Prediction method for future annual average consumption of spare parts, electronic equipment and storage medium

Abstract

The invention is suitable for the field of nuclear power, and provides a method and a device for predicting the future annual average consumption of spare parts, electronic equipment and a storage medium, and the method comprises the steps: determining the function position of a target spare part in a nuclear power station, the function positions comprise a function position of a planned A-type demand, a function position of a planned B-type demand and a function position of a sudden demand; according to the historical receiving data of the target spare parts, calculating the receiving quantity of the target spare parts required by the various demands, and correcting the receiving quantity of the corresponding demands by using the functional position quantity distribution of the planned B-type demands and / or the sudden demands of the target spare parts and the annual failure probability distribution of the same-type spare parts in the calculation process; and summarizing the receiving quantity of the target spare parts required by various demands to obtain the future annual average receiving quantity of the target spare parts in the nuclear power station.

Description

Technical Field

[0001] The present application belongs to the field of nuclear power, and in particular relates to a method and device for predicting the future average annual usage of spare parts, an electronic device, and a computer-readable storage medium. Background Art

[0002] The demand for nuclear power spare parts comes from the maintenance activities of nuclear power plants. There are many types of maintenance activities. Compared with the spare parts demand forecasting method that does not consider the type of maintenance, the demand forecasting method that classifies the spare parts demand according to maintenance activities is more accurate. Nuclear power maintenance activities can be divided into planned (preventive) maintenance and emergency (corrective) maintenance. Planned maintenance work mainly refers to maintenance activities that are actively carried out according to the maintenance outline or corresponding maintenance plan; emergency maintenance work refers to maintenance activities other than planned maintenance work. The spare parts required for planned maintenance can be subdivided into planned Class A spare parts and planned Class B spare parts. Planned Class A spare parts refer to spare parts that must be replaced during maintenance activities, and planned Class B spare parts refer to spare parts that are selectively replaced during maintenance activities according to the status of on-site equipment.

[0003] The calculation process of planned Class B demand and / or sudden demand depends on historical use records. For spare parts with no historical use or large fluctuations in use, there are two potential problems:

[0004] The first problem is that for projects with no records of use, the subsequent predicted spare parts demand quantity is 0, which is inconsistent with the actual situation and easily causes spare parts guarantee problems (inventory shortage). The spare parts installed in a certain functional position have never failed in the past, but it does not mean that they will not fail in the future. As the unit's operating time increases, the equipment will fail at some point in the future, and there will be a need to replace spare parts.

[0005] The second problem is that the use of a single spare part is highly random. If only the use data of the spare part itself is used without effectively using the use distribution data of this type of spare parts, there is a certain probability that the predicted demand quantity value will be too large or too small. Summary of the invention

[0006] The embodiments of the present application provide a method and device for predicting the future average annual usage of spare parts, an electronic device and a computer-readable storage medium, which can solve the problem in the related art that the calculation process of planned Class B demand and / or sudden demand for spare parts depends on historical usage records.

[0007] In the first aspect, an embodiment of the present application provides a method for predicting the future average annual usage of spare parts, the method comprising: determining the functional position of the target spare parts in the nuclear power plant, the functional position including the functional position of the planned Class A demand, the functional position of the planned Class B demand and the functional position of the sudden demand; calculating the usage quantity of the target spare parts required for each type of demand according to the historical usage data of the target spare parts, and in the calculation process, using the functional position quantity distribution of the planned Class B demand and / or sudden demand of the target spare parts and the annual failure probability distribution of spare parts of the same type to correct the usage quantity of the corresponding demand; summarizing the usage quantity of the target spare parts required for each type of demand to obtain the future average annual usage of the target spare parts in the nuclear power plant.

[0008] In the second aspect, an embodiment of the present application provides a device for predicting the future average annual usage of spare parts, the device comprising: a determination module, used to determine the functional position of the target spare parts in the nuclear power plant, the functional position including the functional position of the planned Class A demand, the functional position of the planned Class B demand and the functional position of the sudden demand; a calculation module, used to calculate the usage quantity of the target spare parts required for each type of demand according to the historical usage data of the target spare parts, and in the calculation process, the functional position quantity distribution of the planned Class B demand and / or sudden demand of the target spare parts and the annual failure probability distribution of spare parts of the same type are used to correct the usage quantity of the corresponding demand; a summary module, used to summarize the usage quantity of the target spare parts required for each type of demand to obtain the future average annual usage of the target spare parts in the nuclear power plant.

[0009] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable by the processor, and when the processor executes the computer program, the spare parts inventory parameter allocation method described in the first aspect above is implemented.

[0010] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the spare parts inventory parameter allocation method described in the first aspect above.

[0011] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the spare parts inventory parameter allocation method described in the first aspect above.

[0012] Compared with the prior art, the embodiments of the present application have the following beneficial effects: by determining the functional position of the target spare parts in the nuclear power plant, the functional position includes the functional position of the planned Class A demand, the functional position of the planned Class B demand and the functional position of the sudden demand; according to the historical use data of the target spare parts, the use quantity of the target spare parts required for each type of demand is calculated respectively, and in the calculation process, the functional position quantity distribution of the planned Class B demand and / or sudden demand of the target spare parts and the annual failure probability distribution of the same type of spare parts are used to correct the use quantity of the corresponding demand; the use quantity of the target spare parts required for each type of demand is summarized to obtain the future annual average use quantity of the target spare parts in the nuclear power plant. The use quantity of the planned Class B demand and / or sudden demand is corrected by the annual failure probability distribution of the same type of spare parts, which reduces the dependence on historical use records in the process of predicting the planned Class B demand and / or sudden demand, and improves the accuracy of the prediction, especially for spare parts with no historical use or large fluctuations in use. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0014] Figure 1 is a structural schematic diagram of an electronic device provided by an embodiment of the present application;

[0015] Figure 2 It is a flowchart of a spare parts inventory parameter allocation method provided in an embodiment of the present application;

[0016] Figure 3 is a schematic diagram of determining type data of a spare part in a specific example of the present application;

[0017] Figure 4 It is a schematic diagram of a process of allocating power plant-level inventory parameters to target spare parts with an inventory parameter type of VB by using total cost minimization in a specific example of the present application;

[0018] Figure 5 It is a schematic diagram of a process of allocating power plant-level inventory parameters to target spare parts with an inventory parameter type of ZB using the total cost minimization principle in a specific example of the present application;

[0019] Figure 6 It is a schematic diagram of a process of allocating power plant-level inventory parameters to target spare parts with an inventory parameter type of VB using the allocation minimization principle in a specific example of the present application;

[0020] Figure 7It is a schematic diagram of a process of allocating power plant-level inventory parameters to target spare parts with an inventory parameter type of ZB using the allocation minimization principle in a specific example of the present application;

[0021] Figure 8 yes Figure 5-Figure 7 Schematic diagram of the summary and comparison of the results of the inventory parameter allocation at the power plant level;

[0022] Fig. 9 It is a flow chart of a method for allocating inventory parameters of a batch of spare parts in a specific embodiment of the present application;

[0023] Fig.10 It is a flow chart of a method for predicting the average annual usage of spare parts in the future provided by an embodiment of the present application;

[0024] Fig.11 It is a flow chart of a method for predicting the average annual usage of spare parts in the future provided by an embodiment of the present application;

[0025] Fig.12 is a schematic diagram of a spare parts inventory prediction model in a specific example of the present application;

[0026] Fig.13 It is a distribution diagram of the triggered purchase application quantity and the approved purchase quantity of a certain type of spare parts in a specific example of this application;

[0027] Fig.14 It is a schematic diagram of the total purchase application amount and purchase application approval time of a certain type of spare parts in a specific example of this application;

[0028] Fig.15 This is a monthly distribution diagram of the number of annual withdrawals in a specific example of this application;

[0029] Fig.16 It is a flow chart of a method for predicting the average annual usage of spare parts in the future provided by another embodiment of the present application;

[0030] Fig.17 It is a schematic diagram of constructing a sample set and training an inventory prediction network for spare parts with different types of data in a specific example of the present application;

[0031] Fig.18 It is a structural diagram of an inventory prediction network for spare parts with type data of PD+EX in a specific example of the present application;

[0032] Fig.19 It is a structural diagram of an inventory forecasting network for spare parts with type data of VB+EX and ZB+EX in a specific example of the present application;

[0033] Fig. 20It is a structural diagram of an inventory prediction network for spare parts with type data of VB+HB and ZB+HB in a specific example of the present application;

[0034] Fig.21 It is a structural diagram of an inventory forecasting network for spare parts with type data of VB+FX and ZB+FX in a specific example of the present application;

[0035] Fig. 22 It is a structural schematic diagram of a spare parts inventory parameter allocation device provided in an embodiment of the present application;

[0036] Fig.23 This is a schematic diagram of the structure of a device for predicting the average annual usage of spare parts in the future provided by an embodiment of the present application;

[0037] Fig.24 It is a schematic diagram of the structure of a device for predicting the future annual average inventory of spare parts provided in one embodiment of the present application;

[0038] Fig.25 It is a structural schematic diagram of a device for predicting the future average annual inventory of spare parts provided in another embodiment of the present application. DETAILED DESCRIPTION

[0039] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0040] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0041] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0042] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0043] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0044] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0045] The various embodiments of the present application can be combined with each other without conflict.

[0046] The spare parts inventory parameter allocation method, the future annual average use of spare parts prediction method and the future annual average inventory prediction method of spare parts provided in the embodiments of the present application can be applied to electronic devices, including but not limited to servers, server clusters, mobile phones, tablet computers, laptop computers, desktop computers, personal digital assistants, wearable devices and other electronic devices with computing functions. The embodiments of the present application do not impose any restrictions on the specific types of electronic devices.

[0047] Figure 1 FIG. 1 is a block diagram showing a partial structure of an electronic device provided by an embodiment of the present application. Figure 1 , the electronic device includes: a processor 10, a memory 20, a bus 30, an input device 40, an output device 50 and a communication device 60. The processor 10 and the memory 20 are connected to each other through the bus 30, and the input device 40, the output device 50 and the communication device 60 are also connected to the bus 30. Those skilled in the art can understand that Figure 1 The structure of the electronic device shown in the figure does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0048] Combine the following Figure 1 A detailed introduction to the various components of electronic equipment:

[0049] The processor 10 is the control center of the electronic device, and can run the program stored in the memory 20 to perform various functions and process data. The processor 10 can be a central processing unit (CPU), and the processor 10 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In some embodiments, the processor 10 may include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0050] The memory 20 is used to store an operating system, an application program, a boot loader, data, and other programs, such as program codes of computer programs. The memory 20 can also be used to temporarily store data required for executing the program and generated. The memory 20 may include a high-speed random access memory, and may also include a non-volatile memory, such as a flash memory, a hard disk, a multimedia card, a card-type memory, etc. The memory 20 may include a storage unit disposed inside the electronic device, such as a hard disk of the electronic device, and / or a removable external storage unit, such as a mobile hard disk, a USB flash drive, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, etc.

[0051] The input device 40 may include at least one of a keyboard, a mouse, a touch panel, a joystick, etc., and is used to collect user input operations to generate corresponding operation instructions.

[0052] The output device 50 is used to output information to be provided to the user. The output device 50 generally includes a display, and optionally, a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc. can be used. In addition, the output device can further include a speaker.

[0053] The communication device 60 may include a modem, a network card, etc., for establishing a network connection with other electronic devices and communicating with each other.

[0054] The spare parts inventory parameter allocation method, the prediction method of the future average annual use of spare parts, and the prediction method of the future average annual inventory of spare parts provided by the embodiment of the present application can be implemented as a computer software program. For example, the embodiment of the present application provides a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 60, and / or installed from a detachable external storage unit. When the computer program is executed by the processor 10, the various functions defined in the spare parts inventory parameter allocation method, the prediction method of the future average annual use of spare parts, and the prediction method of the future average annual inventory of spare parts provided by the embodiment of the present application are realized.

[0055] Figure 2 A schematic flow chart of a spare parts inventory parameter allocation method provided in an embodiment of the present application is shown. As an example but not a limitation, the method can be applied to the above-mentioned electronic device.

[0056] S11: Obtain the group-plant-level inventory parameters of the target spare parts, and determine the allocation principle of the target spare parts.

[0057] In order to achieve unified management of spare parts, the concept of virtual spare parts library can be introduced on the basis of physical spare parts library. Physical spare parts library refers to the physical warehouse used by nuclear power plants to store physical spare parts, while virtual spare parts library refers to an information platform that integrates spare parts inventory data and other information of different nuclear power plants in a nuclear power group to achieve unified management of spare parts.

[0058] The spare parts discussed in the embodiments of the present application refer to common code spare parts, i.e., spare parts that are used in two or more nuclear power plants (also called nuclear power plants, or power plants for short), which are technically universal and interchangeable in installation. There can be many kinds of common code spare parts used in nuclear power plants, and inventory parameters can be allocated separately for different spare parts. The target spare parts can also be called current spare parts, which refer to spare parts being processed.

[0059] The allocation principle may include at least one of total cost minimization, benefit alignment, and allocation minimization.

[0060] When there is only one allocation principle, one can be selected from total cost minimization, benefit consistency, and allocation minimization according to actual conditions or randomly. When there is more than one allocation principle, the allocation principle can be determined with reference to the unit price of the target spare parts. Specifically, the unit price or item ratio of the target spare parts can be compared with the set threshold to obtain a comparison result, and then the allocation principle of the target spare parts can be determined based on the comparison result. The item ratio is calculated based on the unit price and is generally positively correlated with the unit price. For example, a batch of spare parts can be sorted from small to large according to the unit price, and then the ratio of the serial number of each spare part to the total number of spare parts is calculated as the item ratio.

[0061] There is a risk of damage to spare parts during the allocation and transportation process. For some high-value spare parts, the probability of spare parts allocation needs to be reduced to the greatest extent possible to reduce the risk of spare parts being damaged during allocation and transportation. The corresponding principle can be allocation minimization. For low-priced spare parts, even if they are damaged during transportation, the loss is acceptable. More attention should be paid to the allocation cost. The corresponding principle can be total cost minimization. Total cost minimization refers to minimizing the total allocation cost, that is, combining the number of spare parts allocated and the allocation fee, and minimizing the comprehensive total cost. Therefore, when the number of thresholds is set to 1, allocation minimization can be selected for spare parts whose unit price / number of items is greater than the set threshold, and total cost minimization can be selected for spare parts whose unit price is less than the set threshold.

[0062] In some embodiments, in addition to minimizing the allocation of high unit prices and minimizing the total cost of low unit prices, the uniformity of the benefits of medium unit prices is also introduced to maintain the overall balance of the reserve amounts of each nuclear power plant participating in the shared allocation, that is, after participating in the shared allocation, the overall benefit ratio of each nuclear power plant is basically the same. Specifically, the thresholds are set to include a first threshold and a second threshold. When the unit price or the number of items is less than or equal to the first threshold, the allocation principle is determined to be the minimization of the total cost; when the unit price or the number of items is greater than the first threshold and less than or equal to the second threshold, the allocation principle is determined to be the uniformity of benefits; when the unit price or the number of items is greater than the second threshold, the allocation principle is determined to be the minimization of allocation. For example, for the unit price, the first threshold can be 5000, and the second threshold can be 10000; for the number of items, the first threshold can be 0.7, and the second threshold can be 0.8.

[0063] S12: Allocate the group plant level inventory parameters of the target spare parts to the nuclear power plants according to the allocation principle to obtain the power plant level inventory parameters of the target spare parts.

[0064] The plant group inventory parameters include at least one of the plant group reorder point value, the maximum inventory value and the fixed purchase batch value. Correspondingly, the power plant level inventory parameters include at least one of the plant group reorder point value, the maximum inventory value and the fixed purchase batch value.

[0065] The specific content of the inventory parameters is related to the type data of the spare parts. Therefore, the cluster-level inventory parameters that need to be allocated can be determined based on the type data of the target spare parts at the cluster-level. The cluster-level inventory parameters are allocated to the nuclear power plant according to the allocation principles and the type data of the cluster-level to obtain the power plant-level inventory parameters.

[0066] Type data is mainly used to distinguish the conditions that trigger purchase requisitions for spare parts and determine the purchase batch, including inventory parameter types and purchase batch types.

[0067] Inventory parameter types include the first type (PD), the second type (VB) and the third type (ZB). The first type triggers purchase requisitions based on inventory quantity, regular order quantity and reserved quantity; the second type triggers purchase requisitions based on inventory quantity, regular order quantity and reorder point value; the third type triggers purchase requisitions based on inventory quantity, regular order quantity, reserved quantity and reorder point value.

[0068] Since the VB type does not monitor the reserved quantity of spare parts, the VB type cannot be mixed with the PD and ZB types. If the inventory parameters of the shared-code spare parts group plant are of VB type, then the inventory parameters at the power plant level should all be of VB type, and the sum of the reorder point values ​​at the power plant level must be equal to the reorder point value of the group plant inventory parameters. If the inventory parameters of the shared-code spare parts group plant are of ZB type, and the inventory parameters at the power plant level are of ZB type or PD type, the sum of the reorder point values ​​at the power plant level must be equal to the reorder point value of the group plant inventory parameters. If the inventory parameters of the shared-code spare parts group plant are of PD type, the inventory parameters at the power plant level can only be of PD type.

[0069] The purchase batch type includes the fourth type, the fifth type and the sixth type. The fourth type (EX) is an unspecified purchase batch, which is related to the inventory quantity, the regular order quantity, and possibly the reserved quantity and the reorder point value; the fifth type (FX) is a specified fixed purchase batch, which is related to the set fixed purchase batch value; the sixth type (HB) is a specified variable purchase batch, which is related to the inventory quantity, the regular order quantity, the maximum inventory value, and possibly the reserved quantity and the reorder point value.

[0070] If the group-plant purchasing batch of shared-code spare parts is EX type, the purchasing batch at the power plant level can only be EX type; if the group-plant purchasing batch of shared-code spare parts is FX, the purchasing batch at the power plant level can only be FX type or EX type, and the sum of the fixed batch values ​​at the power plant level must be equal to the fixed batch value at the group-plant level; if the group-plant purchasing batch of shared-code spare parts is HB, the purchasing batch at the power plant level must be HB type or EX type.

[0071] Generally speaking, when the inventory parameter type is PD, the purchase batch type can only be EX; when the inventory parameter type is VB or ZB, the purchase batch type can be any one of EX, FX, and HB.

[0072] Based on the importance, unit price and historical usage data of spare parts, the type data of spare parts in each power plant can be determined. For example, the method of determining the type data of spare parts is as follows: Figure 3 As shown. Spare parts can be divided into high-turnover spare parts, low-turnover spare parts, and unused spare parts by the frequency of use. High-turnover spare parts refer to spare parts that have been used in 3 of the past 5 years or in 2 of the past 3 years; low-turnover spare parts refer to spare parts that have been used in the past 5 years but are not included in the high-turnover spare parts range; unused spare parts refer to spare parts that have not been used in the past 5 years. Spare parts can be divided into high-value spare parts and low-value spare parts by value. High-value spare parts refer to spare parts with a purchase price of more than RMB 10,000; low-value spare parts refer to spare parts that are not included in the high-value spare parts range. Spare parts can be divided into large use fluctuation and small use fluctuation by fluctuation. Generally, it can be considered that spare parts with a small use fluctuation when the standard deviation of the historical use quantity divided by the average use quantity is less than 0.5 are used for spare parts with small use fluctuation, and in other cases, are used for spare parts with large use fluctuation. Spare parts can be divided into important spare parts and non-important spare parts by importance of spare parts. When a spare part has no history of use and is a non-important spare part, when the inventory parameter of the spare part at the group plant level is VB, in order to ensure that the inventory parameter at the power plant level can be allocated normally, its inventory parameter at the power plant level must also be set to VB.

[0073] In inventory parameters, the reorder point value is mainly used to trigger purchase requisitions. The maximum inventory value and fixed purchase batch value are used to determine the purchase batch. The following describes the allocation of different inventory parameters for spare parts with different types of data.

[0074] Reorder point value

[0075] When the inventory parameter type of the plant cluster layer is PD, the reorder point value of the plant cluster layer is not allocated. Since the purchase batch type corresponding to PD is only EX, and when the purchase batch type of the plant cluster layer is EX, the maximum inventory value and fixed purchase batch value of the plant cluster layer are not allocated, it can be seen that spare parts of the PD+EX type are not allocated any inventory parameters.

[0076] When the inventory parameter type of the plant group layer is VB, the reorder point value of the plant group layer is allocated to the nuclear power plant according to the allocation principle to obtain the reorder point value of the power plant layer, wherein the minimum reorder point value of the power plant layer is 1. When the inventory parameter type of the plant group layer is ZB, the reorder point value of the plant group layer is allocated to the nuclear power plant according to the allocation principle to obtain the reorder point value of the power plant layer, and the type data of the power plant layer of the nuclear power plant whose reorder point value of the power plant layer is 0 is modified to PD+EX.

[0077] Allocation using the principle of minimizing the total cost requires the use of the normalized deployment cost matrix F' between nuclear power plants and the future average annual usage matrix R of the target spare parts as shown below.

[0078]

[0079] R = [r1, …, r n T (2)

[0080] where n represents the total number of nuclear power plants in the group, refers to the single deployment cost from nuclear power plant i to nuclear power plant j, i, j = 1, 2, …, n, is determined based on the deployment distance between nuclear power plant i and nuclear power plant j and the fixed cost of a single deployment. r i in the future average annual usage matrix R is the future average annual usage of the target spare part at nuclear power plant i, which can be obtained by predicting the annual usage of the target spare part at nuclear power plant i in a specified number of future years and then calculating the average value. The prediction method of the annual usage can refer to the description of the following embodiments.

[0081] When the inventory parameter type at the group plant level is VB and the allocation principle is to minimize the total cost, initialize the allocation matrix according to the future average annual usage matrix of the target spare parts and the reorder point value at the group plant level; set the reorder point value at the power plant level to 1, and update the reorder point value and the allocation matrix at the group plant level; output the reorder point value at the power plant level when the reorder point value at the group plant level is less than 1, and in other cases, calculate the total deployment cost matrix according to the deployment cost matrix and the allocation matrix, and allocate the reorder point value at the group plant level to the power plant level one by one according to the principle of minimizing the total deployment cost.

[0082] Specifically, when the inventory parameter type at the group plant level is VB and the allocation principle is to minimize the total cost, allocating the reorder point value P1 at the group plant level of the target spare part to the power plant level includes the following steps.

[0083] Step 1: Read the reorder point value P1, the deployment cost matrix F′, initialize and set the reorder point value matrix P of the power plant level parameters, and initialize p i = 0;

[0084] Step 2: Let the number of nuclear power plants using this spare part be m, and select the allocation plan:

[0085] Step 2.1: If P1 < m, the parameters cannot be allocated, feedback that the reorder point value P1 at the group plant is incorrect, and go to Step 4;

[0086] Step 2.2: If P1 = m, the reorder point value p of this spare part at the power plant level parameters i ​If both are 1, go to step 5;

[0087] Step 2.3: If P1>m, go to step 3;

[0088] Step 3: Based on the future average annual demand matrix R of spare parts in each nuclear power plant, calculate the allocation plan:

[0089] Step 3.1: Calculate the allocation matrix Z, let Z = R × P1, then

[0090] Step 3.2: Traverse the nuclear power plant, if z i ≤1, then set the reorder point p of nuclear power plant i i =1, execute P1=P1-1, z i =0; if z i >1, then initialize the setting p i =1, execute P1=P1-1, z i =z i -1; after traversing the nuclear power plant, execute

[0091] Step 3.3: If P1<1, go to step 5; if P1≥1, based on the dispatch cost matrix F′ in formula 5, calculate the total dispatch cost matrix Q, let Q=F′·Z, obtain the nuclear power plant with the smallest Q value, let it be nuclear power plant j, and execute p j =p j +1, P1=P1-1; if z j >1, execute z j =z j -1, otherwise execute z j =0; Update the allocation matrix Z and execute

[0092] Step 3.4: If P1 ≥ 1, go to step 3.3; otherwise go to step 5;

[0093] Step 4: Output the group factory reorder point value P1 is incorrect, P1 value should be greater than m value, end.

[0094] Step 5: Output the common code spare parts at the power plant level re-order point p of each nuclear power plant i Value, end.

[0095] For example, a spare part is used in nuclear power plants 1 to 6, the reorder point value P1 at the cluster level is 11, the value of the allocation cost matrix F′ is shown in Formula 3, and the number of nuclear power plants m is 6. Execute step 2. Since the reorder point value P1 is greater than the number of nuclear power plants m, execute step 3. Calculate the average annual demand R of the spare part in each nuclear power plant in the future, as shown in Formula 4. Execute step 3.1 to obtain the allocation matrix Z, as shown in Formula 5. Execute step 3.2, traverse the nuclear power plants, and update the reorder point value matrix P for setting the parameters at the power plant level, as shown in Formula 6. Since the z of nuclear power plant 3 i The value is 0.43, so execute z i =0; z of other nuclear power plants i >1, execute z i =z i -1. After executing P1=P1-1 operation 6 times, the reorder point value P1 of the group factory is 5. Traverse and execute step 3.2. The updated allocation matrix Z is shown in Formula 7. Execute step 3.3. Since the condition P1≥1 is met, the total allocation cost matrix Q is calculated using the formula Q=F′·Z. The calculation result is shown in Formula 8. It can be obtained that the Q value of nuclear power plant 2 is the smallest. Therefore, the reorder point value p2 of nuclear power plant 2 is updated to 2, z2 is 0.49, and the reorder point value P1 of the cluster plant is 4. Update the allocation matrix Z. Execute step 3.4. Since the condition P1≥1 is met, go to step 3.3 and execute it in a loop. The reorder point value matrix P of the power plant layer parameters calculated in the end is shown in Formula 9. The allocation results under each step are as follows: Figure 4 shown.

[0096]

[0097] R=[2.90,3.10,0.50,1.40,3.20,1.70] T (4)

[0098] Z=[2.49,2.66,0.43,1.20,2.75,1.46] T (5)

[0099] P = [1,1,1,1,1,1] T (6)

[0100] Z=[1.34,1.49,0.00,0.18,1.57,0.41] T (7)

[0101] Q=[2.41,2.39,3.03,3.24,3.16,2.85] T (8)

[0102] P = [2,3,1,1,2,2]T (9)

[0103] When the inventory parameter type of the plant cluster layer is ZB and the allocation principle is to minimize the total cost, the allocation matrix is ​​initialized according to the future average annual usage matrix of the target spare parts and the reorder point value of the plant cluster layer; when the reorder point value of the plant cluster layer is less than 1, the reorder point value of the power plant layer is output; in other cases, the total allocation cost matrix is ​​calculated according to the allocation cost matrix and the allocation matrix, and the reorder point values ​​of the plant cluster layer are allocated to the power plant layer one by one according to the principle of minimizing the total allocation cost.

[0104] Specifically, when the inventory parameter type of the plant group layer is ZB and the allocation principle is to minimize the total cost, allocating the reorder point value P1 of the target spare part at the plant group layer to the power plant layer includes the following steps.

[0105] Step 1: Read the reorder point value P1 of the spare part in the cluster plant, adjust the cost matrix F′, initialize the reorder point value matrix P of the power plant layer parameters, and initialize p i =0;

[0106] Step 2: Based on the future average annual demand matrix R of spare parts in each nuclear power plant, calculate the allocation plan:

[0107] Step 2.1: Calculate the allocation matrix Z, let Z = R × P1, then

[0108] Step 2.2: If P1<1, go to step 3; if P1≥1, based on the dispatch cost matrix F′ in formula 5, calculate the total dispatch cost matrix Q, let Q=F′·Z, obtain the nuclear power plant with the smallest Q value, let it be nuclear power plant j, and execute p j =p j +1, P1=P1-1; if z j >1, execute z j =z j -1, otherwise execute z j =0; Update the allocation matrix Z and execute

[0109] Step 2.3: If P1≥1, go to step 2.2; otherwise, go to step 3;

[0110] Step 3: Output the reorder point p of the common code spare parts at the power plant level of each nuclear power plant i When the reorder point value of a nuclear power plant is 0, change its inventory parameter type to PD and end.

[0111] For example, the target spare part is used in Nuclear Power Plant 1 to Nuclear Power Plant 6. The reorder point value P1 of the group of plants is 11. The value of the deployment cost matrix F′ is as shown in Formula 3. The number of nuclear power plants m is 6. The future average annual demand R of the spare part in each nuclear power plant is as shown in Formula 4. The final calculated distribution result of the reorder point value at the power plant level is as Figure 5 shown.

[0112] When the inventory parameter type at the group of plants level is VB and the allocation principle is minimization of deployment, initialize the allocation matrix according to the future average annual requisition matrix of the target spare part and the reorder point value at the group of plants level; set the reorder point value at the power plant level to 1, and update the reorder point value and the allocation matrix at the group of plants level; round the allocation value in the allocation matrix and allocate it to the reorder point value at the power plant level, and update the reorder point value and the allocation matrix at the group of plants level; output the reorder point value at the power plant level when the reorder point value at the group of plants level is less than 1, and in other cases, allocate the reorder point value at the group of plants level to the power plant level one by one according to the principle of the largest allocation value.

[0113] Specifically, when the inventory parameter type at the group of plants level is VB and the allocation principle is minimization of deployment, allocating the reorder point value P1 at the group of plants level of the target spare part to the power plant level includes the following steps.

[0114] Step 1: Read the reorder point value P1 of the spare part at the group of plants, initialize and set the reorder point value matrix P of the power plant level parameters, and initialize p i = 0;

[0115] Step 2: Let the number of nuclear power plants using the spare part be m, and select the allocation plan:

[0116] Step 2.1: If P1 < m, the parameters cannot be allocated, feedback that the reorder point value P1 at the group of plants is incorrect, and go to Step 4;

[0117] Step 2.2: If P1 = m, the reorder point value p of the spare part at the power plant level parameters i are all 1, and go to Step 5;

[0118] Step 2.3: If P1 > m, execute Step 3;

[0119] Step 3: Based on the future average annual demand matrix R of the spare part in each nuclear power plant, calculate the allocation plan:

[0120] Step 3.1: Calculate the allocation matrix Z, let Z = R × P1, then

[0121] Step 3.2: Traverse the nuclear power plants. If z i ≤ 1, then set the reorder point p of nuclear power plant i i = 1, execute P1 = P1 - 1, z i=0; if z i >1, then initialize the setting p i =1, execute P1=P1-1, z i =z i -1; after traversing the nuclear power plant, execute

[0122] Step 3.3: Traverse the nuclear power plant, if z i ≥1, then execute p i =p i +int(z i ), P1=P1-int(z i ), z i =z i -int(z i ), where int(x) is the integer value of the extracted real number x; if z i <1, no action is performed; after traversing the nuclear power plant, execute

[0123] Step 3.4: If P1<1, go to step 5; if P1≥1, traverse the nuclear power plants, find the nuclear power plant with the largest Z value, set it as nuclear power plant j, and execute p j =p j +1, P1=P1-1; if z j >1, execute z j =z j -1, otherwise execute z j =0; Update the allocation matrix Z and execute

[0124] Step 3.5: If P1 ≥ 1, go to step 3.4; otherwise go to step 5;

[0125] Step 4: Output the group factory reorder point value P1 is incorrect, P1 value should be greater than m value, end.

[0126] Step 5: Output the reorder point p of the common code spare parts at the power plant level of each nuclear power plant i Value, end.

[0127] For example, the target spare parts are used in nuclear power plants 1 to 6, the reorder point value P1 of the group is 11, the number of nuclear power plants m is 6, and the future average annual demand R of spare parts in each nuclear power plant is shown in formula 4. The final calculated reorder point value distribution result of the power plant layer is as follows: Figure 6 shown.

[0128] When the inventory parameter type of the plant cluster layer is ZB and the allocation principle is to minimize allocation, the allocation matrix is ​​initialized according to the future average annual usage matrix of the target spare parts and the reorder point value of the plant cluster layer; the allocation values ​​in the allocation matrix are rounded and allocated to the reorder point value of the power plant layer, and the reorder point value and allocation matrix of the plant cluster layer are updated; when the reorder point value of the plant cluster layer is less than 1, the reorder point value of the power plant layer is output, and in other cases, the reorder point values ​​of the plant cluster layer are allocated to the power plant layer one by one according to the principle of the largest allocation value.

[0129] Specifically, when the inventory parameter type of the plant group layer is ZB and the allocation principle is allocation minimization, allocating the reorder point value P1 of the plant group layer of the target spare part to the power plant layer includes the following steps.

[0130] Step 1: Read the reorder point value P1 of the spare part in the cluster plant, initialize the reorder point value matrix P of the power plant layer parameters, and initialize p i =0;

[0131] Step 2: Based on the future average annual demand matrix R of spare parts in each nuclear power plant, calculate the allocation plan:

[0132] Step 2.1: Calculate the allocation matrix Z, let Z = R × P1, then

[0133] Step 2.2: Traverse the nuclear power plant, if z i ≥1, then execute p i =p i +int(z i ), P1=P1-int(z i ), z i =z i -int(z i ); if z i <1, no action is performed; after traversing the nuclear power plant, execute

[0134] Step 2.3: If P1<1, go to step 5; if P1≥1, traverse the nuclear power plants, find the nuclear power plant with the largest Z value, set it as nuclear power plant j, and execute p j =p j +1, P1=P1-1; if z j >1, execute z j =z j -1, otherwise execute z j =0; Update the allocation matrix Z and execute

[0135] Step 2.4: If P1 ≥ 1, go to step 2.3; otherwise, go to step 3;

[0136] Step 3: Output the reorder point p of the common code spare parts at the power plant level of each nuclear power plant i When the reorder point value of a nuclear power plant is 0, change its inventory parameter type to PD+EX and end.

[0137] For example, the target spare parts are used in nuclear power plants 1 to 6, the reorder point value P1 of the group is 11, the value of the deployment cost matrix F′ is shown in Formula 3, the number of nuclear power plants m is 6, and the future average annual demand for spare parts in each nuclear power plant R is shown in Formula 4. The final calculated power plant-level reorder point value distribution result is as follows: Figure 7 shown.

[0138] In the examples given above, the parameters of the target spare parts are the same. The target spare parts are all used in nuclear power plants 1 to 6. The reorder point value P1 of the group plant is 11. The value of the allocation cost matrix F′ is shown in Formula 3. The number of nuclear power plants m is 6. The average annual demand for spare parts in each nuclear power plant in the future is shown in Formula 4. Depending on the inventory parameter type and allocation principle, the obtained power plant-level reorder point value allocation results will also be different. The allocation results of the above examples are summarized and compared. Figure 8 As shown, method 1 refers to allocating target spare parts of type VB using the principle of total cost minimization; method 2 refers to allocating target spare parts of type ZB using the principle of total cost minimization; method 3 refers to allocating target spare parts of type VB using the principle of deployment minimization; method 4 refers to allocating target spare parts of type ZB using the principle of deployment minimization.

[0139] The allocation based on the principle of uniform returns requires the use of the return matrix K of each nuclear power plant as shown below.

[0140] K=[k1,...,k n ] (10)

[0141]

[0142] where k j is the rate of return of nuclear power plant j, j = 1, 2, ..., n, n is the total number of nuclear power plants. k j The calculation formula is shown in formula 11, where The power plant-level inventory parameters allocated to spare part i in nuclear power plant j after participating in shared deployment, as well as the historical usage data of spare part i in nuclear power plant j (average annual usage, maximum annual usage, annual usage standard deviation, and average annual usage times); is the inventory parameter of spare part i in nuclear power plant j before participating in shared allocation, as well as the historical usage data of spare part i in nuclear power plant j (average annual usage, maximum annual usage, standard deviation of annual usage, average annual usage times); g(·) is the estimation function used to calculate the average annual inventory in the future based on inventory parameters and historical usage data. When the shared code spare part has not completed the power plant level inventory parameter allocation, the g(U) and g(V) of the spare part are both 0 and do not participate in the calculation; is the purchase price of spare part i in nuclear power plant j.

[0143] The estimation function g(·) may be an iterative simulation model, which includes a collection model, a reservation model, an inventory model, an acceptance model, a triggering purchase application model, a purchase application approval model, and a purchase cycle model. In the calculation process of the collection model, the annual failure probability distribution matrix is ​​used to predict the correction terms in the collection quantity corresponding to the planned Class B demand and the sudden demand. Alternatively, the estimation function g(·) may be a neural network corresponding to the type data of the target spare parts. For details, please refer to the description of the subsequent embodiments. For the detailed content of the estimation function g(·), please refer to the description of the subsequent embodiments.

[0144] According to the definition of g(·) in Formula 11, the spare parts using the benefit consistency principle will not contribute to the calculation of the rate of return matrix K before the allocation is completed. That is to say, if the benefit consistency principle is to be used, it is necessary to first calculate the nuclear power plant benefit matrix based on the allocation results of spare parts using other allocation principles other than the benefit consistency principle, and then apply the benefit consistency principle to calculate the inventory parameter allocation plan for the remaining spare parts.

[0145] It can be seen that in a specific embodiment of the present application, Fig. 9 As shown, the inventory parameter allocation method for a batch of spare parts includes the following parts.

[0146] S110: extracting a first spare part whose unit price / item quantity ratio is less than or equal to a first threshold from the batch of spare parts.

[0147] S120: Allocate power plant-level inventory parameters of the first spare part using the total cost minimization principle.

[0148] S130: extracting second spare parts whose unit price / item quantity ratio is greater than a second threshold from the batch of spare parts.

[0149] S140: Allocate power plant-level inventory parameters of the second spare part using the allocation minimization principle.

[0150] There is no restriction on the execution order between S110 - S120 and S130 - S140 .

[0151] S150: Calculate the rate of return matrix of each nuclear power plant based on the power plant level inventory parameters of the first spare parts and the second spare parts.

[0152] S160: Power plant layer inventory parameters for allocating remaining spare parts using the principle of income equalization.

[0153] When the inventory parameter type at the group power plant layer is VB and the allocation principle is income equalization, initialize the allocation matrix according to the future average annual consumption volume matrix of the target spare part and the reorder point value at the group power plant layer; set the reorder point value at the power plant layer to 1, and update the reorder point value and the allocation matrix at the group power plant layer; round the allocation value in the allocation matrix and allocate it to the reorder point value at the power plant layer, and update the reorder point value and the allocation matrix at the group power plant layer; output the reorder point value at the power plant layer when the reorder point value at the group power plant layer is less than 1, and in other cases, obtain the rate of return matrix, and allocate the reorder point value at the group power plant layer to the reorder point value at the power plant layer of the nuclear power plant with the highest rate of return, where the rate of return matrix is calculated based on the future average annual inventory volume of other spare parts for which the inventory parameters at the group power plant layer have been allocated.

[0154] Specifically, when the inventory parameter type at the group power plant layer is VB and the allocation principle is income equalization, allocating the reorder point value P1 at the group power plant layer of the target spare part to the power plant layer includes the following steps.

[0155] Step 1: Read the reorder point value P1 of this spare part at the group power plant, initialize and set the reorder point value matrix P of the power plant layer parameters, initialize p i = 0, and calculate the nuclear power plant income matrix K;

[0156] Step 2: Let the number of nuclear power plants using this spare part be m, and select the allocation plan:

[0157] Step 2.1: If P1 < m, the parameters cannot be allocated, feedback that the reorder point value P1 at the group power plant is incorrect, and go to Step 4;

[0158] Step 2.2: If P1 = m, the reorder point value p of this spare part at the power plant layer parameters i are all 1, and go to Step 5;

[0159] Step 2.3: If P1 > m, execute Step 3;

[0160] Step 3: Based on the future average annual demand volume matrix R of the spare part at each nuclear power plant, calculate the allocation plan:

[0161] Step 3.1: Calculate the allocation matrix Z, let Z = R × P1, then

[0162] Step 3.2: Traverse the nuclear power plants. If z i ≤ 1, then set the reorder point p of nuclear power plant i i = 1, execute P1 = P1 - 1, z i = 0; If zi >1, then initialize the setting p i =1, execute P1=P1-1, z i =z i -1; after traversing the nuclear power plant, execute

[0163] Step 3.3: Traverse the nuclear power plant, if z i ≥1, then execute p i =p i +int(z i ), P1=P1-int(z i ), z i =z i -int(z i ); if z i <1, no action is performed; after traversing the nuclear power plant, execute

[0164] Step 3.4: If P1<1, go to step 5; if P1≥1, traverse the nuclear power plants and obtain the nuclear power plant with the largest K value in the profit matrix, let it be nuclear power plant j, and execute p j =p j +P1, go to step 5;

[0165] Step 4: Output the group factory reorder point value P1 is incorrect, P1 value should be greater than m value, end.

[0166] Step 5: Output the reorder point p of the common code spare parts at the power plant level of each nuclear power plant i Value, end.

[0167] When the inventory parameter type of the plant cluster layer is ZB and the allocation principle is the uniform return, the allocation matrix is ​​initialized according to the future average annual usage matrix of the target spare parts and the reorder point value of the plant cluster layer; the allocation values ​​in the allocation matrix are rounded and allocated to the reorder point value of the power plant layer, and the reorder point value and the allocation matrix of the plant cluster layer are updated; when the reorder point value of the plant cluster layer is less than 1, the reorder point value of the power plant layer is output, and in other cases, the yield matrix is ​​obtained, and the reorder point value of the plant cluster layer is allocated to the reorder point value of the power plant layer of the nuclear power plant with the largest yield, wherein the yield matrix is ​​calculated based on the future average annual inventory of other spare parts that have completed the allocation of the plant cluster layer inventory parameters.

[0168] Specifically, when the inventory parameter type of the plant group layer is ZB and the allocation principle is revenue consistency, allocating the reorder point value P1 of the plant group layer of the target spare part to the power plant layer includes the following steps.

[0169] Step 1: Read the reorder point value P1 of the spare part in the cluster plant, initialize the reorder point value matrix P of the power plant layer parameters, and initialize pi =0, calculate the nuclear power plant profit matrix K;

[0170] Step 2: Based on the future average annual demand matrix R of spare parts in each nuclear power plant, calculate the allocation plan:

[0171] Step 2.1: Calculate the allocation matrix Z, let Z = R × P1, then

[0172] Step 2.2: Traverse the nuclear power plant, if z i ≥1, then execute p i =p i +int(z i ), P1=P1-int(z i ), z i =z i -int(z i ); if z i <1, no action is performed; after traversing the nuclear power plant, execute

[0173] Step 2.3: If P1<1, go to step 3; if P1≥1, traverse the nuclear power plants and obtain the nuclear power plant with the largest K value in the profit matrix, let it be nuclear power plant j, and execute p j =p j +P1;

[0174] Step 3: Output the reorder point p of the common code spare parts at the power plant level of each nuclear power plant i When the reorder point value of a nuclear power plant is 0, change its inventory parameter type to PD+EX and end.

[0175] Maximum inventory value and fixed purchase batch value

[0176] When the purchasing batch type at the plant group level is EX, the maximum inventory value and fixed purchasing batch value at the plant group level are not allocated.

[0177] When the purchasing batch type at the plant cluster level is FX, the target nuclear power plant is selected according to the reorder point value at the power plant level, the fixed purchasing batch value at the plant cluster level is allocated to the target nuclear power plant, and the purchasing batch type at the power plant level of other nuclear power plants except the target nuclear power plant is modified to EX.

[0178] Specifically, when the purchasing batch type at the plant group level is FX, allocating the fixed purchasing batch value at the plant group level of the target spare part to the power plant level includes the following steps.

[0179] Step 1: Read the reorder point p of the spare part at the power plant level i Value, future annual average demand r i value;

[0180] Step 2: Compare the power plant level and order more goods i value, select the nuclear power plant with the largest reorder point value and set it as nuclear power plant j; if there are multiple power plants with the largest reorder point values, then select the nuclear power plant with the largest future annual average demand r i The power plant with the largest value is set as nuclear power plant j;

[0181] Step 3: Change the purchase batch type of nuclear power plant j to FX and fix the batch value to f j , the calculation formula is f j =P3, where P3 is the fixed batch value of the common code spare parts in the group factory, f j The fixed batch value for the common code spare parts at the power plant level of nuclear power plant j. Except for nuclear power plant j, the purchase batch type of other nuclear power plants is changed to EX, and no fixed batch value is set;

[0182] Step 4: Output the fixed batch value of the common code spare parts at the power plant level of each nuclear power plant, and end.

[0183] When the purchasing batch type at the group plant level is HB, the purchasing batch type at the power plant level of each nuclear power plant is counted. There are three situations: the purchasing batch types at all power plant levels are EX, the purchasing batch types at the power plant level include EX and HB, and the purchasing batch types at all power plant levels are HB.

[0184] When the purchase batch type of all power plant layers is EX, the target nuclear power plant is selected according to the reorder point value of the power plant layer, and the maximum inventory value of the group plant layer is allocated to the target nuclear power plant. Specifically, the number of nuclear power plants with the largest reorder point value at the power plant layer can be determined; when the number is 1, the nuclear power plant with the largest reorder point value at the power plant layer is taken as the target nuclear power plant; when the number is greater than 1, the nuclear power plant with the largest reorder point value at the power plant layer and the largest average annual use in the future is taken as the target nuclear power plant. When the purchase batch types at the power plant layer include EX and HB, the maximum inventory value at the group plant layer is allocated to the nuclear power plant with the purchase batch type of HB at the power plant layer according to the average annual use in the future. When the purchase batch type of all power plant layers is HB, the maximum inventory value at the group plant layer is allocated to each nuclear power plant according to the average annual use in the future.

[0185] Specifically, when the purchase batch type at the plant group level is HB and the purchase batch types at all power plant levels are EX, allocating the maximum inventory value of the target spare part at the plant group level to the power plant level includes the following steps.

[0186] Step 1: Read the reorder point p of the spare part at the power plant level i Value, future annual average demand r i Value, number of nuclear power plants m;

[0187] Step 2: Compare the power plant level and order more goods i value, select the nuclear power plant with the largest reorder point value and set it as nuclear power plant j; if there are multiple power plants with the largest reorder point values, then select the nuclear power plant with the largest future annual average demand r i The power plant with the largest value is set as nuclear power plant j;

[0188] Step 3: Change the purchase batch type of nuclear power plant j to HB, and the maximum inventory value to h j , the calculation formula is h j =P2-P1+p j , where P2 is the maximum inventory value of the common code spare parts in the group factory, P1 is the reorder point value of the common code spare parts in the group factory, and p j The reorder point value of the common code spare parts at the power plant level of nuclear power plant j. Except for nuclear power plant j, the purchase batch type of other nuclear power plants is changed to EX, and the maximum inventory value is not set;

[0189] Step 4: Output the maximum inventory value of the common code spare parts at the power plant level of each nuclear power plant, and end.

[0190] Specifically, when the purchase batch type at the plant cluster level is HB, and the purchase batch types at the power plant level include EX and HB, allocating the maximum inventory value of the target spare part at the plant cluster level to the power plant level includes the following steps.

[0191] Step 1: Read the reorder point p of the spare part at the power plant level i Value, future annual average demand r i The number of nuclear power plants with the power plant level purchase batch type set to EX is m1, and the number of nuclear power plants with the power plant level purchase batch type set to HB is m2;

[0192] Step 2: Traverse the nuclear power plants with the purchase batch type set to HB at the power plant level, and calculate the maximum inventory value. Where P2 is the maximum inventory value of the common code spare parts in the group factory, P1 is the reorder point value of the common code spare parts in the group factory, j is the reorder point value of the common code spare parts at the power plant level of nuclear power plant j, r j is the average annual demand of nuclear power plant j in the future, is the sum of the average annual demand for common code spare parts in the nuclear power plant with the purchase batch type set to HB at the power plant level, and round() is the rounding function;

[0193] Step 3: Calculation If the value is greater than 0, execute step 3.1; if the value is less than 0, execute step 3.2; if the value is equal to 0, execute step 4:

[0194] Step 3.1: Compare the maximum inventory values ​​initially allocated to each nuclear power plant, select the nuclear power plant with the largest maximum inventory value, and set it as nuclear power plant j; if there are multiple power plants with the largest maximum inventory values, then select the future average annual demand r from the nuclear power plants with the largest maximum inventory values. i The power plant with the largest value is set as nuclear power plant j; execute h j =h j -1, return to step 3;

[0195] Step 3.2: Compare the maximum inventory values ​​initially allocated to each nuclear power plant, select the nuclear power plant with the largest maximum inventory value, and set it as nuclear power plant j; if there are multiple power plants with the largest maximum inventory values, then select the future average annual demand r from the nuclear power plants with the largest maximum inventory values. i The power plant with the largest value is set as nuclear power plant j; execute h j =h j +1, return to step 3;

[0196] Step 4: Output the maximum inventory value of the common code spare parts at the power plant level of each nuclear power plant, and end.

[0197] Specifically, when the purchasing batch type at the plant group level is HB and the purchasing batch types at all power plant levels are HB, allocating the maximum inventory value of the target spare part at the plant group level to the power plant level includes the following steps.

[0198] Step 1: Read the reorder point p of the spare part at the power plant level i Value, future annual average demand r i Value, number of nuclear power plants m;

[0199] Step 2: Calculate the maximum inventory value allocated to each nuclear power plant. Where P2 is the maximum inventory value of the common code spare parts in the group factory, P1 is the reorder point value of the common code spare parts in the group factory, j is the reorder point value of the common code spare parts at the power plant level of nuclear power plant j, r j is the average annual demand of nuclear power plant j in the future, is the sum of the average annual demand for common code spare parts in all nuclear power plants in the future, and round() is the rounding function;

[0200] Step 3: Calculation If the value is greater than 0, execute step 3.1; if the value is less than 0, execute step 3.2; if the value is equal to 0, execute step 4:

[0201] Step 3.1: Compare the maximum inventory values ​​initially allocated to each nuclear power plant, select the nuclear power plant with the largest maximum inventory value, and set it as nuclear power plant j; if there are multiple power plants with the largest maximum inventory values, then select the future average annual demand r from the nuclear power plants with the largest maximum inventory values. iThe power plant with the largest value is set as nuclear power plant j; execute h j =h j -1, return to step 3;

[0202] Step 3.2: Compare the maximum inventory values ​​initially allocated to each nuclear power plant, select the nuclear power plant with the largest maximum inventory value, and set it as nuclear power plant j; if there are multiple power plants with the largest maximum inventory values, then select the future average annual demand r from the nuclear power plants with the largest maximum inventory values. i The power plant with the largest value is set as nuclear power plant j; execute h j =h j +1, return to step 3;

[0203] Step 4: Output the maximum inventory value of the common code spare parts at the power plant level of each nuclear power plant, and end.

[0204] The allocated power plant-level inventory parameters can be used to allocate purchase applications. Specifically, by setting the group plant inventory parameters of the virtual library common code spare parts, the trigger purchase application modules of multiple power plants are unified into one virtual factory, triggering the group plant total purchase application; based on the virtual library common code spare parts group plant inventory parameters, the inventory parameters of the spare parts in each power plant are allocated, that is, the virtual library common code spare parts power plant-level inventory parameters. Based on the triggered group plant total purchase application and the power plant-level inventory parameters of the spare parts, the purchase application is calculated and allocated, that is, the triggered group plant total purchase application is allocated to each power plant for procurement.

[0205] Through the implementation of this embodiment, the group-plant-level inventory parameters of the target spare parts are obtained, and the allocation principle of the target spare parts is determined, the allocation principle includes at least one of total cost minimization, benefit consistency and allocation minimization; the group-plant-level inventory parameters of the target spare parts are allocated to the nuclear power plant according to the allocation principle, and the power plant-level inventory parameters of the target spare parts are obtained, the group-plant-level inventory parameters include at least one of the group-plant-level reorder point value, the maximum inventory value and the fixed purchase batch value, and the inventory parameter allocation plan of the spare parts at the power plant level can be automatically calculated according to the preset rules, thereby optimizing the overall procurement and allocation of spare parts.

[0206] Fig.10 A schematic flow chart of a method for predicting the future average annual usage of spare parts provided in an embodiment of the present application is shown. As an example but not a limitation, the method can be applied to the above-mentioned electronic equipment.

[0207] S21: Determine the functional position of the target spare part in the nuclear power plant.

[0208] S22: Calculate the target spare parts requisition quantity required for each type of demand based on the historical requisition data of the target spare parts. During the calculation process, use the functional position quantity distribution of the planned Class B demand and / or sudden demand for the target spare parts and the annual failure probability distribution of the same type of spare parts to correct the requisition quantity for the corresponding demand.

[0209] Specifically, during the calculation process, the functional position quantity distribution of the planned Class B demand for target spare parts and the annual failure probability distribution of spare parts of the same type can be used to correct the quantity of planned Class B demand; and / or the functional position quantity distribution of sudden demand for target spare parts and the annual failure probability distribution of spare parts of the same type can be used to correct the quantity of sudden demand.

[0210] S23: Summarize the number of target spare parts required for various demands to obtain the average annual number of target spare parts used in the nuclear power plant in the future.

[0211] The demand for nuclear power spare parts comes from the maintenance activities of nuclear power plants. There are many types of maintenance activities. Compared with the spare parts demand forecasting method that does not consider the type of maintenance, the demand forecasting method that classifies the spare parts demand according to maintenance activities is more accurate. Nuclear power maintenance activities can be divided into planned (preventive) maintenance and emergency (corrective) maintenance. Planned maintenance work mainly refers to maintenance activities that are actively carried out according to the maintenance outline or corresponding maintenance plan; emergency maintenance work refers to maintenance activities other than planned maintenance work. The spare parts required for planned maintenance can be subdivided into planned Class A spare parts and planned Class B spare parts. Planned Class A spare parts refer to spare parts that must be replaced during maintenance activities, and planned Class B spare parts refer to spare parts that are selectively replaced during maintenance activities according to the status of on-site equipment.

[0212] Therefore, the functional positions where the target spare parts are installed in the nuclear power plant can be divided, including the functional positions of planned Class A demand, planned Class B demand and sudden demand. The future annual demand of the target spare parts in each functional position is summarized to obtain the annual requisition quantity of the target spare parts in the future, and then the average annual requisition quantity of the specified number of years in the future is calculated to obtain the future average annual requisition quantity of the target spare parts. The calculation formula is as follows.

[0213]

[0214] in, is the planned total demand for Class A target spare parts in year t+i, where t represents the current year, i.e., the last year used as the basis for the forecast. is the planned total demand for Class B of the target spare parts in year t+i, s3 is the sudden demand for the target spare parts every year in the future, and n1 is the specified quantity.

[0215] The calculation methods of planned Class A demand, planned Class B demand, and sudden demand are introduced below. In this embodiment, the default time unit is year, so the "year" in some names can be omitted.

[0216] For planned Class A demand, we can sort out the corresponding maintenance items based on the functional position of the planned Class A demand where the spare parts are located, calculate the annual demand for each maintenance item in the future, and summarize the planned Class A demand for spare parts in the future. The calculation formula is as follows.

[0217]

[0218] Among them, the planned total demand for Class A target spare parts in year t+i is It is equal to the sum of the planned Class A demand of each maintenance project in year t+i, where the planned Class A demand of the jth maintenance project in year t+i is

[0219] For planned Class B spare parts, directly calculating spare parts demand through standard packages does not have a good overall effect. It is necessary to conduct a comprehensive analysis in combination with the historical requisition records involved in the maintenance project. The potential spare parts demand involved in the maintenance project includes two parts: one is the spare parts mounted in the standard package corresponding to the maintenance project, and the other is the spare parts with historical requisition records but not mounted in the standard package. If only the spare parts mounted in the standard package are referred to, the demand for spare parts not mounted in the standard package cannot be output. Therefore, it is necessary to summarize the spare parts mounted in the standard package and the historical requisition information to sort out the maintenance project and the corresponding potential spare parts demand. By sorting out the requisition records of the maintenance project, a maintenance project-spare parts benchmark list is established for different types of maintenance project-spare parts models to call. The benchmark list contains at least the following: maintenance project, mounted standard package number, spare part code, spare part unit price, spare part importance, spare part type in the standard package, spare part mounting quantity in the standard package, total call times, total requisition quantity, total requisition times, and maximum single requisition quantity. The maintenance project + spare part code is the primary key of the list.

[0220] In the related art, the corresponding maintenance items can be sorted out based on the planned Class B demand functional position of the spare parts, the annual demand for each maintenance item in the future can be calculated, and the planned Class B demand for spare parts in the future can be summarized. The calculation formula is as follows.

[0221]

[0222] Among them, the planned total demand for Class B target spare parts in year t+i is It is equal to the sum of the planned Class B demand of each maintenance project in year t+i, where the planned Class B demand of the jth maintenance project in year t+i is b is the planned Class B demand quantity of the target spare part in a maintenance project in a certain year, θ is the planned Class B demand quantity of the target spare part in the standard package of the maintenance project, w2 is the number of times the target spare part is actually used in the maintenance project. For example, if the target spare part is used 4 times in this maintenance project, the number of times the target spare part is used is 4; w3 is the number of calls, that is, the number of times the maintenance project has been carried out in history. For example, if a maintenance project has been carried out 10 times in history, the number of calls is 10; δ is the probability threshold of use. When the number of uses / the number of calls is less than the threshold, or the number of calls of the maintenance project is 0, the planned Class B demand of the target spare part in the maintenance project is the number of demands θ in the standard package. When the number of uses / the number of calls is greater than the threshold, the demand quantity calculated at this time is Among them, w1 is the total usage quantity, that is, the sum of the actual usage quantities of the target spare parts in the maintenance project. For example, the target spare parts were used 4 times in this maintenance project, and a total of 7 were used, then the total usage quantity is 7; λ is the correction coefficient, which is related to the importance and unit price of the target spare parts. For example, for target spare parts with high importance and low unit price, the λ value is set to 1.3; for target spare parts with unimportant and high unit price, the λ value is set to 0.77; for other types of target spare parts, the λ value is set to 1.

[0223] The calculation process of the above planned Class B demand depends on the historical use records. For spare parts with no historical use or large use fluctuations, if formulas 14 and 15 are used directly to calculate the planned Class B demand quantity, there are two potential problems:

[0224] The first problem is that for projects with no record of use, the subsequent predicted spare parts demand quantity is 0; for example, a spare part is installed in functional position 1, which corresponds to maintenance project 1. Since the spare part has not been used in maintenance project 1, based on formula 15, the planned Class B demand quantity of the spare part in maintenance project 1 is 0, and the demand quantity at all subsequent times (t+1 to t+n1) is 0, which is inconsistent with the actual situation and is likely to cause spare parts guarantee problems (inventory shortage). The fact that a spare part installed in a functional position has never failed in the past does not mean that it will not fail in the future. As the operating time of the unit increases, the equipment (usually called equipment when installed on site, and called spare parts when not installed on site) will fail at some point in the future, and there will be a need to replace the spare parts.

[0225] The second problem is that the use of a single spare part is highly random. If only the use data of the spare part itself is used, and the use distribution data of the spare parts of this type is not effectively used, there is a certain probability that the predicted demand quantity value is too large or too small. For example, the annual failure probability of a single functional position of a certain type of spare part (referring to the same type of spare parts produced by the same manufacturer, such as all thermometer spare parts produced by a certain manufacturer) is 5%. There are 20 functional positions installed on site. According to the average use demand of this type of spare parts, there is a need to replace one spare part every year, but a single spare part may have a large use fluctuation. The spare part was used 3 and 5 times in the first two years, respectively. In the future, the spare part will be calculated to have 4 demands every year, which is quite different from the annual average demand of 1 for this type of spare parts. That is, the calculated annual average use is 4, and there is a high probability that the predicted value is too high, which is easy to cause inventory backlog.

[0226] In order to improve the forecast accuracy of planned Class B demand for spare parts with no historical use or large fluctuations in use, the use quantity of planned Class B demand can be corrected by using the functional position quantity distribution of the planned Class B demand for target spare parts and the annual failure probability distribution of spare parts of the same type during the calculation process.

[0227] Specifically, the correction item can be calculated based on the functional position quantity distribution of the planned Class B demand for the target spare parts and the annual failure probability distribution of the same type of spare parts, and the forecast item can be calculated based on the relevant data of the planned Class B demand in the historical requisition data of the target spare parts; the weighted average of the forecast item and the correction item is calculated as the requisition quantity of the planned Class B demand.

[0228] The L1 norm of the Hadamard product of the functional position quantity distribution of the planned Class B demand for the target spare parts and the annual failure probability distribution of the same type of spare parts is calculated as the correction term. The weight of the prediction term is negatively correlated with the weight of the correction term, and the weight of the prediction term is positively correlated with the historical number of times the target spare parts have been used. Optionally, the weight of the prediction term is greater than the weight of the correction term.

[0229] The corrected formula is shown below.

[0230]

[0231] Among them, ω1 and ω2 are non-negative coefficients. Usually, the value of ω2 is less than or equal to the value of ω1. It can be set according to the following principle: the more records of spare parts being used in maintenance projects, and the lower the volatility of use (which can be calculated by dividing the standard deviation of use by the mean of use), the better the value of The value is close to 0; the fewer the records of spare parts being used in maintenance projects, and the greater the volatility of the use (which can be calculated by dividing the standard deviation of the use by the mean of the use), the greater the value of The value is close to 1. (t+i)is the number distribution of functional locations installed and used corresponding to the planned Class B demand for spare parts calculation in the t+i year, U is the annual failure probability distribution of a single functional location of this type of spare parts at different operating times; M(t+i)×U is the matrix M (t+i) The Hadamard product with the matrix U is the multiplication of the corresponding elements in the matrix; ||M(t+i)×U||1 is the L1 norm of the matrix M(t+i)×U, that is, the sum of the absolute values ​​of all elements in the matrix.

[0232] Normally, after nuclear power spare parts are installed on site, their annual failure probability has a certain relationship with the year of installation and use, and the longer the years of installation and use, the greater the annual failure probability. Therefore, the annual failure probability distribution of spare parts of the same type is calculated, and the future demand for the spare parts is predicted based on this information. Spare parts of the same type refer to spare parts with the same attributes produced by the same manufacturer. For example, when conducting demand forecasting for a certain thermometer, the annual failure probability distribution of all thermometers from the same manufacturer is calculated; since the annual failure probability distribution requires a certain amount of historical use data, if the manufacturer's thermometer has too little historical use data in nuclear power plants, it is impossible to construct the failure probability distribution matrix. The annual failure probability distribution can be constructed using the historical use data of thermometers from all manufacturers.

[0233] The calculation of the annual failure probability data of a spare part needs to be based on the spare part's usage data and installation and service life data. For example, a nuclear power plant was officially put into commercial operation in 2012. A spare part was installed and used in 5 functional positions. Its annual usage data is shown in Table 1.

[0234]

[0235] Table 1

[0236] From functional position 1, we can know that there was one withdrawal in 2020, that is, one withdrawal occurred in the 8th year; from functional position 2, we can know that there was one withdrawal in 2021, that is, one withdrawal occurred in the 9th year; from functional position 3, we can know that no withdrawal occurred; from functional position 4, we can know that there was one withdrawal in 2016 and 2022, among which 2016 corresponded to one withdrawal in the 4th year, and 2022 was another withdrawal after the replacement in 2016, so 2 2022 corresponds to the use in the 6th year, that is, there is one use in the 6th year; from functional position 5, it can be obtained that there is one use in 2015, 2017, and 2021, among which 2015 corresponds to the use in the 3rd year, 2017 is used again on the basis of replacement in 2015, so 2017 corresponds to the use in the 2nd year, 2021 is used again on the basis of replacement in 2017, so 2021 corresponds to the use in the 4th year. The use quantity corresponding to the installation time of the spare part and the annual failure probability corresponding to the installation time are shown in Table 2.

[0237]

[0238] Table 2

[0239] The failure probability of multiple spare parts of the same type as the target spare part in different historical years can be calculated; multiple percentiles of the failure probability are counted to obtain the annual failure probability distribution of spare parts of the same type. For example, multiple percentiles can include the 25% digit, the 50% digit and the 75% digit. The 75% digit, the 50% digit and the 25% digit are also known as the upper 75% value, the median value and the lower 25% value respectively. For example, a nuclear power plant uses 100 thermometers of a certain manufacturer. To calculate the annual failure probability of each thermometer spare part, there are 100 data each year, which are sorted from small to large according to the annual failure probability. Assuming that the annual failure probability of the fifth year is 0.1%, 0.2%, 0.3%, ..., 9.9%, 10% respectively, a total of 100 annual failure probability data, then the annual failure probability corresponding to the 75% digit is 7.5%, the annual failure probability corresponding to the 50% digit is 5%, and the annual failure probability corresponding to the 25% digit is 2.5%.

[0240] For example, the maximum failure year failure probability distribution U is as follows

[0241]

[0242] in are the 75% value, 50% value and 25% value of the annual failure probability of the same type of spare parts in the i-th year, respectively, i=1,2,…,a, a represents the longest year of the statistical annual failure probability, and the part exceeding a in the historical data is identified as a. For example, there are 15 years of historical data, and a is set to 10, then the annual failure probability in the 10th year is the average value of the actual annual failure probabilities in the 10th to 15th years.

[0243] The average failure probability of spare parts of the same type can be calculated based on the annual failure probability distribution of spare parts of the same type. The calculation formula is as follows.

[0244]

[0245] The target percentile is determined based on the installation time and replacement quantity of the target spare parts in the functional position of the planned Class B demand and the average failure probability of spare parts of the same type. Specifically, the target spare parts have been installed and used in a certain functional position for x years (only the time of first installation on site is calculated. For example, a spare part was installed on site in 2012, and replaced in 2014 and 2017 respectively. The current time is 2022, and the installation and use time x of the spare part in this functional position is 10 years). Within x years, a total of y spare parts failed (replaced). If y<ω3×x×u (0.5) If the condition is met, the target percentile of the functional position is 25%; if ω3×x×u (0.5) ≤y≤ω4×x×u (0.5) If y>ω4×x×u (0.5) Condition, the target percentile of the functional position is 75%. ω3 and ω4 are coefficients, which can usually be 0.5 and 1.5.

[0246] By sorting out the functional position data of spare parts installed and used on site and calculating the time when spare parts are installed and used in the corresponding functional positions, the functional position quantity distribution M is obtained. (t+i) .

[0247]

[0248] Where ||M (t+i) ||1 is equal to the sum of the quantities of all functional locations.

[0249] For planned Class B demand, when sorting out the functional location data for on-site installation and use of spare parts, only the functional locations to which maintenance projects are attached are selected, and in the maintenance project, the spare parts are planned Class B demand (unplanned Class A demand).

[0250] The elements corresponding to the year and the target percentile in the functional location quantity distribution of the planned Class B demand for the target spare parts are set to 1, and the remaining elements are set to 0.

[0251] Still taking the examples in Table 1 and Table 2 as an example, assume that the spare parts were installed in these five functional positions when the nuclear power plant was in operation in 2012. The current year is 2022, and the installation and use time x of spare parts in all functional positions is 10 years. Let ω3 and ω4 be set to 0.5 and 1.5 respectively, u (0.50) The calculated result is 8%, ω3×x×u (0.5) The value of is 0.4, ω4×x×u (0.5) The value of is 1.2.

[0252] For functional position 1, a total of 1 spare part failed (replaced) in x years, and the y value is 1, satisfying ω3×x×u (0.5) ≤y≤ω4×x×u (0.5) Condition, so the annual failure probability value of 50% is used;

[0253] For functional location 2, a total of 1 spare part failed (replaced) in x years, and the y value is 1, satisfying ω3×x×u (0.5) ≤y≤ω4×x×u (0.5) Condition, so the annual failure probability value of 50% is used;

[0254] For functional position 3, there are 0 spare parts that failed (replaced) in x years, and the y value is 0, satisfying y<ω3×x×u (0.5) Condition, so the annual failure probability value of 25% is used;

[0255] For functional position 4, a total of 2 spare parts failed (replaced) within x years, and the y value is 2, satisfying y>ω4×x×u (0.5) Condition, so the annual failure probability value of 75% is used;

[0256] For functional position 5, a total of 2 spare parts failed (replaced) within x years, and the y value is 2, satisfying y>ω4×x×u (0.5) conditions, so the annual failure probability value of 75% is used.

[0257] When i is 1, the data for 2023 is calculated. At this time, the installation and use time of functional position 1 is 3 years (the time of the last replacement is taken as the initial time); the installation and use time of functional position 2 is 2 years; the installation and use time of functional position 3 is 11 years. Since the matrix U of Formula 17 has only 10 columns, all data greater than 10 years are corrected to 10 years; the installation and use time of functional position 4 is 1 year; the installation and use time of functional position 5 is 2 years.

[0258] Combined with the annual failure probability values ​​of 75% position value, 50% position value, and 25% position value corresponding to function positions 1 to 5, the calculation result is shown in Formula 20. For the data of the 75% value row, the annual failure probability value of 75% position value is used for function positions 4 and 5. Combined with the installation and use time of function position 4 being 1 year; the installation and use time of function position 5 being 2 years, therefore, in the matrix M (t+1) In the first line of and is 1, and the rest are 0. For the data in the 50% value row, the annual failure probability value of the functional position 1 and the functional position 2 uses the 50% value, combined with the installation and use time of the functional position 1 is 3 years; the installation and use time of the functional position 2 is 2 years, so in the matrix M (t+1) In the second line, and is 1, and the rest are 0. For the data in the 25% value row, the annual failure probability value of the functional location 3 uses the 25% value, combined with the installation and use time of the functional location 3 is 10 years, so in the matrix M (t+1) In the third line of is 1, and the rest are 0.

[0259] Similarly, we can get and is 1, and the rest are 0.

[0260]

[0261] When i is 2, the data for 2024 is calculated. At this time, the installation and use time of functional position 1 is 4 years (the time of the last replacement is the initial time); the installation and use time of functional position 2 is 3 years; the installation and use time of functional position 3 is 12 years. Since the matrix U of formula 17 has only 10 columns, all data greater than 10 years are corrected to 10 years; the installation and use time of functional position 4 is 2 years; the installation and use time of functional position 5 is 3 years. Therefore, the matrix M (t+2) The result is shown in formula 21.

[0262]

[0263] When i is 5, the data for 2027 is calculated. At this time, the installation and use time of functional position 1 is 7 years (the time of the last replacement is the initial time); the installation and use time of functional position 2 is 6 years; the installation and use time of functional position 3 is 15 years. Since the matrix U of formula 17 has only 10 columns, all data greater than 10 years are corrected to 10 years; the installation and use time of functional position 4 is 5 years; and the installation and use time of functional position 5 is 6 years. Therefore, the matrix M (t+5) The result is shown in Formula 22.

[0264]

[0265] In the related art, since planned Class B spare parts are spare parts that are selectively replaced based on the status of on-site equipment, they are related to the annual failure probability of spare parts. Therefore, the historical planned Class B spare parts procurement and emergency procurement data are combined to calculate the annual failure probability of spare parts, and combined with the number of functional positions with emergency demand, the future annual emergency demand for spare parts is calculated. The calculation formula is shown below.

[0266]

[0267] in The planned B-type withdrawal quantity for the j-th maintenance item (functional location) in the ti-th year; is the number of emergency demand collections for the jth functional position in the tith year; m2 is the number of maintenance items (functional positions) for spare parts in planned Class B demand; m3 is the number of spare parts in emergency demand functional positions; n2 is the set year value, that is, the number of historical collection data for the past few years is used to calculate the annual average value. For example, if n2 is 5, the historical collection data of the past 5 years is used to calculate the failure probability of spare parts. Generally, for units with a short commissioning time, the n2 value can be set to 5; for units with a long commissioning time, the n2 value can be set to 10.

[0268] Similarly, for spare parts with no historical use or large fluctuations in use, the calculation method of sudden demand in the related art has the same two problems as the calculation method of planned Class B demand. Therefore, the quantity distribution of the functional position of the sudden demand of the target spare parts and the annual failure probability distribution of the same type of spare parts can be used in the calculation process to correct the quantity of sudden demand.

[0269] Specifically, the calculation formula for the revised sudden demand is as follows.

[0270]

[0271] Among them, ω5 and ω6 are non-negative coefficients. Usually, the value of ω6 is less than or equal to the value of ω5. It can be set according to the following principle: the more records of spare parts being used in maintenance projects, and the lower the volatility of use (which can be calculated by dividing the standard deviation of use by the mean of use), the better the value of The value is close to 0; the fewer the records of spare parts being used in maintenance projects, and the greater the volatility of the use (which can be calculated by dividing the standard deviation of the use by the mean of the use), the greater the value of The value is close to 1. (t+i)is the number distribution of functional locations installed and used corresponding to the sudden demand for spare parts calculation in the t+ith year, U is the annual failure probability distribution of a single functional location of this type of spare parts at different operating times; M(t+i)×U is the matrix M (t+i) The Hadamard product with the matrix U is the multiplication of the corresponding elements in the matrix; ||M(t+i)×U||1 is the L1 norm of the matrix M(t+i)×U, that is, the sum of the absolute values ​​of all elements in the matrix.

[0272] In sorting out the functional location data used in the installation of spare parts on site, obtain the functional location quantity distribution M (t+i) During the process, only the functional locations with sudden demands are sorted out, that is, the functional locations that are not attached with maintenance projects.

[0273] Through the implementation of this embodiment, the functional position of the target spare parts in the nuclear power plant is determined, and the functional positions include the functional positions of planned Class A demands, the functional positions of planned Class B demands, and the functional positions of sudden demands; the procurement quantities of the target spare parts required for each type of demand are calculated respectively according to the historical procurement data of the target spare parts, and the functional position quantity distribution of the planned Class B demands and / or sudden demands of the target spare parts and the annual failure probability distribution of spare parts of the same type are used in the calculation process to correct the procurement quantities of the corresponding demands; the procurement quantities of the target spare parts required for each type of demand are summarized to obtain the future average annual procurement quantity of the target spare parts in the nuclear power plant, and the procurement quantities of planned Class B demands and / or sudden demands are corrected by the annual failure probability distribution of spare parts of the same type, thereby reducing the dependence on historical procurement records in the process of predicting planned Class B demands and / or sudden demands, and improving the accuracy of the prediction, especially for spare parts with no historical procurement or large procurement fluctuations.

[0274] Fig.11 A schematic flow chart of a method for predicting the future average annual inventory of spare parts provided in an embodiment of the present application is shown. As an example but not a limitation, the method can be applied to the above-mentioned electronic equipment.

[0275] S31: Building multiple prediction models for target spare parts based on inventory parameters and historical requisition data of target spare parts.

[0276] S32: Use the forecasting model to estimate the average annual inventory of target spare parts in the future.

[0277] The forecasting model includes the purchase application approval model and the purchase cycle model. In addition, it can also include the collection model, reservation model, inventory model, verification model, and trigger purchase application model (also known as the regular order model). For example, a specific example of the spare parts inventory forecasting model is as follows: Fig.12 shown.

[0278] The prediction model can be used to iteratively calculate the inventory quantity of target spare parts in future time units; the future average annual inventory of target spare parts can be calculated based on the inventory quantity of target spare parts. The spare parts inventory prediction model iterates and predicts in set time units, and the time unit can be selected as "day", "week", "month", or "year". Combined with the current granularity of nuclear power plants in the field of inventory management, "month" is usually preferred as the minimum time unit for simulation. The following article uses "month" as the simulation time to introduce in detail the calculation methods of each sub-model in the spare parts inventory prediction model.

[0279] The inventory model can be used to iteratively calculate the inventory quantity. The specific calculation formula is as follows.

[0280] st t+i =st t+i-1 +y t+i-1 -l t+i-1 (25)

[0281] Among them, st is the inventory quantity, y is the acceptance quantity, l is the collection quantity, t+i is the target time unit, and t+i-1 is the previous time unit.

[0282] The acceptance quantity is the actual purchase quantity that expires in the purchase cycle within the target time unit. The acceptance quantity can be calculated using the acceptance model. The specific calculation formula is as follows.

[0283] Y t+ΔT =Y t+ΔT +NPR t (26)

[0284] NPR t The newly added order quantity is the actual purchase quantity. When a new order is added, a corresponding number of spare parts will be accepted after the purchase cycle time. When initializing the acceptance data, the acceptance quantity of all time periods is set to 0. During the simulation process, it is updated accordingly as new orders are added.

[0285] The actual purchase quantity is calculated based on the purchase requisition triggering model and the purchase requisition approval model.

[0286] The trigger purchase requisition quantity is related to the spare parts inventory quantity (St), reserved quantity (R), inventory parameters (Par), and regular order quantity (PR). The inventory parameters are affected by the spare parts type data. The calculation formula for triggering the purchase requisition model is as follows.

[0287]

[0288] in To calculate the function that triggers the purchase requisition quantity, the purchase trigger logic is shown in Table 3. This logic is converted into a function, and the newly triggered purchase quantity is calculated by inputting inventory parameters, inventory quantity, order quantity, and reserved quantity. When the conditions for triggering the purchase requisition are met, The value is the newly triggered purchase quantity, and the total purchase times of spare parts are increased by 1; when the conditions for triggering the purchase application are not met, The value of is 0.

[0289]

[0290] Table 3

[0291] Reservation is the user's request for future use of spare parts in advance. There are three factors that affect reservation: reservation lead time, reservation ratio, and reservation accuracy. Different types of spare parts may have different reservation lead times, reservation ratios, and reservation accuracy. Since the reservation model is relatively complex to set up, in actual applications, you can choose to use or not use the reservation model as needed. When the reservation model is not used, all spare parts will not be reserved in advance. During the simulation process, the reservation value of all spare parts is 0. During the actual approval process, the number of triggered purchase applications may not be equal to the actual number of purchase applications. In order to make the simulation process more in line with the actual purchase application approval process, it is necessary to establish a purchase application approval model.

[0292] Spare parts order quantity refers to the number of spare parts purchase applications that have been approved and are currently in the purchase process. When a new purchase application is approved, the spare parts order quantity increases, and when the purchased spare parts arrive and are accepted, the spare parts order quantity decreases. The calculation formula for spare parts order quantity is shown in formula 29, where PR is the order quantity, NPR is the newly added order quantity, Y is the acceptance quantity, and t is the simulation time. The spare parts order quantity is related to the order quantity of the previous period, the new order quantity of the previous period, and the acceptance quantity of the previous period.

[0293] The distribution of the number of triggered purchase applications and the number of approved purchases of the same type of spare parts (which may be spare parts of the same type produced by the same manufacturer, for example, all thermometer spare parts produced by a certain manufacturer; if the amount of purchase approval data for the thermometers of this manufacturer in nuclear power plants is relatively small, the data of the thermometers of all manufacturers may be used) can be counted to obtain a first ratio distribution matrix; the first target probability corresponding to the first ratio distribution matrix is ​​obtained based on the generated random number and the number of triggered purchase applications for the target spare parts; the actual purchase quantity is calculated based on the first target probability and the number of triggered purchase applications for the target spare parts.

[0294] For example, the distribution of the triggered purchase requisition quantity and the approved purchase quantity of a certain type of spare parts is as follows: Fig.13As shown, Category 1 refers to the proportion of approved purchase quantities of 0, Category 2 refers to the proportion of approved purchase quantities less than the triggered purchase quantity (excluding the case where the approved purchase quantity is 0), Category 3 refers to the proportion of approved purchase quantities equal to the triggered purchase quantity, and Category 4 refers to the proportion of approved purchase quantities greater than the triggered purchase quantity.

[0295] Build the approved purchase quantity based on the relationship between the trigger purchase requisition quantity and the approved purchase quantity With trigger purchase quantity The first ratio distribution matrix is ​​shown below.

[0296]

[0297] Among them is Refers to the proportion of category 1 when the number of triggered purchase applications is k; Refers to the proportion of category 1 + category 2 when the number of triggered purchase applications is k; Refers to the proportion of category 1+category 2+category 3 when the number of triggered purchase requisitions is k.

[0298] by Fig.13 Take the data of as an example, when the number of triggered purchase applications is 1, 29%, is 29% (since the number of purchase applications in this example is an integer, when the number of triggered purchase applications is 1, if the number of approved purchase applications is not 0, there is no situation where the number of approved purchase applications is less than the number of triggered purchase applications, so the proportion of category 2 is 0). is 62%; when the number of triggered purchase applications is 5, is 17%, is 24%, is 75%.

[0299] Based on the purchase application quantity k triggered by this type of spare parts, obtain the corresponding probability value according to the ratio distribution matrix data of formula 28 Generate a random number x between [0,1] random , if satisfied relationship, the calculated approved purchase quantity is 0, and the operation ends; otherwise, determine whether the If the relationship is satisfied, the calculated approved purchase quantity is Otherwise, determine whether If the relationship is satisfied, the calculated approved purchase quantity is k; otherwise, it must satisfy relationship, the calculated approved purchase quantity is

[0300] Example 1: The number of triggered purchase applications is 1, and the corresponding probability value is The random number x generated is 29%, 29%, and 62%. random is 0.89, then the calculated approved purchase quantity is Right now The calculated approved purchase quantity is 2.

[0301] Example 2: The number of triggered purchase applications is 5, and the corresponding probability value is The random number x is generated. random is 0.19, then the calculated approved purchase quantity is Right now The calculated approved purchase quantity is 1.

[0302] The procurement cycle is calculated based on the procurement cycle model. The specific calculation formula is shown below.

[0303] ΔT=ΔT1+ΔT2 (29)

[0304] Among them, ΔT is the purchase cycle value of spare parts, ΔT1 is the time from the purchase application triggering to the purchase application completion approval, and ΔT2 is the time from the purchase application completion approval to the purchase of spare parts into the warehouse. For the ΔT2 time, the historical average value of the spare parts is used. At present, some nuclear power plants conduct hierarchical authorization and approval according to the total amount of the purchase application. That is, the higher the total purchase amount, the more approval nodes and levels, and the longer the approval process required. Therefore, the lower the total purchase application amount, the greater the probability of a short approval time; the higher the total purchase application amount, the greater the probability of a long approval time. Therefore, the ΔT1 time can be calculated based on the triggered purchase application amount.

[0305] The distribution of the triggered purchase application amount and the approval completion time of the same type of spare parts is statistically analyzed to obtain the second ratio distribution matrix; the corresponding second target probability in the second ratio distribution matrix is ​​obtained according to the generated random number and the triggered purchase application amount of the target spare part; the procurement cycle is calculated according to the second target probability and the triggered purchase application amount of the target spare part.

[0306] For example, the relationship between the total purchase requisition amount of a certain type of spare parts (i.e. the number of triggered purchase requisitions multiplied by the spare parts purchase unit price) and the purchase requisition approval time is as follows: Fig.14 The second ratio distribution matrix constructed for this type of spare parts is shown below.

[0307]

[0308] Among them is Refers to the percentage of purchase requisitions with approval time within 1 month when the total amount of the triggered purchase requisition is (0,1000); Refers to the percentage of purchase requisitions with approval time within 2 months when the total amount of the triggered purchase requisition is (0,1000); Refers to the percentage of purchase requisitions with approval time within 1 month when the total amount of the triggered purchase requisition is (1000,5000); Refers to the percentage of purchase requisitions with approval time within 2 months when the total amount of the triggered purchase requisition is (1000,5000); Refers to the percentage of purchase requisitions with approval time within 1 month when the total amount of the triggered purchase requisition is (5000,20000); Refers to the percentage of purchase requisitions with approval time within 2 months when the total amount of the triggered purchase requisition is (5000,20000); Refers to the percentage of purchase requisitions with approval time within 1 month when the total amount of the triggered purchase requisition is (20000,100000]; Refers to the percentage of purchase requisitions with approval time within 2 months when the total amount of the triggered purchase requisition is (20000,100000]; Refers to the percentage of purchase requisitions with approval time within 1 month when the total amount of the triggered purchase requisition is (100000,∞]; Refers to the percentage of purchase requisitions with approval time within 2 months when the total amount of the triggered purchase requisitions is (100000,∞].

[0309] Based on the total purchase amount triggered by this type of spare parts, obtain the corresponding probability value f according to the ratio distribution matrix data of formula 30, and generate a random number x between [0,1]. random , if x is satisfied random ≤f (1) relationship, the calculated purchase application approval time ΔT1 is 1 month, and the operation ends; otherwise, determine whether x is satisfied. random ≤f (2) If the relationship is satisfied, the calculated purchase application approval time ΔT1 is 2 months, and the operation ends; otherwise, the calculated purchase application approval time ΔT1 is 3 months.

[0310] Example 1: The number of triggered purchase applications is 1, and the purchase price is 10,000 yuan. It can be calculated that the total purchase price is 10,000 yuan, which belongs to the interval (5000, 20000). The corresponding probability value is The random number x generated is 94.3% and 97.8%. random is 0.491, because it satisfies The calculated purchase requisition approval time ΔT1 is 1 month.

[0311] Example 2: The number of triggered purchase applications is 5, and the purchase price is 10,000 yuan. The total purchase price can be calculated to be 50,000 yuan, which belongs to the interval (20,000, 100,000]. The corresponding probability value is The random number x generated is 90.4% and 96.9%. random is 0.941, because it satisfies The calculated purchase requisition approval time ΔT1 is 2 months.

[0312] The calculation of the requisition model can refer to the relevant description of the above-mentioned annual average requisition quantity prediction method. Specifically, the requisition quantity of the target spare parts in the future time unit can be calculated according to the functional position of the target spare parts in the nuclear power plant and the historical requisition data. The functional position includes at least one of the functional position of planned Class A demand, the functional position of planned Class B demand and the functional position of sudden demand. The specific calculation formula is as follows.

[0313]

[0314] Where L t is the total quantity taken at time t, is the number of items to be collected corresponding to the planned Class A demand at time t, is the number of items to be collected corresponding to the planned B-type demand at time t, is the number of units to be used for the sudden demand at the tth moment. For the sake of convenience, it is assumed that the minimum time unit of the simulation is a month, and a total of 100 months need to be simulated.

[0315] The time unit in the above-mentioned prediction process is year. When the time unit used by the collection model is smaller than year, for example, month, the prediction of the collection amount needs to be split.

[0316] For planned Class A demand, in formula 12, is the planned total demand for Class A spare parts in year t+i. We only need to change the definition of t+i from "year" to "month", that is, is the total planned Class A demand for the spare part in month t+i. By referring to the above method, the planned Class A demand quantity of the spare part in the specified future month can be calculated, and then the corresponding requisition quantity of the planned Class A demand can be obtained.

[0317] Optionally, during the calculation process, a roulette algorithm is used to calculate a correction item in the collection quantity of the target spare parts for planned Class B demand and / or sudden demand based on the annual failure probability distribution of spare parts of the same type.

[0318] Specifically, the average failure probability of spare parts of the same type can be calculated based on the annual failure probability distribution of spare parts of the same type; the target percentile can be determined based on the installation time and replacement quantity of the target spare parts in the functional locations with planned Class B demand and / or sudden demand and the average failure probability of spare parts of the same type; the elements corresponding to the year and the target percentile in the annual failure probability distribution of spare parts of the same type are used as the probability of obtaining the target spare parts in the corresponding year; for planned Class B demand and / or sudden demand, each obtaining time of each functional location is traversed, and according to the relationship between the random number and the corresponding obtaining probability, it is determined whether to update the correction item according to the minimum replacement quantity of the target spare parts in the functional location.

[0319] For planned Class B demand, in formula 16, is the planned total demand for Class B spare parts in year t+i, which includes the forecast item and corrections Two parts.

[0320] against Just change the definition of t+i from "year" to "month", that is is the total planned Class B demand for the spare part in month t+i. Referring to the above method, the planned Class B demand quantity of the spare part in the specified future month can be calculated, and the calculation result is assigned to the corresponding requisition quantity of the planned Class B demand.

[0321] against Since ||M(t+i)×U||1 has been summarized according to the annual collection quantity, it cannot be directly split into monthly collection quantities. It is necessary to reconstruct the demand quantity of ||M(t+i)×U||1 in the specified future months, as shown below:

[0322] Step 1: Calculate basic data:

[0323] Step 1.1: Calculate the spare parts annual failure probability distribution matrix U, and calculate the average failure probability of spare parts of the same type based on the spare parts annual failure probability distribution matrix U;

[0324] Step 1.2: Sort out the list of planned Class B demand function positions and the spare parts collection time of the function positions (determine the time according to the maintenance project and overhaul start time. For example, if the maintenance project corresponding to a certain function position is performed in the overhaul in January 2025 and July 2026, the spare parts collection time corresponding to the function position is January 2025 and July 2026);

[0325] Step 1.3: Determine the target percentile (75% value, 50% value, or 25% value) based on the installation time and replacement quantity of the target spare parts at the functional position of the planned Class B demand and the average failure probability of spare parts of the same type;

[0326] Step 1.4: For each functional position, record the minimum number of spare parts replaced at that functional position (i.e., the number of spare parts installed at that functional position. For example, if a valve functional position has 1 gasket and 4 fixing nuts installed, the minimum number of gasket spare parts replaced at one time is 1, and the minimum number of fixing nut spare parts replaced at one time is 4. Let the minimum number of replacements be n. min ), spare parts collection time (refer to the content of step 1.2, according to the spare parts collection time corresponding to the functional position and the simulation start time, for example, the current time is January 2024, then the collection time of the spare parts is t+12 and t+30), spare parts collection probability (refer to the above-mentioned annual average collection quantity prediction method, combined with the installation years of the functional position and the quantile of the functional position use to calculate the spare parts collection probability, for example, the collection probability of the spare parts using the 50% value in the functional position, January 2025 corresponds to the fifth year of installation, and July 2026 corresponds to the sixth year of installation, then the collection probabilities of t+12 and t+30 are respectively and );

[0327] Step 1.5: Target Part, change the definition of t+i from "year" to "month", that is, is the total planned Class B demand for the spare part in month t+i. Referring to the above method, calculate the planned Class B demand quantity of the spare part in the specified future month, and assign the calculated result to the corresponding requisition quantity of the planned Class B demand.

[0328] Step 2: For each functional location corresponding to the spare part, calculate as follows:

[0329] Step 2.1: Read the minimum replacement quantity, spare parts collection time, and spare parts collection probability data corresponding to the functional position;

[0330] Step 2.2: For each pick time, apply the roulette algorithm to calculate the number of spare parts picked:

[0331] Step 2.2.1: Let the time of collection be t+i and the probability of collection be u t+i , the number of units to be used is n min ,

[0332] Step 2.2.2: Generate a random number x between [0,1] random , determine whether x is satisfied random ≤u t+i Relationship, if the relationship is satisfied, go to 2.2.3, otherwise go to 2.2.4;

[0333] Step 2.2.3: Update the quantity to be collected implement

[0334] Step 2.2.4: If all the collection times under the functional position have completed the roulette algorithm calculation, go to step 2.3; otherwise, update the collection time data and go to step 2.2.1;

[0335] Step 2.3: If all functional location data have been calculated, go to step 3; otherwise, update the functional location and go to step 2.1;

[0336] Step 3: Output the calculation results:

[0337] Step 3.1: Output the corresponding quantity of planned B-type demand

[0338] To further clarify the calculation process, an example is given below. A spare part has planned Class B demand only in two functional positions. The minimum single replacement quantity of the two functional positions is 2 and 4 respectively. The collection time is [t+12, t+30], [t+30, t+48, t+66], and the collection probability is [4.1%, 4.6%], [5.5%, 8.5%, 9.6%]. The number of use for partial calculations is 0, and the coefficient ω1 and coefficient ω2 are both set to 0.5. Calculate the data of the first functional position and the second functional position in turn. For the time t+12 of the first functional position, the random number x random is 0.723. Since 0.723 is greater than 0.041, x is not satisfied. random ≤u t+i Relationship, no need to update For the first functional position at time t+30, the random number x random is 0.043. Since 0.043 is less than 0.046, x random ≤u t+i Relationship, Execution After Update for For the t+30 time of the second functional position, the random number x random is 0.051. Since 0.051 is less than 0.055, x random ≤u t+i Relationship, Execution After Update for For the t+48 time of the second functional position, the random number x random is 0.635. Since 0.635 is greater than 0.085, x is not satisfied. random ≤u t+i Relationship, no need to update For the t+66 time of the second functional position, the random number x random is 0.079. Since 0.079 is less than 0.096, x random ≤u t+i Relationship, Execution After Update for Therefore, in the planned Class B demand for this spare part, the quantity to be collected at t+30 is 3, the quantity to be collected at t+66 is 2, and the quantity to be collected at other times is 0.

[0339] For sudden demand, the prediction term in formula 24 and corrections Both are annual collection quantities and cannot be directly split into monthly collection quantities. It is necessary to reconstruct the two formulas to calculate the demand in the specified future months.

[0340] against It represents the prediction of future spare part demand based on the historical probability data of spare part use. It represents the average annual demand for spare parts based only on the spare part's usage data. It can be combined with the historical average annual number of spare parts to predict the sudden demand for spare parts every month in the future. For example, the average annual usage of spare part j is The historical average annual number of spare parts is k (the average annual number of spare parts is calculated by the number of sudden demand for spare parts in history). First, the k value is rounded to the nearest integer. When the k value is 0, it is corrected to 1. When the k value is greater than 12, it is corrected to 12. Based on the simulation year, k-1 random numbers with a value range of (0,1) are generated each year. The random numbers are sorted from small to large, and their values ​​are r1, r2, r3, ..., rk. -1 , and the monthly collection quantity is calculated accordingly, which are Among them, round() is a rounding function. The monthly distribution relationship of the annual number of withdrawals is as follows: Fig.15 As shown in the figure, when the annual collection frequency k is 1, one month is randomly selected between January and December as the collection month of the year; when the annual collection frequency k is 2, one month is randomly selected between January and June, and one month is randomly selected between July and December as the collection month of the year. Based on the historical annual average number k and Fig.15 The annual usage frequency is distributed in the monthly chart, and the future monthly sudden demand for spare parts can be calculated.

[0341] For example, the average annual number of spare parts s3 is 10, the average annual number of times k is 4, and the simulation time is 2 years (24 months). Based on the simulation year, k-1 random numbers with a value range of (0,1) are generated each year. After 2 years of simulation, a total of 6 random numbers are generated. Assuming that the random numbers are 0.83, 0.19, 0.61, 0.58, 0.73, and 0.26 respectively, the random numbers of each year are sorted from small to large, which are 0.19, 0.61, 0.83, 0.26, 0.58, and 0.73, respectively. The single monthly The quantities to be used are 2(round(10*0.19)), 4(round(10*(0.61-0.19))), 2(round(10*(0.83-0.61))), 2(round(10*(1-0.83))), 3(round(10*0.26)), 3(round(10*(0.58-0.26))), 1(round(10*(0.73-0.58))), 3(round(10*(1-0.73))). Based on Fig.15 Distribution relationship, since the annual number of times k is 4, one month is randomly selected from January to March, April to June, July to September, and October to December to assign the data. Assuming that the randomly assigned months in the first year are January, April, July, and November, and the randomly assigned months in the second year are February, June, August, and November, it is converted into the form of t+i months. The data was collected in the month in formula 32, and the collection quantity in the remaining months is 0.

[0342] I3=[1,6,9,11,14,18,20,23] T (32)

[0343] Similarly, referring to Formula 32, the quantity taken each time is summarized as shown in Formula 33.

[0344]

[0345] For example, the second item in formula 32 is 6, which represents the sixth month. The number of parts to be used is the second item in formula 33, which is 4, which means that 4 spare parts were used in the sixth month. Similarly, when the simulation time is set to 100 months, the demand time of formula 32 and the demand quantity of formula 33 are calculated by referring to the above content. Based on the contents of formula 32 and formula 33, the calculation results are assigned to the number of parts to be used for the sudden demand. The quantity of formula 33 needs to be used After correction, assign it to the quantity to be used From this, we can get The corresponding monthly use data. If the coefficient ω1 and the coefficient ω2 are both set to 0.5, based on the results of formula 32 and formula 33, the calculated for

[0346] against It refers to the historical usage probability data of the spare parts type, combined with the usage data of the spare parts themselves, to predict the future usage demand of the spare parts. The method is as follows:

[0347] Step 1: Calculate basic data:

[0348] Step 1.1: Reference Fig.10 The content of the embodiment shown is to calculate the spare parts annual failure probability distribution matrix U;

[0349] Step 1.2: Sort out the list of function locations with sudden demands;

[0350] Step 1.3: Reference Fig.10 The content of the embodiment shown is to calculate the percentile value (75% percentile value, 50% percentile value, or 25% percentile value) of the use of the functional position based on the data that the spare part has been installed and used at the functional position for x years and a total of y spare parts have failed (replaced) within x years;

[0351] Step 1.4: For each functional position, record the minimum number of spare parts replaced at that functional position (i.e., the number of spare parts installed at that functional position. For example, if a valve functional position has 1 gasket and 4 fixing nuts installed, the minimum number of gasket spare parts replaced at one time is 1, and the minimum number of fixing nut spare parts replaced at one time is 4. Let the minimum number of replacements be n. min ), spare parts collection time (sudden demand may occur in all months of simulation, that is, the collection time is equal to the simulation time), spare parts collection probability (refer to Fig.10 In the embodiment shown, the installation years of the functional position are combined with the quantiles used in the functional position to calculate the probability of spare parts use). For example, the probability of spare parts use 50% of the quantiles in the functional position. The current time is January 2024, and the spare parts were last installed on site in January 2020. Then February 2024 to January 2025 corresponds to the fifth year of installation, and February 2025 to January 2026 corresponds to the sixth year of installation. The probability from t+1 to t+12 is (The annual failure probability has been converted into a monthly use probability), the probability from t+13 to t+24 is And so on;

[0352] Step 1.5: Target Part, refer to the calculation process in the previous article, calculate the number of sudden demands for the spare parts in the specified future months, and assign the calculation result to the number of requisitions corresponding to the sudden demand

[0353] Step 2: For each functional location corresponding to the spare part, calculate as follows:

[0354] Step 2.1: Read the minimum replacement quantity, spare parts collection time, and spare parts collection probability data corresponding to the functional position;

[0355] Step 2.2: For each pick time, apply the roulette algorithm to calculate the number of spare parts picked:

[0356] Step 2.2.1: Let the time of collection be t+i and the probability of collection be u t+i , the number of units to be used is n min ,

[0357] Step 2.2.2: Generate a random number x between [0,1] random , determine whether x is satisfied random ≤u t+i Relationship, if the relationship is satisfied, go to 2.2.3, otherwise go to 2.2.4;

[0358] Step 2.2.3: Update the quantity to be collected implement

[0359] Step 2.2.4: If all the collection times under the functional position have completed the roulette algorithm calculation, go to step 2.3; otherwise, update the collection time data and go to step 2.2.1;

[0360] Step 2.3: If all functional location data have been calculated, go to step 3; otherwise, update the functional location and go to step 2.1;

[0361] Step 3: Output the calculation results:

[0362] Step 3.1: Output the quantity corresponding to the sudden demand

[0363] Starting from the current time unit, the forecast model can be used to iteratively calculate the inventory quantity of the target spare parts in the future time units until the set number of time units is reached. Then the inventory quantity of the target spare parts in the time unit is converted into the future average annual inventory of the target spare parts.

[0364] Through the implementation of this embodiment, multiple prediction models of target spare parts are constructed based on the inventory parameters and historical requisition data of the target spare parts, and the prediction models include a purchase application approval model and a purchase cycle model; the prediction model is used to estimate the future average annual inventory of the target spare parts. By constructing a prediction model, a relationship between inventory parameters and the future average annual inventory quantity of spare parts is established to improve the accuracy of the prediction, and the purchase application approval model and the purchase cycle model are used to further improve the prediction accuracy of the purchase quantity, thereby improving the accuracy of the average annual inventory quantity.

[0365] Fig.16 A schematic flow chart of a method for predicting the average annual usage of spare parts in the future provided by another embodiment of the present application is shown. As an example but not a limitation, the method can be applied to the above-mentioned electronic equipment.

[0366] S41: construct multiple prediction models for multiple spare parts based on their inventory parameters and historical requisition data, and use the prediction models to estimate the future average annual inventory of the spare parts.

[0367] The forecast model is used to iteratively calculate the inventory quantity of spare parts in future time units. The forecast model includes the purchase application approval model and the purchase cycle model. Then, the future annual average inventory of spare parts is calculated based on the inventory quantity of spare parts. For detailed calculation process, please refer to Fig.11 The detailed description of the illustrated embodiments will not be repeated here.

[0368] S42: Use multiple spare parts and their inventory parameters, historical usage data and future annual average inventory to build a sample set, and use the sample set to train the inventory prediction network.

[0369] The above prediction model requires iterative simulation of future spare parts usage data, which requires a large amount of calculation, and the computing power required is positively correlated with the prediction time. If this method is used to predict a large number of spare parts, it will take a lot of time and it will be difficult to perform online calculations.

[0370] To solve this problem, an artificial neural network model can be constructed. First, the iterative simulation method is used to offline calculate the corresponding annual average usage of spare parts under different types of inventory parameters, and then the data is organized into training set data. The training set data is used to train the neural network model. After the training is completed, the neural network model is directly used to predict the annual average inventory quantity of spare parts to improve computing efficiency.

[0371] Sample sets and training inventory prediction networks can be constructed for spare parts with different types of data. Different inventory prediction networks have different inputs, such as Fig.17 shown.

[0372] The input of the inventory forecasting network for spare parts of type data PD+EX includes historical requisition data. An example of the network is Fig.18 The input of the inventory forecasting network for spare parts with type data VB+EX and ZB+EX includes historical requisition data and reorder point value. An example of the network is shown in Fig.19 The input of the inventory forecasting network for spare parts with type data of VB+HB and ZB+HB includes historical requisition data, reorder point value and maximum inventory value. An example of the network is shown in Fig. 20 The input of the inventory forecasting network for spare parts with type data VB+FX and ZB+FX includes historical requisition data, reorder point value and fixed batch value. An example of the network is shown in Fig.21 shown.

[0373] S43: Input the inventory parameters and historical requisition data of the target spare parts into the corresponding inventory quantity prediction network to obtain the future annual average inventory quantity of the target spare parts.

[0374] In some embodiments, this step may be omitted or performed by other execution entities.

[0375] Through the implementation of this embodiment, multiple prediction models for spare parts are constructed for multiple spare parts based on their inventory parameters and historical requisition data, and the prediction models are used to estimate the future average annual inventory of spare parts. The prediction models include a purchase application approval model and a purchase cycle model; a sample set is constructed using multiple spare parts and their inventory parameters, historical requisition data, and future average annual inventory, and the sample set is used to train the inventory prediction network. By constructing a prediction model, a relationship between inventory parameters and the future average annual inventory quantity of spare parts is established to improve the accuracy of the prediction, and the purchase application approval model and the purchase cycle model are used to further improve the prediction accuracy of the purchase quantity, thereby improving the accuracy of the average annual inventory quantity. In addition, the neural network model is used to predict the average annual inventory quantity of spare parts in advance to improve computing efficiency and achieve online computing.

[0376] Fig. 22 A schematic structural diagram of a spare parts inventory parameter allocation device provided in an embodiment of the present application is shown. The spare parts inventory parameter allocation device includes an acquisition module 11 and an allocation module 12.

[0377] The acquisition module 11 is used to acquire the group-plant-level inventory parameters of the target spare parts and determine the allocation principle of the target spare parts, where the allocation principle includes at least one of minimizing total cost, unifying benefits, and minimizing allocation.

[0378] The allocation module 12 is used to allocate the group plant level inventory parameters of the target spare parts to the nuclear power plants according to the allocation principle to obtain the power plant level inventory parameters of the target spare parts. The group plant level inventory parameters include at least one of the group plant level reorder point value, maximum inventory value and fixed purchase batch value.

[0379] Fig.23A schematic diagram of the structure of a device for predicting the average annual usage of spare parts in the future provided by an embodiment of the present application is shown. The spare parts inventory parameter allocation device includes a determination module 21, a calculation module 22 and a summary module 23.

[0380] The determination module 21 is used to determine the functional position of the target spare parts in the nuclear power plant, and the functional position includes the functional position of the planned Class A demand, the functional position of the planned Class B demand and the functional position of the sudden demand.

[0381] The calculation module 22 is used to calculate the target spare parts collection quantity required for each type of demand based on the historical collection data of the target spare parts. During the calculation process, the collection quantity of the corresponding demand is corrected using the functional position quantity distribution of the planned Class B demand and / or sudden demand for the target spare parts and the annual failure probability distribution of the same type of spare parts.

[0382] The summarizing module 23 is used to summarize the number of target spare parts required for various needs to obtain the average annual number of target spare parts used in the nuclear power plant in the future.

[0383] Fig.24 A schematic diagram of the structure of a device for predicting the future average annual inventory of spare parts provided in an embodiment of the present application is shown. The device for allocating spare parts inventory parameters includes a construction module 31 and an estimation module 32.

[0384] The construction module 31 is used to construct multiple prediction models of the target spare parts based on the inventory parameters and historical requisition data of the target spare parts. The prediction models include a purchase application approval model and a purchase cycle model.

[0385] The estimation module 32 is used to estimate the target spare parts' annual average inventory and the target spare parts' use quantity based on the forecast model to obtain the target spare parts' annual average use quantity in the nuclear power plant.

[0386] Fig.25 A schematic diagram of the structure of a device for predicting the future average annual inventory of spare parts provided in another embodiment of the present application is shown. The device for allocating spare parts inventory parameters includes a prediction module 41 and a training module 42.

[0387] The prediction module 41 is used to construct multiple prediction models for multiple spare parts based on their inventory parameters and historical requisition data, and use the prediction models to estimate the future average annual inventory of spare parts. The prediction models include a purchase application approval model and a purchase cycle model.

[0388] The training module 42 is used to construct a sample set using a plurality of spare parts and their inventory parameters, historical usage data and future annual average inventory, and use the sample set to train an inventory prediction network.

[0389] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0390] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0391] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0392] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the camera / electronic device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0393] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0394] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0395] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0396] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0397] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for predicting the average annual usage of spare parts in the future, characterized in that: The method comprises: Determine the functional position of the target spare parts in the nuclear power plant, wherein the functional position includes the functional position of the planned Class A demand, the functional position of the planned Class B demand, and the functional position of the sudden demand; The number of target spare parts required for each type of demand is calculated according to the historical use data of the target spare parts. During the calculation process, the number of functional positions of the planned Class B demand and / or the sudden demand of the target spare parts and the annual failure probability distribution of spare parts of the same type are used to correct the number of target spare parts required for the corresponding demand. The number of target spare parts required for each type of demand is summarized to obtain the average annual number of target spare parts to be used in the nuclear power plant in the future.

2. The method according to claim 1, characterized in that The correction of the number of functional positions of the planned Class B demand and / or the sudden demand for the target spare parts and the annual failure probability distribution of spare parts of the same type to the corresponding demand includes: Calculate the correction item based on the functional position quantity distribution of the planned Class B demand and / or the sudden demand of the target spare parts and the annual failure probability distribution of spare parts of the same type, and calculate the forecast item based on the relevant data of the planned Class B demand and / or the sudden demand in the historical use data of the target spare parts; A weighted average of the predicted item and the revised item is calculated as the collection quantity of the corresponding demand.

3. The method according to claim 2, characterized in that The calculation of the correction item based on the functional position quantity distribution of the planned Class B demand and / or the sudden demand of the target spare parts and the annual failure probability distribution of spare parts of the same type includes: The L1 norm of the Hadamard product of the functional position quantity distribution of the planned Class B demand and / or the sudden demand for the target spare parts and the annual failure probability distribution of spare parts of the same type is calculated as the correction term.

4. The method according to claim 2, characterized in that The weight of the prediction item is negatively correlated with the weight of the correction item, and the weight of the prediction item is positively correlated with the historical number of times the target spare part has been used.

5. The method according to claim 2, characterized in that The weight of the prediction term is greater than the weight of the correction term.

6. The method according to claim 1, characterized in that Before the calculation process uses the planned Class B demand and / or the sudden demand of the target spare parts, and the annual failure probability distribution of spare parts of the same type to correct the corresponding demand, the method further includes: Calculate the failure probability of the same type of spare parts in different historical years; Multiple percentiles of the failure probability are counted to obtain the annual failure probability distribution of the same type of spare parts.

7. The method according to claim 6, characterized in that Before the calculation process uses the planned Class B demand and / or the sudden demand of the target spare parts, and the annual failure probability distribution of spare parts of the same type to correct the corresponding demand, the method further includes: Calculate the average failure probability of spare parts of the same type according to the annual failure probability distribution of spare parts of the same type; Determine the target percentile according to the installation time and replacement quantity of the target spare parts at the functional location of the planned Class B demand and / or the sudden demand and the average failure probability of the same type of spare parts; The elements corresponding to the year and the target percentile in the functional location quantity distribution of the planned Class B demand and / or the sudden demand for the target spare parts are set to 1, and the remaining elements are set to 0.

8. The method according to claim 6, characterized in that The plurality of percentiles include the 25th percentile, the 50th percentile, and the 75th percentile.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images