Prediction method and prediction device

Abstract

This prediction method includes: calculating the likelihood that a crack will occur in inspected equipment, which is equipment that has already been subjected to a crack inspection, by using the crack results of a plurality of first heat-affected parts of the inspected equipment (step S120); predicting the hardness of the plurality of first heat-affected parts for each candidate, by using each of a plurality of candidates comprising a pair consisting of a maximum temperature and the maximum temperature-holding time at the time of welding, and calculating a predicted occurrence likelihood, which is a prediction value for the likelihood that a crack will occur in inspected equipment by using the hardness prediction results (step S130); selecting an optimal pair for maximum temperature and holding time, by using the occurrence likelihood and the predicted occurrence likelihood for each candidate (step S140); and predicting the hardness of a second heat-affected part in prediction target equipment, which is equipment targeted for crack occurrence likelihood prediction, by using the selected optimal pair (step S150).

Description

Prediction method and prediction device

[0001] This application claims the benefit of priority from Japanese Patent Application No. 2024-082004, filed May 20, 2024, the contents of which are incorporated herein by reference.

[0002] Tanks for storing fuel and the like are made of metal materials such as steel, aluminum, and nickel. Stress corrosion cracking can occur in such metal materials due to deterioration over time. If stress corrosion cracking occurs in an existing tank, fuel and the like will leak from the tank. For this reason, there is a demand for predicting the occurrence of stress corrosion cracking in advance.

[0003] Therefore, as a technique for predicting sulfide stress cracking in a corrosive environment, creating a prediction model by machine learning has been proposed (for example, Patent Document 1). In the technique of Patent Document 1, a sulfide stress cracking test is performed using a plurality of test materials, and then machine learning is performed on learning data in which the strength grade of the test materials and test conditions such as hydrogen sulfide partial pressure are used as input data and the results of the sulfide stress cracking test are used as output data.

[0004] JP 2023-174500 A

[0005] In recent years, there has been a demand for predicting cracks in tank weld joints, such as stress corrosion cracking and hydrogen embrittlement cracking. Because the occurrence of cracks is affected by the hardness of the metal material, if the hardness of the weld joint can be predicted, the occurrence of cracks can also be predicted.

[0006] However, the hardness of welded joints is difficult to predict because the microstructure and properties of the base material are changed by welding.

[0007] In view of the above problems, the present disclosure aims to provide a prediction method and a prediction device that can easily predict the hardness of the heat-affected zone caused by welding.

[0008] In order to solve the above problem, a prediction method according to one embodiment of the present disclosure includes: calculating the probability of crack occurrence in inspected equipment using the results of cracks in multiple first heat-affected zones of inspected equipment, which is equipment for which crack inspection has already been performed; predicting the hardness of multiple first heat-affected zones for each candidate using multiple candidates for pairs of maximum temperature and maximum temperature holding time during welding, and calculating a predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment, using the occurrence probability and the predicted occurrence probability for each candidate; determining an optimal pair of maximum temperature and holding time using the occurrence probability and the predicted occurrence probability for each candidate; and predicting the hardness of a second heat-affected zone in prediction target equipment, which is equipment for which the probability of crack occurrence is to be predicted, using the determined optimal pair.

[0009] In addition, calculating the probability of cracks occurring in the inspected equipment may include calculating the probability of cracks occurring in the inspected equipment using the results of cracks obtained from an open inspection performed on the inspected equipment.

[0010] Furthermore, determining the optimum pair may include determining a different optimum pair for each parameter that affects whether or not a crack occurs.

[0011] Furthermore, the facility and the prediction target facility may be tanks that store the same type of liquefied gas.

[0012] In order to solve the above problem, a prediction device according to one aspect of the present disclosure includes one or more processors and one or more memories connected to the processors, and the processor performs processing including: calculating the probability of crack occurrence in inspected equipment using the results of cracks in multiple first heat-affected zones of inspected equipment, which is equipment for which crack inspection has already been performed; predicting the hardness of multiple first heat-affected zones for each candidate using multiple candidates for pairs of maximum temperature and maximum temperature holding time during welding, and calculating a predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment, using the occurrence probability and the predicted occurrence probability for each candidate; determining an optimal pair of maximum temperature and holding time using the occurrence probability and the predicted occurrence probability for each candidate; and predicting the hardness of the second heat-affected zone in prediction target equipment, which is equipment for which the probability of crack occurrence is to be predicted, using the determined optimal pair.

[0013] According to the present disclosure, it is possible to easily predict the hardness of the heat-affected zone caused by welding.

[0014] Fig. 1 is a diagram illustrating a prediction device according to an embodiment of the present disclosure. Fig. 2 is a diagram illustrating an example of inspected equipment and a first heat-affected zone according to the embodiment. Fig. 3 is a block diagram illustrating an example of a functional configuration of a control device according to the embodiment. Fig. 4 is a flowchart illustrating a processing flow of a prediction method according to the embodiment. Fig. 5 is a diagram illustrating an example of a CCT diagram according to the embodiment.

[0015] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Dimensions, materials, and other specific numerical values ​​shown in the embodiments are merely examples for ease of understanding and do not limit the present disclosure unless otherwise specified. In this specification and drawings, elements having substantially the same functions and configurations are designated by the same reference numerals to avoid redundant explanation. Elements not directly related to the present disclosure are not shown.

[0016] [Prediction device 100] Fig. 1 is a diagram illustrating a prediction device 100 according to an embodiment of the present disclosure. Fig. 2 is a diagram illustrating an example of an inspected facility 10 and a first heat-affected zone 12 according to the embodiment.

[0017] As shown in FIG. 1 , a prediction device 100 according to this embodiment includes one or more processors 102 and one or more memories 104 connected to the processors 102. The processor 102 includes, for example, a CPU (Central Processing Unit). The memory 104 includes, for example, a ROM (Read Only Memory) and a RAM (Random Access Memory). The ROM is a storage element that stores programs and calculation parameters used by the CPU. The RAM is a storage element that temporarily stores data such as variables and parameters used in processing executed by the CPU.

[0018] The prediction device 100 according to this embodiment predicts the hardness of the second heat-affected zone in the prediction target equipment using the results of cracking of multiple first heat-affected zones 12 in the inspected equipment 10 shown in FIG. 2 . The inspected equipment 10 is equipment for which a crack inspection has already been conducted. The prediction target equipment is equipment for which the probability of crack occurrence is predicted. The inspected equipment 10 and the prediction target equipment are, for example, tanks that store the same type of liquefied gas. The liquefied gas is, for example, liquefied ammonia or liquefied hydrogen.

[0019] The heat affected zone (HAZ) is a portion where the microstructure and properties of the base material are changed by welding. The first heat affected zone 12 is, for example, a welded joint of the inspected equipment 10, as shown in Fig. 2. The second heat affected zone is, for example, a welded joint of the equipment to be predicted.

[0020] The cracking is either or both of stress corrosion cracking (SCC) and hydrogen embrittlement cracking.

[0021] 3 is a block diagram showing an example of the functional configuration of the prediction device 100 according to this embodiment. For example, as shown in FIG. 3, the prediction device 100 also functions as an acquisition unit 210, a processing unit 212, and a storage unit 214.

[0022] Note that various processes, including the processes described below, performed by one or more selected from the group consisting of the acquisition unit 210, the processing unit 212, and the storage unit 214, may be executed by the processor 102. In detail, the various processes are executed by the processor 102 executing a program stored in the memory 104. The function of the storage unit 214 is realized by the memory 104. However, the functions of the prediction device 100 may be divided among multiple devices, or multiple functions may be realized by a single device.

[0023] The acquisition unit 210 acquires results of cracks in the multiple first heat-affected zones 12 of the inspected equipment 10. The acquisition unit 210 may acquire, for example, crack results obtained by an open inspection performed on the inspected equipment 10. In addition, the acquisition unit 210 acquires, for example, mill sheets of materials constituting the inspected equipment 10, welding records of the first heat-affected zones 12 of the inspected equipment 10, mill sheets of materials constituting the prediction target equipment, and welding records of the second heat-affected zones of the prediction target equipment.

[0024] The processing unit 212 calculates the probability of crack occurrence in the inspected equipment 10 using the results of crack detection in the multiple first heat-affected zones 12 of the inspected equipment 10. The processing unit 212 calculates the probability of crack occurrence in the inspected equipment 10 using, for example, the results of crack detection obtained through an open inspection of the inspected equipment 10. Next, the processing unit 212 predicts the hardness of the multiple first heat-affected zones 12 for each candidate pair of maximum temperature and maximum temperature holding time during welding. The processing unit 212 then calculates a predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment 10, using the hardness prediction results. The processing unit 212 also determines an optimal pair of maximum temperature and holding time using the occurrence probability and the predicted occurrence probability for each candidate. The processing unit 212 then predicts the hardness of the second heat-affected zone in the equipment to be predicted using the determined optimal pair. The processing by the processing unit 212 will be described in detail below in the prediction method section.

[0025] The storage unit 214 stores the information acquired by the acquisition unit 210 .

[0026] [Prediction Method] Next, a prediction method using the prediction device 100 will be described. Fig. 4 is a flowchart showing the processing flow of the prediction method according to this embodiment. As shown in Fig. 4, the prediction method according to this embodiment includes, for example, an acquisition processing step S110, an occurrence probability calculation processing step S120, a predicted occurrence probability calculation processing step S130, a pair determination processing step S140, and a prediction processing step S150. Each step will be described below.

[0027] [Acquisition Processing Step S110] In the acquisition processing step S110, the acquisition unit 210 acquires the results of cracks in the multiple first heat-affected zones 12 of the inspected equipment 10.

[0028] Tanks that store liquefied gas are required by the national government, local governments, etc. to undergo open inspections at regular intervals. Open inspections involve emptying the tank, setting up scaffolding inside the tank, and conducting flaw detection tests such as penetrant testing and magnetic particle testing to inspect the first heat-affected zone 12 for cracks. Therefore, when the inspected equipment 10 is a tank that stores liquefied gas, the acquisition unit 210 acquires the results of the open inspection. Note that the open inspection can acquire whether the first heat-affected zone 12 is cracked, but it cannot measure the hardness of the first heat-affected zone 12 when it is cracked.

[0029] The results of the open inspection include, for example, position information of the first heat-affected zone 12, the results of cracking in the first heat-affected zone 12, and the liquid contact state of the first heat-affected zone 12. The results of cracking include the presence or absence of cracks. The liquid contact state includes, for example, a contact state with either or both of the liquid and the gas stored in the tank, such as constant contact with the liquid phase, constant contact with the gas phase, or a gas-liquid interface.

[0030] The acquisition unit 210 also acquires mill sheets of the materials constituting the inspected equipment 10 and a welding record of the first heat-affected zone 12 of the inspected equipment 10. The mill sheets of the materials constituting the inspected equipment 10 include the composition of the first heat-affected zone 12 and the plate thickness of the first heat-affected zone 12. The welding record of the first heat-affected zone 12 of the inspected equipment 10 includes the welding method and welding conditions for the first heat-affected zone 12. Welding methods include, for example, arc welding, electron beam welding, laser welding, plasma arc welding, etc. Welding conditions include, for example, welding current, welding voltage, welding speed, and initial temperature of the material to be welded. Note that the welding method to be applied depending on the welding position is predetermined, and the welding conditions to be applied depending on the plate thickness are predetermined. Welding positions include, for example, downward, horizontal, upward, downward, and overhead.

[0031] The memory unit 214 stores the results of the open inspection acquired by the acquisition unit 210, the mill sheets of the materials that make up the inspected equipment 10, and the welding record of the first heat-affected zone 12 of the inspected equipment 10 as inspection result information.

[0032] [Occurrence Probability Calculation Processing Step S120] In the occurrence probability calculation processing step S120, the processing unit 212 refers to the inspection result information stored in the storage unit 214, and calculates the occurrence probability of cracks in the inspected equipment 10 using the results of cracks in the multiple first heat-affected zones 12 of the inspected equipment 10. For example, the processing unit 212 calculates the occurrence probability of cracks in the inspected equipment 10 by dividing the number of first heat-affected zones 12 that have cracks by the total number of first heat-affected zones 12 that have been inspected for cracks, as shown in the following formula (1): Probability of crack occurrence in inspected equipment 10 = Number of first heat-affected zones 12 that have cracks / Total number of first heat-affected zones 12 that have been inspected for cracks ... formula (1)

[0033] [Predicted Occurrence Probability Calculation Processing Step S130] In the predicted occurrence probability calculation processing step S130, the processing unit 212 uses each of a plurality of candidates for a pair of a maximum temperature and a holding time at the maximum temperature during welding to predict the hardness of the plurality of first heat-affected zones 12 for each candidate, based on information including, for example, the composition and cooling rate of the first heat-affected zone 12. Then, the processing unit 212 uses the hardness prediction results to calculate a predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment 10. The predicted occurrence probability calculation processing step S130 includes selecting a candidate, predicting the hardness of the first heat-affected zone 12, and calculating the predicted occurrence probability. Each process included in the predicted occurrence probability calculation processing step S130 will be described in detail below.

[0034] (Selection of Candidates) The processing unit 212 selects a plurality of candidates for pairs of maximum temperature and maximum temperature holding time during welding. For example, the processing unit 212 selects the plurality of candidates using a design of experiments. Note that the processing unit 212 may select the plurality of candidates using a method other than the design of experiments.

[0035] (Prediction of Hardness of First Heat-Affected Zone 12) The processing unit 212 uses each of the selected candidates to predict the hardness of the first heat-affected zones 12 for each candidate.

[0036] First, the processing unit 212, for example, references the inspection result information stored in the memory unit 214 and calculates the cooling rate of the first heat-affected zone 12 using the plate thickness and welding record of the first heat-affected zone 12. The cooling rate is calculated, for example, using a known calculation formula (Terasaki Toshio, Akiyama Tetsuya, Ishimoto Kenji, Mori Yasuhiko, "Proposal of an Estimation Formula for Cooling Time t8 / 5," Proceedings of the Japan Welding Society, Vol. 6, No. 2, pp. 301-305, 1988) or by heat conduction analysis simulating welding. The cooling rate is, for example, a cooling rate from 800°C to 500°C.

[0037] The processing unit 212 then acquires the composition of the first heat-affected zone 12, for example, by referring to the inspection result information stored in the memory unit 214. The processing unit 212 creates a Continuous Cooling Transformation (CCT) diagram using the acquired composition of the first heat-affected zone 12, the calculated cooling rate of the first heat-affected zone 12, and the selected candidates (pairs of maximum temperature and maximum temperature holding time during welding) as input data. The processing unit 212 then predicts the hardness of the first heat-affected zone 12 from the CCT diagram.

[0038] Fig. 5 is a diagram showing an example of a CCT diagram according to this embodiment. In Fig. 5, the vertical axis represents temperature [°C], and the horizontal axis represents time. In Fig. 5, F represents ferrite, P represents pearlite, B represents bainite, and M represents martensite. In Fig. 5, Tmax represents the maximum temperature, and Rt represents the holding time at the maximum temperature.

[0039] 5 , the processing unit 212 estimates the microstructure (structure) of the first heat-affected zone 12 based on the CCT diagram. Then, the processing unit 212 predicts the hardness of the first heat-affected zone 12 based on the estimated structure of the first heat-affected zone 12.

[0040] The processing unit 212 creates a CCT diagram for each selected candidate for one first heat-affected zone 12, and predicts the hardness of the first heat-affected zone 12 for each selected candidate from each CCT diagram. Then, the processing unit 212 predicts the hardness for each selected candidate for all first heat-affected zones 12 that have been subjected to open inspection.

[0041] For example, a case will be described in which a first candidate, a second candidate, a third candidate, ..., an Nth candidate are selected as candidates for a pair of maximum temperature and maximum temperature holding time. In this case, the processing unit 212 first predicts the hardness of all first heat-affected zones 12 using the first candidate as described above. Next, the processing unit 212 predicts the hardness of all first heat-affected zones 12 using the second candidate as described above. Similarly, the processing unit 212 predicts the hardness of all first heat-affected zones 12 for the other candidates. In this way, the processing unit 212 predicts the hardness of all first heat-affected zones 12 for each of the first candidate, second candidate, third candidate, ..., an Nth candidate.

[0042] (Calculation of predicted occurrence probability) The processing unit 212 calculates the predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment 10, using the hardness prediction result. For example, the processing unit 212 assumes that the first heat-affected zones 12 whose hardness prediction result is equal to or greater than a predetermined value are the locations where cracks have occurred. Then, as shown in the following formula (2), the processing unit 212 calculates the predicted occurrence probability in the inspected equipment 10 by dividing the number of first heat-affected zones 12 whose hardness prediction result is equal to or greater than a predetermined value by the total number of first heat-affected zones 12 that have been inspected for cracks. Note that the predetermined value is, for example, Vickers hardness 210 (Hv210). Predicted occurrence probability = Number of first heat-affected zones 12 whose hardness prediction result is equal to or greater than a predetermined value / Total number of first heat-affected zones 12 that have been inspected for cracks ... formula (2)

[0043] Furthermore, the processing unit 212 calculates a predicted occurrence probability for each of the selected candidates (pairs of maximum temperature and maximum temperature holding time during welding).

[0044] For example, a case will be described in which the first candidate, second candidate, third candidate, ..., Nth candidate are selected as candidates for a pair of maximum temperature and maximum temperature holding time, as described above. In this case, the processing unit 212 first calculates the predicted occurrence probability as described above using the hardness prediction result when the first candidate is used. Next, the processing unit 212 calculates the predicted occurrence probability as described above using the hardness prediction result when the second candidate is used. Similarly, the processing unit 212 calculates the predicted occurrence probability for the other candidates as described above using the hardness prediction results. In this way, the processing unit 212 calculates the predicted occurrence probability for each of the first candidate, second candidate, third candidate, ..., Nth candidate.

[0045] 4 , in the pair determination process step S140, the processing unit 212 determines an optimal pair of maximum temperature and holding time using the occurrence probability calculated by the above formula (1) and the predicted occurrence probability for each candidate calculated by the above formula (2). The optimal pair is a pair of maximum temperature and holding time at maximum temperature for which the predicted occurrence probability is close to or coincides with the occurrence probability. The optimal pair is a pair of maximum temperature and holding time at maximum temperature for which the structure of the second heat-affected zone, which will be described later, can be calculated with high accuracy.

[0046] The optimal pair is, for example, a pair of the maximum temperature and the holding time of the maximum temperature that minimizes the difference between the occurrence probability and the predicted occurrence probability. In this case, the processing unit 212 uses, for example, Bayesian optimization, with the pair of the maximum temperature and the holding time of the maximum temperature as explanatory variables and the error between the occurrence probability and the predicted occurrence probability as objective variables, and searches for a pair of the maximum temperature and the holding time of the maximum temperature that minimizes the objective variable. The error can be, for example, a root mean squared error (RMSE), a mean absolute error (MAE), a decision coefficient (R ２ The processing unit 212 may determine the optimal pair of the maximum temperature and the holding time using a method other than Bayesian optimization, for example, experimental design.

[0047] [Prediction Processing Step S150] Then, the processing unit 212 predicts the hardness of the second heat-affected zone in the prediction target equipment using the determined optimal pair. For example, the acquisition unit 210 acquires mill sheets of the materials constituting the prediction target equipment and welding records of the second heat-affected zone of the prediction target equipment, and stores them in the memory unit 214. The mill sheets of the materials constituting the prediction target equipment include the composition of the second heat-affected zone and the plate thickness of the second heat-affected zone. The welding records of the second heat-affected zone of the prediction target equipment include the welding method and welding conditions for the second heat-affected zone.

[0048] The processing unit 212 then refers to the mill sheets of the materials constituting the equipment to be predicted and the welding record of the second heat-affected zone of the equipment to be predicted, and calculates the cooling rate of the second heat-affected zone using the plate thickness of the second heat-affected zone and the welding record. The calculation of the cooling rate of the second heat-affected zone is similar to the calculation method of the cooling rate of the first heat-affected zone 12 described above, and therefore a detailed description thereof will be omitted here.

[0049] Next, the processing unit 212 references the mill sheet of the material constituting the equipment to be predicted and obtains the composition of the second heat-affected zone. The processing unit 212 then creates a CCT diagram using the composition of the second heat-affected zone, the cooling rate of the second heat-affected zone, and the optimal pair determined in the pair determination process step S140 as input data. The processing unit 212 also predicts the hardness of the second heat-affected zone from the CCT diagram. Predicting hardness based on the CCT diagram is similar to predicting the hardness of the first heat-affected zone 12, and therefore will not be described in detail here.

[0050] In this embodiment, the processing unit 212 predicts the hardness of all second heat-affected zones in the equipment to be predicted.

[0051] The processing unit 212 then determines that the second heat-affected zone having a predicted hardness equal to or greater than a predetermined value (e.g., Hv210) is a second heat-affected zone having a high probability of cracking. Then, the second heat-affected zone having a high probability of cracking is subjected to crack prevention work. The crack prevention work is, for example, one or more selected from the group consisting of shot peening, laser peening, water jet peening, and zinc spraying.

[0052] As described above, the prediction device 100 of this embodiment and the prediction method using the same include calculating the probability of crack occurrence in the inspected equipment 10 using the results of cracking in multiple first heat-affected zones 12 of the inspected equipment 10, which is equipment for which a crack inspection has already been performed; predicting the hardness of multiple first heat-affected zones 12 for each candidate using each of multiple candidates for a pair of maximum temperature and holding time at the maximum temperature during welding, and calculating a predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment 10, using the occurrence probability and the predicted occurrence probability for each candidate; determining an optimal pair of maximum temperature and holding time using the occurrence probability and the predicted occurrence probability for each candidate; and predicting the hardness of the second heat-affected zone in the prediction target equipment, which is equipment for which the probability of crack occurrence is to be predicted, using the determined optimal pair.

[0053] Specifically, the prediction device 100 and a prediction method using the same according to this embodiment estimate the structure of the first heat-affected zone 12 and predict the hardness of the first heat-affected zone 12 based on the structure of the first heat-affected zone 12. The structure of the heat-affected zone varies depending on the composition of the heat-affected zone and the maximum temperature and the holding time at the maximum temperature during welding. Therefore, the prediction device 100 and a prediction method using the same according to this embodiment predicts the hardness of the first heat-affected zone 12 for each of multiple candidates for the maximum temperature and the holding time at the maximum temperature during welding, which are important parameters for estimating the structure of the first heat-affected zone 12. Next, the prediction device 100 and a prediction method using the same according to this embodiment calculates a predicted occurrence probability using these. Then, the prediction device 100 and a prediction method using the same according to this embodiment determine an optimal pair of the maximum temperature and the holding time at the maximum temperature during welding, which is an important parameter for estimating the structure of the heat-affected zone, based on the occurrence probability of cracks in the inspected equipment 10 that was actually inspected and the predicted occurrence probability. Therefore, the prediction device 100 according to the present embodiment and the prediction method using the same can determine a suitable optimal pair. Furthermore, the prediction device 100 according to the present embodiment and the prediction method using the same predict the hardness of the second heat-affected zone in the equipment to be predicted using the determined optimal pair, and therefore it is possible to predict the hardness of the second heat-affected zone in the equipment to be predicted with high accuracy.

[0054] Furthermore, when determining the optimal pair experimentally, it was necessary to create a huge number of test specimens, calculated by multiplying the number of welding methods for the first heat-affected zone 12, the number of welding conditions for the first heat-affected zone 12, and the number of plate thicknesses of the first heat-affected zone 12, and then conduct experiments. In contrast, the prediction device 100 according to the present embodiment and the prediction method using the same can determine the optimal pair by calculation. Therefore, the prediction device 100 according to the present embodiment and the prediction method using the same can reduce the effort required for experiments.

[0055] Furthermore, the prediction device 100 according to the present embodiment and the prediction method using the same can predict the hardness of the second heat-affected zone in the equipment to be predicted by effectively utilizing the inspection results of the inspected equipment 10, which is equipment that has already been inspected for cracking. Therefore, the prediction device 100 according to the present embodiment and the prediction method using the same can predict the hardness of the second heat-affected zone simply and at low cost, compared to a comparative example in which multiple heat-affected zone test specimens are prepared and a crack test is performed on each test specimen to predict hardness.

[0056] Furthermore, as described above, the prediction device 100 according to the present embodiment and the prediction method using the same may calculate the probability of crack occurrence in inspected equipment using the results of cracking obtained from an overhaul inspection performed on the inspected equipment. As a result, the prediction device 100 according to the present embodiment and the prediction method using the same can predict the hardness of the second heat-affected zone more easily and at lower cost than a comparative example in which multiple heat-affected zone test specimens are prepared and cracking tests are performed on each test specimen to predict hardness. Furthermore, the prediction device 100 according to the present embodiment and the prediction method using the same can predict the hardness of the second heat-affected zone in the equipment to be predicted with higher accuracy.

[0057] As described above, the inspected equipment 10 and the prediction target equipment may be tanks that store the same type of liquefied gas. This allows the prediction device 100 according to the present embodiment and the prediction method using the same to more accurately predict the hardness of the second heat-affected zone in the prediction target equipment.

[0058] Although the embodiments have been described above with reference to the accompanying drawings, it goes without saying that the present disclosure is not limited to the above-described embodiments. It is clear that a person skilled in the art can conceive of various modifications or alterations within the scope of the claims, and it is understood that these also naturally fall within the technical scope of the present disclosure.

[0059] For example, in the above embodiment, the inspected facility 10 and the prediction target facility are tanks that store the same type of liquefied gas. However, the inspected facility 10 and the prediction target facility may be tanks that store different types of liquefied gas. Furthermore, the inspected facility 10 and the prediction target facility may be different facilities, such as a tank and a pipe.

[0060] In the above embodiment, the processing unit 212 determines one optimal pair. However, the processing unit 212 may determine a different optimal pair for each parameter that affects whether or not a crack occurs. The parameters that affect whether or not a crack occurs include, for example, parameters related to the material of the first heat-affected zone 12, parameters related to welding of the first heat-affected zone 12, and parameters related to the environment to which the first heat-affected zone 12 is exposed. The material parameters include, for example, the type of material and the plate thickness. The material type includes, for example, the material composition and the steel plate standard. The welding parameters include, for example, the welding method. The environment parameters include, for example, the liquid contact state.

[0061] For example, in the pair determination processing step S140, an optimal pair may be determined for each steel plate standard of the first heat-affected zone 12. In this case, in the prediction processing step S150, the optimal pair may be used when predicting the hardness of a second heat-affected zone having the same steel plate standard as the first heat-affected zone 12 used to determine the optimal pair.

[0062] Similarly, in the pair determination process step S140, an optimal pair may be determined for each welding method of the material of the first heat-affected zone 12. In this case, in the prediction process step S150, the optimal pair may be used when predicting the hardness of a second heat-affected zone that is welded by the same method as the first heat-affected zone 12 used to determine the optimal pair.

[0063] Similarly, in the pair determination processing step S140, an optimal pair may be determined for each liquid contact state of the first heat-affected zone 12. In this case, in the prediction processing step S150, the optimal pair may be used when predicting the hardness of a second heat-affected zone having the same liquid contact state as the first heat-affected zone 12 used to determine the optimal pair.

[0064] As a result, the prediction device and the prediction method using the same are able to predict the hardness of the second heat-affected zone in the equipment to be predicted with even greater accuracy.

[0065] The present disclosure can contribute, for example, to Goal 12 of the Sustainable Development Goals (SDGs), "Ensure sustainable consumption and production patterns."

[0066] 10: Inspected equipment 12: First heat-affected zone 100: Prediction device 102: Processor 104: Memory

Claims

1. A prediction method comprising: calculating the probability of crack occurrence in inspected equipment, which is equipment that has already been inspected for cracks, using the results of cracks in multiple first heat-affected zones; using each of multiple candidates for a pair of a maximum temperature during welding and a holding time at the maximum temperature, predicting the hardness of the multiple first heat-affected zones for each candidate, and calculating a predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment, using the hardness prediction results; determining an optimal pair of the maximum temperature and the holding time using the occurrence probability and the predicted occurrence probability for each candidate; and using the determined optimal pair, predicting the hardness of the second heat-affected zone in prediction target equipment, which is equipment that is the target for predicting the probability of crack occurrence.

2. The prediction method according to claim 1, wherein calculating the probability of crack occurrence in the inspected equipment includes calculating the probability of crack occurrence in the inspected equipment using the results of cracks obtained from an overhaul inspection performed on the inspected equipment.

3. A prediction method according to claim 1 or 2, wherein determining the optimum pair includes determining a different optimum pair for each parameter that influences whether or not cracks will occur.

4. A prediction method as described in claim 1 or 2, wherein the inspected equipment and the equipment to be predicted are tanks storing the same type of liquefied gas.

5. A prediction device comprising one or more processors and one or more memories connected to the processors, wherein the processor performs processes including: calculating the probability of crack occurrence in inspected equipment, which is equipment that has already been inspected for cracks, using the results of cracks in multiple first heat-affected zones of the inspected equipment; predicting the hardness of the multiple first heat-affected zones for each of multiple candidates for a pair of a maximum temperature during welding and a holding time at the maximum temperature, and calculating a predicted occurrence probability, which is a predicted value of the probability of crack occurrence in the inspected equipment, using each of multiple candidates for a pair of a maximum temperature and a holding time at the maximum temperature, and using the hardness prediction results; determining an optimal pair of the maximum temperature and the holding time using the occurrence probability and the predicted occurrence probability for each candidate; and predicting the hardness of the second heat-affected zone in prediction target equipment, which is equipment that is the target for predicting the probability of crack occurrence, using the determined optimal pair.

Images