least one of the manufacturing cost of manufacturing the device, the inspection related to the device, and the operation history of the device to make the corrected cost function,
Correction means for correcting the correction damage probability obtained from the correction damage probability function according to the service life with the correction coefficient as the correction damage probability as a correction coefficient indicating the importance indicating the importance of each device,
When the value obtained from the selection function according to the service life is greater than a predetermined determination value, the device is selectively used according to the corrected cost probability and the damage probability function as a selection function according to the corrected failure probability. Determining means for determining the replacement of
A risk management device, wherein a risk of the plant equipment is managed in accordance with the modified failure probability function and the modified cost function. A risk management device for evaluating and managing secondary damage as a secondary risk in consideration of the social impact of plant equipment,
Assumption setting means for performing a past similar case analysis relating to the plant equipment, and setting an assumption based on the analysis result,
Evaluation means for evaluating at least economic loss and social loss according to the assumption setting to obtain an evaluation result,
A risk estimating means for estimating a risk cost using the evaluation result as a loss factor.   Assess at least human damage, property damage, damage to own facilities, operation loss, business compensation, inspection compensation, and increase in power generation costs as the economic loss, and at least traffic obstacles, evacuation, and third A risk management device characterized by evaluating a business loss of a business operator. A risk management device that performs risk management of the plant equipment according to characteristics of the plant equipment when managing risk of the plant equipment,
Correction failure probability by correcting the original failure probability function according to the characteristics of the plant equipment as a failure probability function indicating the relationship between the service life and the failure probability for each device constituting the plant equipment as an original failure probability function Correction means for finding a function;
A risk management unit that performs risk management of the plant equipment based on the corrected failure probability function. A first means for obtaining a damage occurrence time equation represented by a reciprocal of the damage sensitivity of the equipment material;
Second means for identifying a correction value according to the damage occurrence time equation;
Third means for determining a temperature and stress distribution for the device according to a normal distribution;
Fourth means for obtaining a damage occurrence time distribution by a Monte Carlo method according to the damage occurrence time formula, the temperature and stress distribution, and the correction value;
13. The risk management apparatus according to claim 12, further comprising: fifth means for adapting the failure time distribution to a predetermined probability distribution to obtain the corrected failure occurrence probability curve. A risk management device that performs risk management of the plant equipment by quantitatively evaluating the value related to the inspection of the plant equipment when managing the risk of the plant equipment,
Optimizing a repair plan using the current failure probability function using the current failure probability function as a current failure probability function indicating the relationship between the service life and the failure probability for each component of the plant equipment. Optimization means,
Using the current failure probability function, when inspection is performed using a plurality of inspection methods different from each other, an inspection result prediction means that calculates a plurality of expected inspection results for each inspection method,
A failure probability function updating means for determining a failure probability curve after inspection update as an updated failure probability function for each of the expected inspection results,
A second optimization means for optimizing a repair plan for each update failure probability function using the update failure probability function to obtain a plurality of optimization measures;
A risk evaluation amount for the expected inspection result is evaluated for the optimization measure, and a risk evaluation amount at the time of performing the inspection is compared with a risk amount of the optimization measure obtained by the first optimization means, and the inspection is performed. A risk management device comprising: an evaluation unit that evaluates a value. When managing the risk of plant equipment, a failure probability function indicating the relationship between the service life and the failure probability for each device constituting the plant equipment according to the number of inspections, inspection intervals, and inspection methods for each plant equipment A risk management device that updates the risk of the plant equipment by updating
First means for obtaining, as a first probability, a probability of being the inspection result based on the inspection result according to the type of the failure probability for each device constituting the plant facility;
A second means for obtaining the probability as a second probability for each failure probability type for the inspection result using Bayes' theorem,
Third means for setting the second probability to a third failure probability of the failure probability type by updating based on Bayes theory;
And a fourth means for obtaining the failure probability function as an average of the third failure probabilities. A risk management device that performs a risk management of the plant equipment according to the selected maintenance method by selecting a maintenance method according to the inspection result for each plant equipment when managing the risk of the plant equipment,
First means for selecting an inspection period for each plant equipment;
Second means for selecting the inspection method for each plant equipment,
A third means for obtaining an expected inspection result when the inspection method selected by the second means is used;
Fourth means for updating the original failure probability curve indicating the relationship between the service life and the failure probability for each device constituting the plant equipment based on the predicted inspection result to obtain an updated failure probability curve;
A fifth means for calculating an estimated cost amount while obtaining an optimum measure in accordance with the update failure probability curve,
A risk management device, wherein a maintenance method is selected according to the optimization measure and expected cost. A risk management device that evaluates the maintenance effect of the plant equipment by technology development when managing the risk of the plant equipment and performs risk management of the plant equipment,
First means for obtaining a failure probability distribution indicating a relationship between the service life and the damage probability for each device constituting the plant equipment;
Second means for correcting the failure probability distribution by introducing the technology development to obtain a corrected failure probability function;
A third means for calculating a risk reduction when the corrected failure probability distribution is used.   18. The risk according to claim 17, further comprising: a fourth unit for calculating an investment amount for the technology development based on a cumulative risk reduction according to a timing of inputting the technology development and an application period after the technology development. Management device.

Images