Nuclear power plant spare parts demand forecasting method, system, computer equipment and storage medium

Through nuclear power plant maintenance plans and preset constraints, a target demand forecast model is established and a planned spare parts demand list is generated, which solves the problem that traditional spare parts preparation methods cannot accurately predict spare parts demand during nuclear power plant maintenance, ensures sufficient spare parts and guarantees Stable operation of nuclear power plants.

CN113869602BActive Publication Date: 2024-12-03CHINA GENERAL NUCLEAR POWER OPERATION +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111224764.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-10-21
Publication Date
2024-12-03
Estimated Expiration
2041-10-21

AI Technical Summary

Technical Problem

Traditional nuclear power plant spare parts preparation methods rely on historical experience and maintenance records, and cannot accurately predict faults and spare parts requirements during actual operation, resulting in the shortage of spare parts during maintenance.

Method used

Determine maintenance projects through the nuclear power plant maintenance plan, carry out project equivalent processing, obtain change rate forms and preset constraints, establish a benchmark list, determine the target demand forecast model, and generate a planned spare parts demand list to ensure sufficient spare parts.

Benefits of technology

Accurate prediction of spare parts demand during nuclear power plant maintenance has been achieved, shortage of spare parts is avoided, and the stable operation of nuclear power plant is ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113869602B_ABST
    Figure CN113869602B_ABST
Patent Text Reader

Abstract

The present application relates to a method, system, computer equipment and storage medium for predicting spare parts demand for nuclear power plants. The method comprises: performing project equivalence processing based on maintenance projects, generating a preliminary maintenance project list, and screening maintenance projects from the preliminary maintenance project list according to a maintenance project change rate form. Determine a first-category spare parts demand list based on preset constraints and maintenance standard package data. Integrate the requisition data based on the requisition records, and establish a first benchmark list of maintenance projects and spare parts, and a second benchmark list of standard packages and spare parts. Determine a target demand forecasting model based on the maintenance project requisition data, the standard package requisition data and the spare parts data. Generate a second-category spare parts demand list based on the target demand forecasting model. Generate a planned spare parts demand quantity list based on the first-category spare parts demand list and the second-category spare parts demand list. This method can accurately determine the spare parts demand of different categories during the maintenance process of a nuclear power plant.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, system, computer equipment and storage medium for predicting demand for spare parts of a nuclear power plant. Background Art

[0002] With the development of computer technology and the gradual maturity of the production method of using nuclear power plants for power generation, the requirements for the stable operation of nuclear power plants are increasing to ensure the safety of power generation and use. In order to maintain the stable operation of nuclear power plants, the importance of maintenance work in nuclear power plants is self-evident.

[0003] Traditionally, nuclear power plants are inspected and repaired regularly, and different parts that need to be used are prepared in time to avoid the situation where spare parts are lacking and repairs cannot be carried out in time when a failure occurs. Specifically, it is necessary to prepare different types of spare parts required during the maintenance of nuclear power plants in advance, such as spare parts required for historical maintenance experience or routine maintenance, in advance to reduce the shortage of spare parts during the maintenance process.

[0004] However, the traditional spare parts preparation methods mostly rely on the maintenance experience of the staff, the historical maintenance records of the nuclear power plant, and the conventional maintenance spare parts, etc. However, since the actual operating environment faced by the nuclear power plant during actual operation is not consistent, the possible faults or problems cannot be accurately determined based on historical records or accumulated experience, and there are still problems such as insufficient spare parts preparation and inability to meet spare parts needs during maintenance. Summary of the invention

[0005] Based on this, it is necessary to provide a nuclear power plant spare parts demand prediction method, system, computer equipment and storage medium that can accurately determine the spare parts demand of nuclear power plants during the maintenance process in order to prepare sufficient maintenance spare parts in advance in response to the above technical problems.

[0006] A method for predicting demand for spare parts of a nuclear power plant, the method comprising:

[0007] Determine the corresponding maintenance items according to the maintenance plan of the nuclear power plant, perform project equivalence processing based on the maintenance items, and generate a corresponding preliminary maintenance item list;

[0008] Acquire a maintenance item change rate form corresponding to the preliminary maintenance item list, and select maintenance items from the preliminary maintenance item list according to the maintenance item change rate form;

[0009] Acquire preset constraints corresponding to the maintenance project, and determine a first category spare parts requirement list corresponding to the maintenance project according to the preset constraints and maintenance standard package data;

[0010] Acquire the collection records corresponding to the preliminary maintenance item list, and integrate the collection data based on the collection records to obtain updated collection record data;

[0011] Establishing a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts according to the updated requisition record data;

[0012] Obtaining maintenance item requisition data corresponding to the first benchmark list and standard package requisition data corresponding to the second benchmark list, and determining a target demand forecasting model corresponding to the maintenance item based on the maintenance item requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list;

[0013] Based on the target demand forecasting model, forecast the demand for second-category spare parts for the maintenance project, and generate a corresponding demand list for second-category spare parts;

[0014] Based on the first category spare parts requirement list and the second category spare parts requirement list, a planned spare parts requirement quantity list is summarized and generated.

[0015] A nuclear power plant spare parts demand forecasting system, the system comprising:

[0016] A preliminary maintenance item list generation module is used to determine corresponding maintenance items according to the maintenance plan of the nuclear power plant, and perform item equivalence processing based on the maintenance items to generate a corresponding preliminary maintenance item list;

[0017] A maintenance item screening module, used to obtain a maintenance item change rate form corresponding to the preliminary maintenance item list, and screen maintenance items from the preliminary maintenance item list according to the maintenance item change rate form;

[0018] A first-category spare parts requirement list determination module, used to obtain preset constraints corresponding to the maintenance project, and determine the first-category spare parts requirement list corresponding to the maintenance project according to the preset constraints and the maintenance standard package data;

[0019] A collection record data updating module, used to obtain the collection record corresponding to the preliminary maintenance item list, and integrate the collection data based on the collection record to obtain updated collection record data;

[0020] A reference list establishment module, used to establish a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts according to the updated requisition record data;

[0021] a target demand forecast model determination module, configured to obtain maintenance item requisition data corresponding to the first benchmark list and standard package requisition data corresponding to the second benchmark list, and determine a target demand forecast model corresponding to the maintenance item based on the maintenance item requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list;

[0022] A second-category spare parts demand list generation module, configured to perform a second-category spare parts demand forecast for the maintenance project based on the target demand forecast model, and generate a corresponding second-category spare parts demand list;

[0023] The module for generating the planned spare parts requirement quantity list is used to summarize and generate the planned spare parts requirement quantity list based on the first category spare parts requirement list and the second category spare parts requirement list.

[0024] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0025] Determine the corresponding maintenance items according to the maintenance plan of the nuclear power plant, perform project equivalence processing based on the maintenance items, and generate a corresponding preliminary maintenance item list;

[0026] Acquire a maintenance item change rate form corresponding to the preliminary maintenance item list, and select maintenance items from the preliminary maintenance item list according to the maintenance item change rate form;

[0027] Acquire preset constraints corresponding to the maintenance project, and determine a first category spare parts requirement list corresponding to the maintenance project according to the preset constraints and maintenance standard package data;

[0028] Acquire the collection records corresponding to the preliminary maintenance item list, and integrate the collection data based on the collection records to obtain updated collection record data;

[0029] Establishing a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts according to the updated requisition record data;

[0030] Obtaining maintenance item requisition data corresponding to the first benchmark list and standard package requisition data corresponding to the second benchmark list, and determining a target demand forecasting model corresponding to the maintenance item based on the maintenance item requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list;

[0031] Based on the target demand forecasting model, forecast the demand for second-category spare parts for the maintenance project, and generate a corresponding demand list for second-category spare parts;

[0032] Based on the first category spare parts requirement list and the second category spare parts requirement list, a planned spare parts requirement quantity list is summarized and generated.

[0033] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0034] Determine the corresponding maintenance items according to the maintenance plan of the nuclear power plant, perform project equivalence processing based on the maintenance items, and generate a corresponding preliminary maintenance item list;

[0035] Acquire a maintenance item change rate form corresponding to the preliminary maintenance item list, and select maintenance items from the preliminary maintenance item list according to the maintenance item change rate form;

[0036] Acquire preset constraints corresponding to the maintenance project, and determine a first category spare parts requirement list corresponding to the maintenance project according to the preset constraints and maintenance standard package data;

[0037] Acquire the collection records corresponding to the preliminary maintenance item list, and integrate the collection data based on the collection records to obtain updated collection record data;

[0038] Establishing a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts according to the updated requisition record data;

[0039] Obtaining maintenance item requisition data corresponding to the first benchmark list and standard package requisition data corresponding to the second benchmark list, and determining a target demand forecasting model corresponding to the maintenance item based on the maintenance item requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list;

[0040] Based on the target demand forecasting model, forecast the demand for second-category spare parts for the maintenance project, and generate a corresponding demand list for second-category spare parts;

[0041] Based on the first category spare parts requirement list and the second category spare parts requirement list, a planned spare parts requirement quantity list is summarized and generated.

[0042] In the above-mentioned nuclear power plant spare parts demand prediction method, system, computer equipment and storage medium, the corresponding maintenance items are determined according to the nuclear power plant maintenance plan, and the project equivalent processing is performed based on the maintenance items to generate the corresponding preliminary maintenance item list, and the maintenance items are screened from the preliminary maintenance item list by obtaining the maintenance item change rate form corresponding to the preliminary maintenance item list, and according to the maintenance item change rate form, the maintenance items are screened from the preliminary maintenance item list. Then, by obtaining the preset constraints corresponding to the maintenance items, and according to the preset constraints and the maintenance standard package data, the first category spare parts demand list corresponding to the maintenance items is determined. By obtaining the requisition records corresponding to the preliminary maintenance item list, and integrating the requisition data based on the requisition records, the updated requisition record data is obtained, and then according to the updated requisition record data, the first benchmark list between the maintenance items and the spare parts, and the second benchmark list between the standard package and the spare parts are established. By obtaining the maintenance project collection data corresponding to the first benchmark list and the standard package collection data corresponding to the second benchmark list, and according to the maintenance project collection data, the standard package collection data and the spare parts data corresponding to the spare parts list, the target demand forecasting model corresponding to the maintenance project is determined, and based on the target demand forecasting model, the second category spare parts demand forecast is performed for the maintenance project to generate the corresponding second category spare parts demand list. Then, according to the first category spare parts demand list and the second category spare parts demand list, a planned spare parts demand quantity list is summarized and generated. It realizes the accurate determination of the different categories of spare parts demand of the nuclear power plant during the maintenance process, and can be purchased in advance according to the obtained planned spare parts demand quantity list to avoid spare parts shortage during the maintenance process, so as to prepare sufficient maintenance spare parts in advance to meet maintenance needs and further ensure the stable operation of the nuclear power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 An application environment diagram of a method for predicting demand for spare parts in a nuclear power plant in one embodiment;

[0044] Figure 2 A schematic diagram of a flow chart of a method for predicting spare parts demand for a nuclear power plant in one embodiment;

[0045] Figure 3 A schematic diagram of classification of planned spare parts for a nuclear power plant in one embodiment;

[0046] Figure 4 A schematic diagram of the change rate of each professional outline item of a nuclear power plant in an embodiment;

[0047] Figure 5 A schematic diagram of sorting out the first category spare parts requirement list in an embodiment;

[0048] Figure 6 A schematic diagram of setting constraints corresponding to the first category spare parts requirement list in an embodiment;

[0049] Figure 7 A schematic diagram of updating the requisition record data of a nuclear power plant spare parts demand forecasting method in one embodiment;

[0050] Figure 8 A schematic diagram of potential spare parts demand relationships involved in a maintenance project in one embodiment;

[0051] Fig. 9 A schematic diagram of the relationship between the maintenance items corresponding to the standard package and the historical use records in an embodiment;

[0052] Fig.10 A schematic diagram of the relationship between the standard package corresponding to the spare parts and the historical requisition records in an embodiment;

[0053] Fig.11 A schematic diagram of a model selection process corresponding to the second category spare parts requirement list in an embodiment;

[0054] Fig.12 A schematic diagram of a framework for sorting out the required quantities corresponding to the second category spare parts requirement list in an embodiment;

[0055] Fig.13 A schematic diagram of a process for determining a target demand forecasting model corresponding to a maintenance project in one embodiment;

[0056] Fig.14 A schematic diagram of a framework of a first type of maintenance item and a spare parts model in an embodiment;

[0057] Fig.15 A schematic diagram of a framework of a first type of standard package and spare parts model in an embodiment;

[0058] Fig.16 A schematic diagram of a framework of a first type of spare parts model in an embodiment;

[0059] Fig.17 A schematic diagram of a process for determining a target demand forecasting model corresponding to a maintenance project in another embodiment;

[0060] Fig.18 It is a schematic diagram of the framework of the second type of maintenance items and spare parts model in one embodiment;

[0061] Fig.19 A schematic diagram of the structure of a maintenance project and a spare parts model library in one embodiment;

[0062] Fig. 20 A schematic diagram of a framework of a second type of standard package and spare parts model in one embodiment;

[0063] Fig.21 A schematic diagram of a framework of a standard package and a spare parts model library in one embodiment;

[0064] Fig. 22 A schematic diagram of a framework of a second type of spare parts model in an embodiment;

[0065] Fig.23 A schematic diagram of a spare parts model library in one embodiment;

[0066] Fig.24 A schematic diagram of a neural network spare parts demand prediction process in one embodiment;

[0067] Fig.25 A schematic flow chart of a method for predicting spare parts demand for a nuclear power plant in another embodiment;

[0068] Fig.26 A schematic diagram of a framework of a purchase prediction data model in one embodiment;

[0069] Fig. 27 A schematic diagram of a flow chart of a method for predicting spare parts demand for a nuclear power plant in yet another embodiment;

[0070] Fig.28 A schematic diagram of comparing actual on-site spare parts requisition data and a planned spare parts requirement quantity list in one embodiment;

[0071] Fig.29 A schematic diagram of a grid distribution of a prediction effect in one embodiment;

[0072] Fig.30 A structural block diagram of a nuclear power plant spare parts demand forecasting system in one embodiment;

[0073] Fig.31 It is a structural block diagram of a nuclear power plant spare parts demand prediction system in another embodiment;

[0074] Fig.32 A schematic diagram of a query interface of a nuclear power plant spare parts demand forecasting system in one embodiment;

[0075] Fig.33 A schematic diagram of a spare parts historical use data analysis interface of a nuclear power plant spare parts demand forecasting system in one embodiment;

[0076] Fig.34 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0077] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0078] The nuclear power plant spare parts demand forecasting method provided in this application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The server 104 determines the corresponding maintenance items according to the maintenance plan of the nuclear power plant, and performs project equivalence processing based on the maintenance items to generate the corresponding preliminary maintenance item list, and obtains the maintenance item change rate form corresponding to the preliminary maintenance item list, and screens out the maintenance items from the preliminary maintenance item list according to the maintenance item change rate form. The server 104 further obtains the preset constraints corresponding to the maintenance items, and determines the first category spare parts demand list corresponding to the maintenance items based on the preset constraints and the maintenance standard package data, and obtains the corresponding use records of the preliminary maintenance items list, and integrates the use data based on the use records to obtain the updated use record data, and then can establish the first benchmark list between the maintenance items and the spare parts, and the second benchmark list between the standard package and the spare parts according to the updated use record data. By obtaining the maintenance project collection data corresponding to the first benchmark list and the standard package collection data corresponding to the second benchmark list, and according to the maintenance project collection data, the standard package collection data and the spare parts data corresponding to the spare parts list, the target demand forecasting model corresponding to the maintenance project is determined, and then based on the target demand forecasting model, the second category spare parts demand forecasting for the maintenance project can be performed to generate the corresponding second category spare parts demand list. Finally, the server 104 can summarize and generate a planned spare parts demand quantity list based on the first category spare parts demand list and the second category spare parts demand list, and feed the planned spare parts demand quantity list back to the terminal 102. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers and portable wearable devices, and the server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0079] In one embodiment, Figure 2 As shown in the figure, a method for predicting the demand for spare parts in nuclear power plants is provided. Figure 1 The server in the example is used as an example to illustrate the specific steps:

[0080] Step S202, determining corresponding maintenance items according to the maintenance plan of the nuclear power plant, and performing item equivalence processing based on the maintenance items to generate a corresponding preliminary maintenance item list.

[0081] Specifically, by obtaining the preventive maintenance program of the nuclear power plant, and determining the corresponding nuclear power plant maintenance plan according to the preventive maintenance program of the nuclear power plant, and then screening the overhaul rounds based on the nuclear power plant maintenance plan, the maintenance items corresponding to the current overhaul rounds are obtained. Further, by obtaining the functional position and text content of each item corresponding to the maintenance item, and performing item equivalence processing based on the functional position and text content of each item corresponding to the maintenance item, the corresponding preliminary maintenance item list is generated.

[0082] Specifically, the preventive maintenance program of a nuclear power plant can be the ten-year preventive maintenance program for overhaul of a nuclear power plant as shown in Table 1 below. According to the ten-year preventive maintenance program for overhaul of a nuclear power plant, the corresponding round of overhaul plans can be sorted out. Referring to Table 1, it can be seen that the ten-year preventive maintenance program for overhaul of a nuclear power plant includes the maintenance cycle of the project, the execution of major overhaul cycles in history, and the execution plan of major overhaul cycles in the future. For projects with a maintenance cycle greater than 1C (1 overhaul cycle), there are situations where overhaul projects are carried out in advance or are equivalent. It is necessary to first extract the round of the most recent overhaul carried out on the project, and combine the project equivalence to comprehensively determine its implementation plan in future overhaul cycles.

[0083] For example, if the cycle of a project is 4C and its most recent implementation record is the 15th round of overhaul, if there is no equivalent record after the 15th round of overhaul, the project will be implemented in the 19th and 23rd rounds of overhaul. If the project has an equivalent record in the 17th round of overhaul, the project will be implemented in the 21st and 25th rounds of overhaul.

[0084] Table 1

[0085]

[0086] As shown in Table 2, a sample table of the maintenance plan of a project in a future overhaul round is provided. Referring to Table 2, it can be seen that the maintenance plan of a project in a future overhaul round includes specific data such as functional positions, project texts, main work centers, cycles, overhaul rounds, task list groups, and task lists corresponding to different maintenance items. Referring to the specific data recorded in Table 2, it can be further known whether there are maintenance items with the same functional position or the same text content, which is conducive to further project equivalence processing.

[0087] Table 2

[0088]

[0089]

[0090] Furthermore, after sorting out the overhaul maintenance plan of the project, the corresponding maintenance items of the overhaul can be obtained by screening the corresponding overhaul rounds. Since the same equipment has similar maintenance items (but different maintenance cycles), for example, for a motor equipment, the equipment needs to be disassembled and inspected every 1C, and some wearing parts in the motor need to be replaced. A fixed replacement is performed every 4C, that is, the entire motor is replaced. When executing the overall replacement project of 4C, it is not necessary to execute the disassembly inspection project of 1C, so it is necessary to perform project equivalence, thereby eliminating the disassembly inspection project of 1C.

[0091] Among them, by obtaining the functional position of each project and the project text content corresponding to the maintenance project, and based on the functional position of each project and the project text content corresponding to the maintenance project, project equivalence processing is performed. Specifically, it is necessary to embed the project equivalence module into the system, that is, in the same round of overhaul, if two or more projects have the same functional position and the same project text, the system directly eliminates the project with a short cycle and retains the project with a long cycle. When the functional position is the same and the similarity of the project text content is higher than a certain threshold, the information is displayed to the engineer to remind the engineer that such projects are suspected to be the same, so as to facilitate further manual project equivalence work.

[0092] In one embodiment, Figure 3 As shown in the figure, a schematic diagram of the classification of planned spare parts for nuclear power plants is provided. Figure 3 It can be seen that the process of classifying planned spare parts for nuclear power plants specifically includes: obtaining the preventive maintenance outline, determining the nuclear power plant maintenance plan according to the preventive maintenance outline, attaching spare parts data and maintenance plans according to the maintenance standard package, and further determining maintenance items. The maintenance items may include the first category of spare parts (i.e. Figure 3 Category A spare parts shown in the figure), and Category II spare parts (i.e. Figure 3 The first-class spare parts can be directly determined according to the first-class spare parts, and the corresponding first-class spare parts demand list (i.e. Figure 3 According to the structured data of spare parts use of the second category of spare parts and the historical maintenance projects, the second category of spare parts demand list (i.e. Figure 3 Planned Category B spare parts requirements as shown in ).

[0093] Step S204, obtaining a maintenance item change rate form corresponding to the preliminary maintenance item list, and screening maintenance items from the preliminary maintenance item list according to the maintenance item change rate form.

[0094] Specifically, after completing the project equivalent processing, a preliminary maintenance project list can be obtained. Due to the adjustment of the project implementation plan and the optimization of the project, the maintenance projects have certain changes, and the change rate is related to the lead time. The change rates of different projects in different specialties are also different. Therefore, it is necessary to calculate the annual change rate of each project in the ten-year preventive maintenance outline (where the data needs to be updated once after each round of overhaul), and record it in the maintenance project change rate form. By obtaining the maintenance project change rate form corresponding to the preliminary maintenance project list, the maintenance projects can be screened out from the preliminary maintenance project list according to the maintenance project change rate form.

[0095] Among them, the maintenance project change rate form should at least include the following information: maintenance project, power plant, 1-year change rate, 2-year change rate, 3-year change rate, 4-year change rate and 5-year change rate, etc.

[0096] In one embodiment, Figure 4 As shown in the figure, a schematic diagram of the change rate of each professional outline project in a nuclear power plant is provided. The change rate of maintenance projects is related to the lead time, and the change rates of different professional projects are also different. Figure 4 It can be seen that the change rates of electrical, instrumentation, transfer and static machines of the same years are different, and the change rates of the same items are also different in different years.

[0097] Step S206, obtaining preset constraints corresponding to the maintenance project, and determining a first category spare parts requirement list corresponding to the maintenance project according to the preset constraints and the maintenance standard package data.

[0098] Specifically, by obtaining the total amount threshold of project spare parts, the long procurement cycle threshold of project spare parts, and the standard package data corresponding to the maintenance project, and establishing a first mapping form between the standard package data and the total amount threshold of project spare parts, and the long procurement cycle threshold of project spare parts, then according to each maintenance project, the maintenance project change rate form and the first mapping form, determine the first category spare parts requirement list corresponding to the maintenance project.

[0099] Among them, maintenance projects can obtain the spare parts demand list required for maintenance projects by calling maintenance standard packages. In order to filter maintenance projects by the total amount of spare parts involved in the project and the spare parts procurement cycle, it is necessary to establish a form for the spare parts amount and procurement cycle corresponding to the standard package. The form should at least contain the following information: standard package, power plant, spare parts code, procurement cycle, unit price, quantity and other data.

[0100] In one embodiment, Figure 5 As shown, a schematic diagram of sorting out the first category of spare parts requirements list is provided. Figure 5 It can be seen that when sorting out the first type of spare parts demand list, the specific data that needs to be applied includes the maintenance project change rate form, the first mapping form (i.e. Figure 5 The standard package corresponding to the spare parts amount shown in the table is the procurement cycle form), maintenance items and constraints, etc., where the constraints can be understood as screening conditions, that is, screening is performed through dimensions such as project change rate, spare parts long cycle threshold, and spare parts amount involved in the project. The corresponding change rate of the maintenance project must be calculated first, and the spare parts requirement list corresponding to the standard package must be obtained. By entering the above screening conditions, the corresponding first-category spare parts requirement list can be screened out.

[0101] Among them, the constraints mainly refer to the total amount threshold of spare parts involved in a single project, the change rate of a project, and the long procurement cycle threshold. By setting the constraints reasonably, we can find a balance point among the concentration of demand submissions, the degree of spare parts demand guarantee, and the risk of inventory backlog, and then determine the corresponding first-category spare parts demand list.

[0102] In one embodiment, Figure 6 As shown, a schematic diagram of constraint condition setting corresponding to the first type of spare parts requirement list is provided, referring to Figure 6 It can be seen that for the first type of spare parts, that is, Figure 6 For the maintenance project shown in the figure - Class A spare parts, firstly, it is necessary to determine whether the total price of the project and spare parts is less than RMB 10,000. If it is determined that the total price of the project and spare parts is less than RMB 10,000, the spare parts corresponding to the project are included in the first category spare parts demand list (i.e. Figure 6 Secondly, when it is determined that the total price of the project and spare parts is not less than 10,000 yuan, it is further determined whether the three-year change rate of the project is less than 15%. When it is determined that the three-year change rate of the project is less than 15%, the spare parts corresponding to the project are included in the first category spare parts demand list (i.e. Figure 6 The planned Class A spare parts requirement list is shown in the table above.

[0103] Among them, when it is determined that the three-year change rate of the project is not less than 15%, it is further determined whether the spare parts procurement cycle is greater than 12 months. When it is determined that the spare parts procurement cycle is greater than 12 months, the spare parts corresponding to the project are included in the first category spare parts demand list (i.e. Figure 6 When the spare parts procurement cycle is no more than 12 months, the spare parts corresponding to the project will not be included in the first category spare parts demand list.

[0104] Further, refer to Figure 6It can be seen that the maintenance projects have been sorted out through the previous process, and the total amount threshold of spare parts involved in a single project is set to 10,000 yuan, that is, the spare parts demand involved in projects below 10,000 yuan is directly included in the planned Class A spare parts demand list. The amount is low, the impact of inventory backlog is small, and the degree of security and concentration can be improved. The 3-year change rate is set to within 15%, that is, the spare parts demand involved in projects with a 3-year change rate of 15% is directly included in the planned Class A spare parts demand list. The change rate is low, the risk of inventory backlog is small, and the degree of security and concentration can be improved. The long spare parts procurement cycle is set to 12 months, that is, for projects that do not meet the amount and change rate requirements, if there are spare parts involving a long procurement cycle, only the spare parts with a long procurement cycle are included in the planned Class A spare parts demand list. In this case, the risk of inventory backlog is high, and the guarantee of spare parts with a long procurement cycle is difficult, so only spare parts with a long procurement cycle are included in the Class A spare parts demand list. However, in actual applications, corresponding constraints can be set or adjusted according to the inventory assessment requirements of each power plant, the maturity of unit operation, and other conditions.

[0105] Step S208, obtaining the collection records corresponding to the preliminary maintenance item list, and integrating the collection data based on the collection records to obtain updated collection record data.

[0106] Specifically, by obtaining the receipt records corresponding to the preliminary maintenance item list, and filtering the receipt records to obtain the duplicate spare parts records and the replacement spare parts records, and then integrating the receipt data based on the duplicate spare parts records, the replacement spare parts records and the receipt records, the updated receipt record data is obtained.

[0107] Among them, since the spare parts of nuclear power plants have duplicate code mergers and item substitutions, it is necessary to integrate the spare parts requisition data involving duplicate codes and substitutions, that is, to integrate the requisition data of the original code into the requisition data of the new code. Taking the database of a nuclear power plant as an example, the spare parts codes involving duplicate code mergers will add "= new code" to the material description of the original code, and the spare parts codes involving item substitutions will add new code information to the subsequent materials of the original code. Therefore, by extracting the material description and the subsequent material field information, the corresponding relationship between the original code and the new code can be judged and extracted, and the requisition data can be integrated.

[0108] Furthermore, if Figure 7 In the schematic diagram of updating the requisition record data of the nuclear power plant spare parts demand forecasting method shown, spare part A is replaced by spare part B through recoding or substitution, and spare part B is replaced by spare part C through recoding or substitution. Finally, the requisition data of spare parts A and B are integrated into spare part C to obtain updated requisition record data.

[0109] Step S210: establishing a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts based on the updated requisition record data.

[0110] Specifically, based on the updated requisition record data, a first benchmark list between maintenance items and spare parts is established, wherein the first benchmark list includes at least the following columns: maintenance item, outline change rate, attachment standard package number, spare parts code, spare parts unit price, spare parts importance, spare parts type in the standard package, spare parts attachment quantity in the standard package, total maintenance item call-up times, total spare parts requisition quantity, maximum single requisition quantity of spare parts, and spare parts requisition fluctuation coefficient and other data.

[0111] Similarly, based on the updated requisition record data, a second benchmark list between standard packages and spare parts is prepared, wherein the second benchmark list includes at least the following: standard package number, spare parts code, spare parts unit price, spare parts importance, spare parts type in the standard package, spare parts attachment quantity in the standard package, total number of standard package calls, total spare parts requisition quantity, maximum single requisition quantity of spare parts, and spare parts requisition fluctuation coefficient and other data.

[0112] Among them, the potential spare parts demand involved in the maintenance project includes two parts. One part is the spare parts mounted in the standard package corresponding to the maintenance project, and the other part is the spare parts that have historical use records but are not mounted in the standard package. If only the spare parts mounted in the standard package are referred to, the demand for spare parts that have been historically used, that is, 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 use information, and sort out the maintenance projects and the corresponding potential spare parts demand.

[0113] Furthermore, by combing through the records of maintenance project requisition, a maintenance project-spare parts benchmark list, i.e., the corresponding first benchmark list, is established for different types of maintenance project-spare parts models to call. Among them, the first benchmark list contains at least the following columns: maintenance project, outline change rate, attachment standard package number, spare part code, spare part unit price, spare part importance, spare part type in the standard package, spare part attachment quantity in the standard package, total maintenance project call times, total spare parts requisition quantity, maximum single spare parts requisition quantity, and spare parts requisition fluctuation coefficient, etc. Among them, maintenance project + spare part code is the primary key of the first benchmark list.

[0114] In one embodiment, Figure 8 As shown in the figure, a potential spare parts demand relationship involved in a maintenance project is provided. Figure 8 It can be seen that the potential required spare parts involved in the maintenance project include the spare parts mounted in the standard package corresponding to the maintenance project, corresponding to Figure 8 Spare parts 1, 2, and 3 in the standard package, as well as spare parts with historical use records but not attached in the standard package, correspond to Figure 8Spare parts 4 and 5 in the standard package. It is understandable that if only the spare parts mounted in the standard package are referred to, the demand for spare parts 4 and 5 cannot be output, so it is necessary to summarize the spare parts mounted in the standard package and the historical requisition information, and sort out the maintenance items and the corresponding potential spare parts demand.

[0115] In one embodiment, Fig. 9 As shown, a schematic diagram of the relationship between the maintenance items corresponding to a standard package and the historical use records is provided. Fig. 9 It can be seen that since there is no or few historical records of requisition, it is impossible to infer the spare parts demand corresponding to the maintenance project through historical requisition information. In this case, the historical requisition records of the standard package information attached to the maintenance project can be calculated. Fig. 9 For example, maintenance project 2 has no historical use record, but maintenance project 1, maintenance project 2, maintenance project 3 and other projects are attached to the same standard package, and maintenance project 1, maintenance project 3 and other projects have historical use data. At this time, the historical use information of the standard package can be calculated.

[0116] Furthermore, by combing the use records of the standard package, a standard package-spare parts benchmark list, i.e., the corresponding second benchmark list, is established for different types of standard package-spare parts models to call. Among them, the second benchmark list at least includes the following columns: standard package number, spare part code, spare part unit price, spare part importance, spare part type in the standard package, number of spare parts attached in the standard package, total number of calls to the standard package, total number of spare parts used, maximum number of spare parts used at a time, and spare parts use fluctuation coefficient, etc. Among them, the standard package number + spare part code is the primary key of the second benchmark list.

[0117] Step S212, obtain the maintenance project collection data corresponding to the first benchmark list and the standard package collection data corresponding to the second benchmark list, and determine the target demand forecasting model corresponding to the maintenance project based on the maintenance project collection data, the standard package collection data and the spare parts data corresponding to the spare parts list.

[0118] Specifically, when it is determined that the maintenance project requisition data meets the first requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the maintenance project and spare parts model. When it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data meets the second requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the standard package and spare parts model.

[0119] Similarly, when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data does not meet the second requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is a spare parts model.

[0120] In one embodiment, Fig.10 As shown, a schematic diagram of the relationship between the spare parts corresponding standard package and the historical use record is provided. Fig.10 It can be seen that for some standard packages, since there is no or few historical records of use, it is impossible to infer the spare parts demand corresponding to the standard package through historical use information. In this case, the call and use records of the spare parts attached to the standard package in other standard packages can be calculated. Fig.10 For example, a spare part is attached to standard package 1, standard package 2, standard package 3, and standard package 4, among which standard package 2 has no historical use record. At this time, the call and use data of the spare part in the other standard packages can be calculated. By sorting out the use records of spare parts, a benchmark list of spare parts use is established for use by different types of spare parts models. The benchmark list contains at least the following columns: spare part code, spare part unit price, spare part importance, total number of spare part calls, total number of spare parts used, maximum number of spare parts used at a time, and spare part use fluctuation coefficient. Among them, the spare part code is the primary key of the spare parts list.

[0121] In one embodiment, Fig.11 As shown, a schematic diagram of the model selection process corresponding to the second type of spare parts requirement list is provided. Fig.11 It can be seen that after sorting out the maintenance project list, the selected model is determined by judging whether the maintenance project and standard package collection data are sufficient. Specifically, for example, if the historical collection data of a maintenance project is sufficient (for example, there are 3 or more collection data in history), the maintenance project and spare parts model are directly selected to calculate the spare parts demand. If the historical call data of the maintenance project is insufficient, it is judged whether the collection data of the corresponding standard package is sufficient. If the standard package collection data is sufficient, the standard package and spare parts model are called, otherwise the spare parts model is called. If in the spare parts model, the spare part has no collection record in each standard package, you can choose to output the demand quantity of the spare part as 0, or directly use the demand quantity of the spare part in the standard package.

[0122] Step S214: Based on the target demand forecasting model, forecast the demand for second-category spare parts for the maintenance project, and generate a corresponding demand list for second-category spare parts.

[0123] Specifically, the total number of calls, total number of requisitions, outline change rate, spare parts importance, spare parts unit price and fluctuation coefficient in the first benchmark list are extracted, and the first number of calls in this round is obtained. Based on the total number of calls, total number of requisitions, outline change rate, spare parts importance, spare parts unit price and fluctuation coefficient in the first benchmark list, and the first number of calls in this round, the first input data corresponding to the maintenance item and spare parts model is obtained.

[0124] Similarly, according to the maintenance items and spare parts model, the first input data is subjected to demand forecasting to obtain the corresponding first forecast demand data, and the total number of calls, total number of requisitions, importance of spare parts, unit price of spare parts, and fluctuation coefficient in the second benchmark list are extracted to obtain the second number of calls in this round. Then, according to the total number of calls, total number of requisitions, importance of spare parts, unit price of spare parts, fluctuation coefficient in the second benchmark list, and the second number of calls in this round, the second input data corresponding to the standard package and spare parts model is obtained;

[0125] Similarly, according to the standard package and spare parts model, the second input data is subjected to demand forecasting to obtain the corresponding second forecast demand data, and the total number of calls, total number of withdrawals, importance of spare parts, unit price of spare parts and fluctuation coefficient in the spare parts list are extracted to obtain the third number of calls in this round. Then, according to the total number of calls, total number of withdrawals, importance of spare parts, unit price of spare parts, fluctuation coefficient and the third number of calls in this round in the spare parts list, the third input data corresponding to the spare parts model is obtained, and then according to the spare parts model, the third input data is subjected to demand forecasting to obtain the corresponding third forecast demand data.

[0126] Furthermore, a corresponding second category spare parts demand list is generated based on the first predicted demand data, the second predicted demand data and the third predicted demand data.

[0127] In one embodiment, Fig.12 , provides a schematic diagram of the framework for sorting out the quantity of requirements corresponding to the second category spare parts requirement list, refer to Fig.12 It can be seen that for each maintenance project, three factors need to be considered: maintenance project, standard package and spare parts, which can be specifically divided into maintenance project-spare parts factor, standard package-spare parts factor and spare parts factor. Among them, for the maintenance project-spare parts factor, it is necessary to determine the corresponding target demand forecasting model from the maintenance project and spare parts model library, and then perform demand forecasting based on the target demand forecasting model to obtain the first forecast demand data.

[0128] Similarly, for the standard package-spare parts factor, it is necessary to determine the corresponding target demand forecast model from the standard package and spare parts model library, and then perform demand forecasting based on the target demand forecast model to obtain the second forecast demand data. For the spare parts factor, it is also necessary to determine the corresponding target demand forecast model from the spare parts model library, and then perform demand forecasting based on the target demand forecast model to obtain the third forecast demand data. Among them, the models in the model library can be diversified, and different types of models can be set according to needs, and the model with the best forecasting effect can be selected for different application scenarios.

[0129] Step S216: Based on the first category spare parts requirement list and the second category spare parts requirement list, a planned spare parts requirement quantity list is generated by summarizing.

[0130] Specifically, the corresponding first-category spare parts demand quantity can be obtained based on the first-category spare parts demand list, and the corresponding second-category spare parts demand data can be obtained based on the second-category spare parts demand list. Further, based on the first-category spare parts demand quantity and the second-category spare parts demand quantity, all planned spare parts demand quantities can be summarized to generate the corresponding planned spare parts demand quantity list.

[0131] In the above-mentioned nuclear power plant spare parts demand forecasting method, the corresponding maintenance items are determined according to the nuclear power plant maintenance plan, and the project equivalent processing is performed based on the maintenance items to generate the corresponding preliminary maintenance item list, and the maintenance items are screened from the preliminary maintenance item list according to the maintenance item change rate form. Then, by obtaining the preset constraints corresponding to the maintenance items, and according to the preset constraints and the maintenance standard package data, the first category spare parts demand list corresponding to the maintenance items is determined. By obtaining the corresponding receipt records of the preliminary maintenance item list, and integrating the receipt data based on the receipt records, the updated receipt record data is obtained, and then according to the updated receipt record data, the first benchmark list between the maintenance items and the spare parts, and the second benchmark list between the standard package and the spare parts are established. By obtaining the maintenance project collection data corresponding to the first benchmark list and the standard package collection data corresponding to the second benchmark list, and according to the maintenance project collection data, the standard package collection data and the spare parts data corresponding to the spare parts list, the target demand forecasting model corresponding to the maintenance project is determined, and based on the target demand forecasting model, the second category spare parts demand forecast is performed for the maintenance project to generate the corresponding second category spare parts demand list. Then, according to the first category spare parts demand list and the second category spare parts demand list, a planned spare parts demand quantity list is summarized and generated. It realizes the accurate determination of the different categories of spare parts demand of the nuclear power plant during the maintenance process, and can be purchased in advance according to the obtained planned spare parts demand quantity list to avoid spare parts shortage during the maintenance process, so as to prepare sufficient maintenance spare parts in advance to meet maintenance needs and further ensure the stable operation of the nuclear power plant.

[0132] In one embodiment, Fig.13 As shown, the step of determining the target demand forecast model corresponding to the maintenance project, that is, the step of determining the target demand forecast model corresponding to the maintenance project according to the maintenance project requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list, specifically includes:

[0133] Step S1302: When it is determined that the maintenance project requisition data meets the first requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is a maintenance project and spare parts model.

[0134] Specifically, when it is determined that the maintenance project requisition data meets the first requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be a maintenance project and spare parts model, wherein, in this embodiment, the maintenance project and spare parts model is a first type of maintenance project and spare parts model.

[0135] Furthermore, if Fig.14 As shown, a schematic diagram of the framework of a first type of maintenance project and spare parts model is provided, referring to Fig.14 It can be seen that the first type of maintenance project and spare parts model needs to be applied to the total number of calls D, the total number of requisitions L, the outline change rate R, the spare parts importance I, the spare parts unit price P, the fluctuation coefficient B and the prior knowledge X, among which the total number of calls D, the total number of requisitions L, the outline change rate R, the spare parts importance I, the spare parts unit price P, and the fluctuation coefficient B are extracted from the first benchmark list between maintenance projects and spare parts, and the prior knowledge X is supplemented manually. Then, based on the specific data involved above, the demand quantity is predicted to obtain the corresponding first predicted demand data.

[0136] Specifically, the following formula (1) is used to calculate the first predicted demand data:

[0137]

[0138] Among them, Y is the predicted demand quantity, L is the total number of withdrawals, D is the total number of calls, and K is the correction coefficient.

[0139] The calculation formula of the correction coefficient K is shown in the following formula (2):

[0140] K=f(R,I,P,B,X);(2)

[0141] Among them, the correction coefficient L is determined by the outline change rate R, spare parts importance I, spare parts unit price P, fluctuation coefficient B, and prior knowledge X. The higher the outline change rate value, the higher the probability of cancellation of the maintenance project, and less reserves should be made. The higher the spare parts importance, the more serious the risk of spare parts shortage, and more reserves should be made. The higher the spare parts unit price, the higher the risk of inventory backlog, and less reserves should be made. The fluctuation coefficient refers to the standard deviation of historical call-up and the average value of historical call-up and requisition. The higher its value, the greater the fluctuation of historical spare parts requisition, the risk of backlog, and less reserves should be made. Prior knowledge is the user's estimate of equipment failure. If the estimated failure probability of the equipment at the next overhaul is higher than the historical average, more reserves should be made. If the estimated failure probability of the equipment at the next overhaul is lower than the historical average, less reserves should be made. If the user has no effective expectation of the failure of the equipment, the prior knowledge item should not be applied. In summary, R, P, B are negatively correlated with K, and I, X are positively correlated with K.

[0142] Step S1304, when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data meets the second requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is the standard package and spare parts model.

[0143] Specifically, when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement, but the standard package requisition data meets the second requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is the standard package and spare parts model. In this embodiment, the standard package and spare parts model is a first-class standard package and spare parts model.

[0144] Furthermore, if Fig.15 As shown, a schematic diagram of the framework of a first type of standard package and spare parts model is provided, referring to Fig.15 It can be seen that the first type of standard package and spare parts model needs to be applied to the total number of calls D, the total number of requisitions L, the number of calls in this round N, the importance of spare parts I, the unit price of spare parts P, the fluctuation coefficient B and the prior knowledge X. Among them, the total number of calls D, the total number of requisitions L, the importance of spare parts I, the unit price of spare parts P, and the fluctuation coefficient B are extracted from the second benchmark list between standard packages and spare parts, the prior knowledge X is supplemented manually, and the number of calls in this round N is also supplemented manually. Then, based on the above-mentioned specific data, the demand quantity is predicted to obtain the corresponding second predicted demand data.

[0145] Specifically, the following formula (3) is used to calculate the second forecast demand data:

[0146]

[0147] Among them, Y is the predicted demand quantity, L is the total number of withdrawals, D is the total number of calls, N is the number of calls in this round. Since the same standard package may correspond to multiple maintenance projects, the number of calls in this round refers to how many maintenance projects have called the standard package in this round of overhaul, and K is the correction coefficient.

[0148] The calculation formula of the correction coefficient K is shown in the following formula (4):

[0149] K=f(I,P,B,X); (4)

[0150] Among them, the correction coefficient K is comprehensively determined by the importance of spare parts I, the unit price of spare parts P, the fluctuation coefficient B, and the prior knowledge X. The higher the importance of spare parts, the more serious the risk of spare parts shortage, and more reserves should be reserved. The higher the unit price of spare parts, the higher the risk of inventory backlog, and less reserves should be reserved. The fluctuation coefficient refers to the standard deviation of historical call-up and the average value of historical call-up and requisition. The higher its value, the greater the fluctuation of historical spare parts requisition, the risk of backlog, and less reserves should be reserved. Prior knowledge is the user's estimate of the overall failure situation of each device corresponding to the standard package. If the estimated failure probability of this type of equipment at the next overhaul is higher than the historical average, more reserves should be reserved. If the estimated failure probability of this type of equipment at the next overhaul is lower than the historical average, less reserves should be reserved. If the user has no effective expectation of the failure of this type of equipment, the prior knowledge item should not be applied. In summary, P, B and K are negatively correlated, and I, X and K are positively correlated.

[0151] Step S1306, when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data does not meet the second requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is a spare parts model.

[0152] Specifically, when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data does not meet the second requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is a spare parts model. In this embodiment, the spare parts model is a first type spare parts model.

[0153] Furthermore, if Fig.16 As shown, a schematic diagram of the framework of a first type of spare parts model is provided, referring to Fig.16 It can be seen that the first type of spare parts model needs to be applied to the total number of calls D, the total number of requisitions L, the number of calls in this round N, the importance of spare parts I, the unit price of spare parts P, and the fluctuation coefficient B, among which the total number of calls D, the total number of requisitions L, the number of calls in this round N, the importance of spare parts I, the unit price of spare parts P, and the fluctuation coefficient B are obtained from the spare parts list. Then, according to the above-mentioned specific data, the demand quantity is predicted to obtain the corresponding third predicted demand data.

[0154] Specifically, the following formula (5) is used to calculate the third forecast demand data:

[0155]

[0156] Among them, Y is the predicted demand quantity, L is the total number of withdrawals, D is the total number of calls, and N is the number of calls in this round. Since the same standard package may correspond to multiple maintenance projects, the number of calls in this round refers to how many maintenance projects have called the standard package in this round of overhaul, and K is the correction coefficient.

[0157] The calculation formula of the correction coefficient K is shown in the following formula (6):

[0158] K=f(I,P,B); (6)

[0159] Among them, the correction coefficient K is determined by the importance of spare parts I, the unit price of spare parts P, and the fluctuation coefficient B. The higher the importance of spare parts, the more serious the risk of spare parts shortage, and more reserves should be reserved. The higher the unit price of spare parts, the higher the risk of inventory backlog, and less reserves should be reserved. The fluctuation coefficient refers to the standard deviation of historical call-up and the average value of historical call-up and requisition. The higher its value, the greater the fluctuation of historical spare parts requisition, the risk of backlog, and less reserves should be reserved. In summary, P, B and K are negatively correlated, and I and K are positively correlated.

[0160] In this embodiment, when it is determined that the maintenance project requisition data meets the first requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the maintenance project and spare parts model. When it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data meets the second requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the standard package and spare parts model. When it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data does not meet the second requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the spare parts model. Comprehensive consideration is made based on the maintenance project requisition data and the standard package requisition data to screen out a matching target demand forecasting model, and then when the determined target demand forecasting model is used to perform demand data in the subsequent process, the accuracy of the obtained forecast demand data can be improved, the situation of repeated demand forecasting can be reduced, and the accuracy of demand forecasting and work efficiency can be improved.

[0161] In one embodiment, Fig.17 As shown, the step of determining the target demand forecast model corresponding to the maintenance project, that is, the step of determining the target demand forecast model corresponding to the maintenance project according to the maintenance project requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list, specifically includes:

[0162] Step S1702, determining the corresponding maintenance item category according to the maintenance item requisition data, determining the corresponding standard package category according to the standard package requisition data, and determining the corresponding spare parts category according to the spare parts data.

[0163] Specifically, the corresponding maintenance project category is determined according to the maintenance project requisition data, such as maintenance project category A, maintenance project category B, ... maintenance project category X; the corresponding standard package category is determined according to the standard package requisition data, such as standard package category A, standard package category B, ... standard package category X; the corresponding spare parts category is determined according to the spare parts data, such as spare parts category one, spare parts category two, ... spare parts category N, etc.

[0164] Step S1704, based on the combined data of the maintenance item category and the spare part category, the corresponding maintenance item and spare part model is matched from the maintenance item and spare part model library, and determined as the target demand forecasting model.

[0165] Specifically, based on the combined data of maintenance project categories and spare parts categories, for example, based on the combined data of maintenance project categories A, maintenance project categories B, ... maintenance project categories X, and spare parts categories 1, spare parts categories 2, ... spare parts categories N, the corresponding maintenance project and spare parts models are matched from the maintenance project and spare parts model library and determined as the corresponding target demand forecasting model. In this embodiment, the target demand forecasting model is the second type of maintenance project and spare parts model.

[0166] Furthermore, if Fig.18 As shown, a schematic diagram of the framework of the second type of maintenance items and spare parts model is provided, referring to Fig.18 It can be seen that the second type of maintenance project and spare parts model is an artificial neural network model, and the input data of the second type of maintenance project and spare parts model are: total number of calls D, total number of requisitions L, maximum single requisition M, maintenance project outline cycle T, outline change rate R and maintenance project category. These input data are input through the input layer of the artificial neural network, processed through the middle layer of the neural network, and the corresponding first predicted demand data are obtained, which are output through the output layer of the artificial neural network.

[0167] Among them, the total number of calls D, the total number of requisitions L, the maximum single requisition M, and the outline change rate R are extracted from the first benchmark list between maintenance items and spare parts. The maintenance item outline cycle T is the maintenance cycle of the maintenance item in the maintenance outline. Generally, for maintenance items of the same type, the longer the maintenance cycle, the higher the probability of replacing spare parts. The maintenance item category can be divided according to the responsible profession of the maintenance item. By refining the category of maintenance items, the overall effectiveness of demand forecasting can be improved.

[0168] In one embodiment, Fig.19 As shown, a schematic diagram of the structure of a maintenance project and spare parts model library is provided. Fig.19It can be seen that the maintenance project and spare parts model library provides combined data obtained by combining different maintenance project categories and spare parts categories, and the corresponding maintenance project and spare parts models. Among them, in the maintenance project and spare parts model library, maintenance projects are divided according to the equipment function location dimension, and spare parts are divided according to the material category dimension, and a maintenance project-spare parts model library is established. When applied, according to the category of maintenance project and spare parts category, the model in the model library is automatically selected for demand forecasting.

[0169] For example, maintenance item category A can be combined with spare parts category one, spare parts category two to spare parts category N respectively. Similarly, maintenance item categories B to maintenance item category X can also be combined with spare parts category one, spare parts category two to spare parts category N respectively. Therefore, each combination of maintenance item category and spare parts category can be matched with a corresponding model in the maintenance item and spare parts model library.

[0170] Among them, since there are a large number of maintenance items and spare parts in nuclear power plants, the usage probabilities of different maintenance items and spare parts are different. If only one artificial neural network model is used, the same input during training may result in different demand quantity outputs, and the artificial neural network model is difficult to converge. To solve this problem, we can model them separately according to the maintenance item category and spare parts category, set up multiple artificial neural network models, and improve the effectiveness of the prediction.

[0171] Step S1706, based on the combined data of the standard package category and the spare part category, the corresponding standard package and spare part model is matched from the standard package and spare part model library to be determined as the target demand forecasting model.

[0172] Specifically, based on the combined data of the standard package category and the spare parts category, for example, based on the combined data of the standard package category A, the standard package category B, ... the standard package category X, and the spare parts category 1, the spare parts category 2, ... the spare parts category N, the corresponding standard package and spare parts model is matched from the standard package and spare parts model library and determined as the target demand forecasting model. In this embodiment, the target demand forecasting model is the second type of standard package and spare parts model.

[0173] Furthermore, if Fig. 20 As shown, a schematic diagram of the framework of a second type of standard package and spare parts model is provided, referring to Fig. 20 It can be seen that the second type of standard package and spare parts model is an artificial neural network model. The input data of the second type of standard package and spare parts model are: total number of calls D, total number of requisitions L, maximum single requisition M, maintenance project outline cycle T, number of calls in this round N and standard package category. These input data are input through the input layer of the artificial neural network, processed through the middle layer of the neural network, and the corresponding second predicted demand data are obtained, which are output through the output layer of the artificial neural network.

[0174] Among them, the total number of calls D, the total number of requisitions L, and the maximum number of single requisitions M are extracted from the second benchmark list between the standard package and the spare parts, and the maintenance project outline cycle is the average maintenance cycle of the maintenance project that calls the standard package in this round of overhaul. The standard package category can be divided according to the responsible profession of the standard, and the overall effectiveness of demand forecasting can be improved by refining the category of the standard package.

[0175] In one embodiment, Fig.21 As shown, a schematic diagram of the structure of a standard package and spare parts model library is provided. Fig.21 It can be seen that in the standard package and spare parts model library, the standard package is divided according to the equipment function location dimension, and the spare parts are divided according to the material category dimension, and the standard package and spare parts model library is established. When applied, according to the category of the standard package and the category of the spare parts, the model in the model library is automatically selected for demand forecasting.

[0176] For example, standard package category A can be combined with spare parts category one, spare parts category two to spare parts category N respectively. Similarly, standard package category B to standard package category X can also be combined with spare parts category one, spare parts category two to spare parts category N respectively. Therefore, each combination of standard package category and spare parts category can be matched with a corresponding model in the standard package and spare parts model library.

[0177] Step S1708, obtaining supplier data for providing spare parts of different categories, and determining the corresponding supplier category according to the supplier data.

[0178] Specifically, data of suppliers providing different categories of spare parts are obtained, and corresponding supplier categories are determined according to the supplier data, such as supplier A, supplier B, ..., supplier X.

[0179] Step S1710, based on the combined data of spare parts category and supplier category, a corresponding spare parts model is matched from the spare parts model library and determined as a target demand forecasting model.

[0180] Specifically, based on the combined data of spare parts category and supplier category, for example, based on the combined data of supplier A, supplier B, ... supplier X, and spare parts category 1, spare parts category 2, ... spare parts category N, a corresponding spare parts model is matched from the spare parts model library and determined as the target demand forecasting model. In this embodiment, the spare parts model is a second-category spare parts model.

[0181] Furthermore, if Fig. 22 As shown, a schematic diagram of the framework of a second type of spare parts model is provided, referring to Fig. 22It can be seen that the second type of spare parts model is an artificial neural network model. The input data of the second type of spare parts model are: total number of calls D, total number of requisitions L, maximum single requisition M, number of calls in this round N and spare parts category. These input data are input through the input layer of the artificial neural network, processed through the middle layer of the neural network, and the corresponding third predicted demand data are obtained, which are output through the output layer of the artificial neural network.

[0182] Among them, the total number of calls D, the total number of requisitions L, and the maximum number of single requisitions M are obtained from the spare parts list. Spare parts categories can be divided according to the standard responsible profession. By refining the categories of spare parts, the overall effectiveness of demand forecasting can be improved.

[0183] In one embodiment, Fig.23 As shown, a schematic diagram of the structure of a spare parts model library is provided. Fig.23 It can be seen that in the spare parts model library, spare parts are divided according to the material category dimension, and suppliers are divided according to supplier categories to establish a spare parts model library. It should be noted that spare parts are divided according to responsible units in the artificial neural network model, and they are divided according to material categories in the model library, that is, the same spare part will be divided in two dimensions. According to the spare parts category and supplier category, the model in the model library is automatically selected for demand forecasting.

[0184] For example, supplier A can be combined with spare parts category one, spare parts category two to spare parts category N respectively. Similarly, supplier B to supplier X can also be combined with spare parts category one, spare parts category two to spare parts category N respectively. Therefore, each combination of supplier and spare parts category can be matched with a corresponding model in the spare parts model library.

[0185] In one embodiment, Fig.24 As shown, a neural network spare parts demand forecasting process is provided. Fig.24 It can be seen that for the models in different model libraries, the overhaul training data of the N-1th round and before are first summarized. Taking the maintenance project and spare parts model as an example, the training input is 6 columns of data (total number of calls, total number of requisitions, maximum single requisition, maintenance project outline cycle, outline change rate, maintenance project category), and the output is 1 column of data (actual overhaul requisition data). After the Nth round of overhaul is completed, the planned requisition data of the Nth round of overhaul is extracted, the training data table is updated, and a new total training data table for the Nth round and before overhaul is created. At this time, the data is used to train the artificial neural network model and update the threshold and weight of the artificial neural network model.

[0186] Secondly, after sorting out the N+1 round of overhaul project list, organize the data according to the 6-column input requirements of the artificial neural network model, create a new N+1 round of overhaul forecast data input list, and use the artificial neural network model to predict the spare parts demand for the N+1 round of overhaul. After the N+1 round of overhaul is completed, extract the N+1 round of overhaul collection data, train the artificial neural network model again, and iterate to predict the spare parts demand for the N+2 round of overhaul. In practical applications, the Nth round of overhaul artificial neural network model can also be used directly to predict the spare parts demand for the N+2, N+3 and other rounds of overhaul.

[0187] In this embodiment, the corresponding maintenance project category is determined according to the maintenance project requisition data, the corresponding standard package category is determined according to the standard package requisition data, and the corresponding spare parts category is determined according to the spare parts data, and then based on the combined data of the maintenance project category and the spare parts category, the corresponding maintenance project and spare parts model are matched from the maintenance project and spare parts model library to be determined as the target demand forecast model. Based on the combined data of the standard package category and the spare parts category, the corresponding standard package and spare parts model are matched from the standard package and spare parts model library to be determined as the target demand forecast model. By obtaining the supplier data for providing spare parts of different categories, and determining the corresponding supplier category according to the supplier data, and then based on the combined data of the spare parts category and the supplier category, the corresponding spare parts model is matched from the spare parts model library to be determined as the target demand forecast model. It is possible to accurately determine the matching target demand forecasting model based on the combined data of maintenance project category and spare parts category, the combined data of standard package category and spare parts category, and the combined data of spare parts category and supplier category. This can improve the accuracy of the predicted demand data when the determined target demand forecasting model is used to perform demand data in the future, reduce the need for repeated demand forecasting, and improve demand forecasting accuracy and work efficiency.

[0188] In one embodiment, Fig.25 As shown, a method for predicting the demand for spare parts of a nuclear power plant is provided, which specifically includes the following steps:

[0189] Step S2502: Determine the corresponding virtual reserved quantity of spare parts according to the planned spare parts requirement quantity list.

[0190] Specifically, the planned A-type spare parts demand quantity, i.e., the first-type spare parts demand list, and the planned B-type spare parts demand quantity, i.e., the second-type spare parts demand list, are calculated and summarized into a planned spare parts demand quantity list. The planned spare parts procurement quantity can be calculated by inputting the planned spare parts demand quantity list into the procurement forecast quantity module.

[0191] Among them, the planned spare parts demand quantity list can be converted into a virtual reservation quantity. The sum of the virtual reservation and the real reservation of the spare parts is the spare parts reservation quantity. Combined with the spare parts inventory quantity, the regular order quantity, and the inventory parameters, the spare parts MRP (Material Requirements Planning) operation module is used to calculate the spare parts purchase quantity. Among them, the virtual reservation is a temporary reservation before the corresponding maintenance project has a formal work order reservation. Once the maintenance project generates a formal work order reservation, it is equivalent to the virtual reservation being converted into a formal reservation. The corresponding virtual reservation is automatically deleted to avoid two reservations for the same maintenance project, which leads to excessive procurement.

[0192] Step S2504, obtaining spare parts real reservation data corresponding to the planned spare parts requirement quantity list, and obtaining the corresponding actual spare parts reservation quantity according to the spare parts real reservation data and the spare parts virtual reservation quantity.

[0193] Specifically, the real spare parts reservation data corresponding to the planned spare parts demand quantity list is obtained, and the sum of the real spare parts reservation data and the virtual spare parts reservation quantity is calculated to obtain the corresponding actual spare parts reservation quantity.

[0194] Step S2506, obtain the inventory quantity, regular order quantity and inventory parameters corresponding to the spare parts list, and generate the fourth input data corresponding to the trained procurement forecast data model based on the actual spare parts reserved quantity, inventory quantity, regular order quantity and inventory parameters.

[0195] Specifically, by obtaining the inventory quantity, the ordered quantity and the inventory parameters corresponding to the spare parts list and combining them with the calculated actual spare parts reserved quantity, the fourth input data corresponding to the trained procurement forecast data model is generated.

[0196] Step S2508: According to the trained purchase forecast data model, the purchase quantity of the fourth input data is predicted to generate a corresponding planned spare parts purchase list.

[0197] Specifically, by inputting the fourth input data into the trained procurement forecast data model, the actual spare parts reserved quantity, inventory quantity, order quantity and inventory parameters are predicted according to the MRP (Material Requirements Planning) operation module set in the trained procurement forecast data model, and the corresponding planned spare parts procurement forecast quantity is generated, and then based on the planned spare parts procurement forecast quantity, the corresponding planned spare parts procurement list is generated.

[0198] Further, if Fig.26 As shown, a framework for a procurement forecast data model is provided. Fig.26It can be seen that according to the planned demand quantity of Class A spare parts and the planned demand quantity of Class B spare parts, the planned spare parts demand quantity list can be summarized and calculated, and the planned spare parts demand quantity list can be further input into the procurement forecast data model. Combined with spare parts inventory parameters, spare parts inventory, spare parts order and spare parts reservation data, the spare parts MRP (material requirement planning) operation module is used to predict the purchase quantity of spare parts and make corresponding spare parts procurement applications.

[0199] In this embodiment, the corresponding virtual reserved quantity of spare parts is determined according to the planned spare parts demand quantity list, and the real reserved data of spare parts corresponding to the planned spare parts demand quantity list is obtained, and then the corresponding actual reserved quantity of spare parts is obtained according to the real reserved data of spare parts and the virtual reserved quantity of spare parts. The inventory quantity, the corrected quantity and the inventory parameters corresponding to the spare parts list are further obtained, and the fourth input data corresponding to the trained procurement forecast data model is generated according to the actual reserved quantity of spare parts, the inventory quantity, the corrected quantity and the inventory parameters, and then the procurement quantity of the fourth input data can be predicted according to the trained procurement forecast data model to generate the corresponding planned spare parts procurement list. It is realized that according to the actual reserved quantity of spare parts, the inventory quantity, the corrected quantity and the inventory parameters, combined with the trained procurement forecast data model, the procurement quantity can be further accurately predicted to avoid excessive procurement or the procurement quantity cannot meet the actual maintenance needs, thereby improving the maintenance processing efficiency of the nuclear power plant.

[0200] In one embodiment, Fig. 27 As shown, a method for predicting the demand for spare parts of a nuclear power plant is provided, which specifically includes the following steps:

[0201] Step S2702, obtaining the actual spare parts collection data on site, and comparing the planned spare parts demand quantity list with the actual spare parts collection data on site to generate corresponding true positive data, false positive data and false negative data.

[0202] Specifically, by obtaining the actual spare parts collection data on site, the planned spare parts demand quantity list is compared with the actual spare parts collection data on site. Fig.28 As shown in the figure, a comparison diagram of the actual spare parts collection data on site and the planned spare parts demand quantity list is provided. Fig.28It can be seen that by comparing the predicted demand data with the actual collection data, it can be divided into three categories. The first category is true positive data, that is, the predicted demand is the same as the actual collection, the second category is false positive data, that is, there is predicted demand but no actual collection, and the third category is false negative data, that is, there is no predicted demand but there is actual collection. After calculating the true positive data, false positive data, and false negative data, the accuracy and recall rate of the model can be further calculated. Among them, the number of spare parts demand predicted by the planned spare parts demand forecasting model is the predicted demand data, and the number of spare parts actually collected on site is the actual collection data.

[0203] In order to effectively evaluate the pros and cons of different models, a spare parts demand forecasting evaluation method is established, using a quantitative method to simply and objectively evaluate the demand forecasting effect, so that the system can automatically select and push the spare parts demand forecast of the optimal model. The forecasting model of the first category of spare parts demand list and the forecasting model of the second category of spare parts demand list have different structures, so the forecasting effects of the two categories of spare parts models need to be calculated separately.

[0204] Step S2704, based on the true positive data, the false positive data and the false negative data, determine the accuracy and recall rate corresponding to the target demand prediction model.

[0205] Specifically, after calculating the true positive data, false positive data, and false negative data, the accuracy and recall rate corresponding to the target demand prediction model can be further determined based on the true positive data, false positive data, and false negative data.

[0206] in, The accuracy rate represents the proportion of correctly predicted samples to all predicted samples of the model. The value range is [0,1]. The larger the value, the better the prediction ability of the model.

[0207] in, The recall rate represents the proportion of correctly predicted samples to all true value samples. Its value range is [0,1]. The larger the value, the better the model's prediction ability.

[0208] Step S2706, evaluating the effect of the target demand prediction model based on the accuracy and recall rate, and generating corresponding effect evaluation data.

[0209] Specifically, the target demand forecasting model is evaluated based on the accuracy and recall rate, which may be the overall forecasting effect of the target demand forecasting model, thereby obtaining corresponding effect evaluation data.

[0210] Furthermore, the prediction data of Model 1 and Model 2 for the samples are shown in Tables 3 and 4 respectively. Referring to Tables 3 and 4, it can be seen that the accuracy and recall rates of Model 1 and Model 2 corresponding to different overhaul rounds have different or identical data values. For example, the accuracy and recall rates of Model 1 and Model 2 at overhaul 1 are different, while the accuracy rates of Model 1 and Model 2 at overhaul 6 are different, but the recall rates are consistent.

[0211]

[0212] In one embodiment, if Fig.29 As shown in the figure, a grid distribution diagram of the prediction effect is provided, with the recall rate as the horizontal axis and the accuracy rate as the vertical axis, and the following is established: Fig.29 The prediction effect grid distribution diagram is shown in the figure. The higher the recall rate and accuracy, the better the prediction effect of the model. The prediction effect grid distribution diagram is divided into 25% levels, and a total of 7 areas are established, namely A, B, C, D, E, F, and G. In this case, Model 1 accounts for 0% in area A, 0% in area B, 50% in area C, 33.3% in area D, 16.7% in area E, 0% in area F, and 0% in area G. Model 2 accounts for 0% in area A, 0% in area B, 16.7% in area C, 83.3% in area D, 0% in area E, 0% in area F, and 0% in area G. In comprehensive comparison, Model 1 has a better overall prediction effect.

[0213] Furthermore, the prediction effect grid distribution diagram can facilitate users to observe the prediction effects of different models in different application scenarios more intuitively. In practical applications, the average distance of the model from the theoretical optimal value (accuracy 100%, recall 100%) can also be directly calculated. The smaller the average distance, the better the overall effect of the model. Taking Model 1 and Model 2 as examples, the theoretical optimal positions of the overhaul distances of each round in Model 1 are: 0.85, 0.75, 0.87, 0.45, 0.57, 0.56, and the average value is 0.67; similarly, the theoretical optimal positions of the overhaul distances of each round in Model 2 can be calculated as: 0.79, 0.79, 0.84, 0.72, 0.69, 0.59, and the average value is 0.74. Since the average distance of Model 1 is smaller, it means that the overall prediction effect of Model 1 is better.

[0214] In one embodiment, the prediction effect can be comprehensively evaluated from two dimensions: the number of spare parts and the amount of spare parts. The number of spare parts will affect the order quantity of back-end procurement, and the amount of spare parts will affect the inventory control index of spare parts. Among them, the prediction data and the actual use data are shown in Table 5 and Table 6 respectively. With reference to Table 5, the prediction amount of different predicted spare parts, such as spare part 1, spare part 2, ..., spare part 7, can be obtained. According to Table 6, the use amount of the used spare parts, such as spare part 1, spare part 2, ..., spare part 8, can be obtained.

[0215]

[0216] Furthermore, based on the predicted data shown in Table 5 and the actual usage data shown in Table 6, the true positive, false positive and false negative data distribution table shown in Table 7 is further calculated. Referring to Table 7, it can be seen that the number of items and the amount correspond to the corresponding true positive data, false positive data and false negative data respectively.

[0217] Table 7

[0218]

[0219] The calculation process of the number of items-true positive is: among the predicted spare parts, spare parts 1, spare parts 3, and spare parts 5 are the same as the actual ones, so the number of items is 3. The calculation process of the number of items-false positive is: spare parts 6 and spare parts 7 are not in the requisition list, so the number of items is 2. The calculation process of the number of items-false negative is: spare parts 2, spare parts 4, and spare parts 8 are not in the predicted list, so the number of items is 3.

[0220] Similarly, the calculation process of amount-true positive is: 40>20 in spare part 1, the actual amount of use is 20, the actual amount of use in spare part 3 is 20, and the actual amount of use in spare part 5 is 30, then the amount-true positive data is 20+20+30=70. The calculation process of amount-false positive is: 40>20 in spare part 1, the over-prediction amount is 20, spare part 6 is not used, the over-prediction amount is 40, spare part 7 is not used, the over-prediction amount is 50, then the amount-false positive is 20+40+50=110.

[0221] Among them, the calculation process of amount minus false negative is as follows: for spare part 2, there is no prediction, and the under-prediction amount is 20; for spare part 4, there is no prediction, and the under-prediction amount is 30; for spare part 5, the under-prediction amount is 20; for spare part 8, there is no prediction, and the under-prediction amount is 10. Then the amount minus false negative data is 20+30+20+10=80.

[0222] Furthermore, based on the calculated number of items, accuracy of the amount, recall rate, and theoretical optimal distance data, the prediction effect table in Table 8 is obtained. Among them, the accuracy of the number of items is 60%, the recall rate is 50%, and the theoretical optimal distance is 0.64. Similarly, the accuracy of the amount is 38.9%, the recall rate is 46.7%, and the theoretical optimal distance is 0.81.

[0223] Table 8

[0224]

[0225] In this embodiment, by obtaining the actual spare parts collection data on site, and comparing the planned spare parts demand quantity list with the actual spare parts collection data on site, the corresponding true positive data, false positive data and false negative data are generated, and then based on the true positive data, false positive data and false negative data, the accuracy and recall rate corresponding to the target demand forecasting model are determined. Therefore, the effect of the target demand forecasting model can be evaluated according to the accuracy and recall rate, and the corresponding effect evaluation data can be generated. According to the accuracy and recall rate corresponding to the target demand forecasting model, the effect of the target demand forecasting model is evaluated, and the corresponding effect evaluation data is generated. Then, according to the corresponding effect evaluation data, it can be determined whether each model in the model library meets the spare parts demand forecasting requirements of the nuclear power plant, and the model can be updated or adjusted in time, thereby improving the accuracy of the spare parts demand forecasting results.

[0226] It should be understood that, although the steps in each flow chart involved in the above-described embodiment are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a portion of the steps in each flow chart involved in the above-described embodiment may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the steps or stages in other steps.

[0227] In one embodiment, Fig.30 As shown, a nuclear power plant spare parts demand forecasting system is provided, including: a preliminary maintenance project list generation module 3002, a maintenance project screening module 3004, a first-category spare parts demand list determination module 3006, a collection record data update module 3008, a benchmark list establishment module 3010, a target demand forecasting model determination module 3012, a second-category spare parts demand list generation module 3014 and a planned spare parts demand quantity list generation module 3016, wherein:

[0228] The preliminary maintenance item list generating module 3002 is used to determine the corresponding maintenance items according to the maintenance plan of the nuclear power plant, and perform item equivalence processing based on the maintenance items to generate the corresponding preliminary maintenance item list.

[0229] The maintenance item screening module 3004 is used to obtain a maintenance item change rate form corresponding to the preliminary maintenance item list, and screen maintenance items from the preliminary maintenance item list according to the maintenance item change rate form.

[0230] The first category spare parts requirement list determination module 3006 is used to obtain preset constraints corresponding to the maintenance project, and determine the first category spare parts requirement list corresponding to the maintenance project according to the preset constraints and the maintenance standard package data.

[0231] The collection record data updating module 3008 is used to obtain the collection records corresponding to the preliminary maintenance item list, and integrate the collection data based on the collection records to obtain updated collection record data.

[0232] The reference list establishing module 3010 is used to establish a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts according to the updated requisition record data.

[0233] The target demand forecasting model determination module 3012 is used to obtain the maintenance project requisition data corresponding to the first benchmark list and the standard package requisition data corresponding to the second benchmark list, and determine the target demand forecasting model corresponding to the maintenance project based on the maintenance project requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list.

[0234] The second-category spare parts demand list generating module 3014 is used to perform second-category spare parts demand forecasting for the maintenance project based on the target demand forecasting model, and generate a corresponding second-category spare parts demand list.

[0235] The planned spare parts requirement quantity list generating module 3016 is used to generate a planned spare parts requirement quantity list based on the first category spare parts requirement list and the second category spare parts requirement list.

[0236] The above-mentioned nuclear power plant spare parts demand prediction system can accurately determine the different types of spare parts needs of the nuclear power plant during the maintenance process. It can make purchases in advance based on the obtained planned spare parts demand quantity list to avoid spare parts shortages during the maintenance process, so as to prepare sufficient maintenance spare parts in advance to meet maintenance needs and further ensure the stable operation of the nuclear power plant.

[0237] In one embodiment, the first category spare parts requirement list determination module is further used to:

[0238] Obtain the total amount threshold of project spare parts, the long procurement cycle threshold of project spare parts, and standard package data corresponding to the maintenance project; establish a first mapping form between the standard package data and the total amount threshold of project spare parts, and the long procurement cycle threshold of project spare parts; determine the first category spare parts requirement list corresponding to the maintenance project according to each maintenance project, the maintenance project change rate form and the first mapping form.

[0239] In one embodiment, the target demand forecasting model determination module is further used to:

[0240] When it is determined that the maintenance project requisition data meets the first requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the maintenance project and spare parts model; when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data meets the second requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the standard package and spare parts model; when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data does not meet the second requisition data threshold requirement, the target demand forecasting model corresponding to the maintenance project is determined to be the spare parts model.

[0241] In one embodiment, the target demand forecasting model determination module is further used to:

[0242] Determine the corresponding maintenance project category based on the maintenance project requisition data, determine the corresponding standard package category based on the standard package requisition data, and determine the corresponding spare parts category based on the spare parts data; match the corresponding maintenance project and spare parts model from the maintenance project and spare parts model library based on the combined data of the maintenance project category and the spare parts category, and determine it as the target demand forecasting model; or match the corresponding standard package and spare parts model from the standard package and spare parts model library based on the combined data of the standard package category and the spare parts category, and determine it as the target demand forecasting model; or obtain supplier data for providing spare parts of different categories, and determine the corresponding supplier category based on the supplier data; match the corresponding spare parts model from the spare parts model library based on the combined data of the spare parts category and the supplier category, and determine it as the target demand forecasting model.

[0243] In one embodiment, the second type spare parts requirement list generating module is further used to:

[0244] Extract the total number of calls, total number of requisitions, outline change rate, spare parts importance, spare parts unit price and fluctuation coefficient in the first benchmark list, and obtain the first number of calls in this round; obtain the first input data corresponding to the maintenance project and spare parts model based on the total number of calls, total number of requisitions, outline change rate, spare parts importance, spare parts unit price and fluctuation coefficient in the first benchmark list, and the first number of calls in this round; perform demand forecasting on the first input data based on the maintenance project and spare parts model to obtain the corresponding first forecast demand data; extract the total number of calls, total number of requisitions, spare parts importance, spare parts unit price and fluctuation coefficient in the second benchmark list, and obtain the second number of calls in this round; according to the total number of calls, total number of requisitions, spare parts importance, spare parts unit price and fluctuation coefficient in the second benchmark list, and the second call count of this round, to obtain the second input data corresponding to the standard package and spare parts model; perform demand forecast on the second input data according to the standard package and spare parts model, to obtain the corresponding second forecast demand data; extract the total call count, total number of requisitions, importance of spare parts, unit price of spare parts and fluctuation coefficient in the spare parts list, and obtain the third call count of this round; obtain the third input data corresponding to the spare parts model according to the total call count, total number of requisitions, importance of spare parts, unit price of spare parts, fluctuation coefficient and the third call count in this round in the spare parts list; perform demand forecast on the third input data according to the spare parts model, to obtain the corresponding third forecast demand data; generate the corresponding second category spare parts demand list according to the first forecast demand data, the second forecast demand data and the third forecast demand data.

[0245] In one embodiment, a nuclear power plant spare parts demand forecasting system is provided, further comprising a planned spare parts procurement list generation module, which is used to:

[0246] According to the planned spare parts demand quantity list, determine the corresponding virtual reserve quantity of spare parts; obtain the real reserve data of spare parts corresponding to the planned spare parts demand quantity list, and obtain the corresponding actual reserve quantity of spare parts based on the real reserve data of spare parts and the virtual reserve quantity of spare parts; obtain the inventory quantity, the regular order quantity and inventory parameters corresponding to the spare parts list, and generate the fourth input data corresponding to the trained procurement forecast data model based on the actual reserve quantity of spare parts, the inventory quantity, the regular order quantity and the inventory parameters; perform procurement quantity forecast on the fourth input data according to the trained procurement forecast data model, and generate the corresponding planned spare parts procurement list.

[0247] In one embodiment, a nuclear power plant spare parts demand forecasting system is provided, further comprising an effect evaluation data generating module for:

[0248] The actual spare parts usage data on site is obtained, and the planned spare parts demand quantity list is compared with the actual spare parts usage data on site to generate corresponding true positive data, false positive data and false negative data; based on the true positive data, false positive data and false negative data, the accuracy and recall rate corresponding to the target demand forecasting model are determined; according to the accuracy and recall rate, the effect of the target demand forecasting model is evaluated to generate corresponding effect evaluation data.

[0249] In one embodiment, the acquisition record data update module is further used to:

[0250] Obtain the receipt records corresponding to the preliminary maintenance item list, and filter out duplicate spare parts records and replacement spare parts records based on the receipt records; integrate the receipt data based on the duplicate spare parts records, replacement spare parts records and receipt records to obtain updated receipt record data.

[0251] In one embodiment, the preliminary maintenance item list generation module is further used to:

[0252] Obtain the preventive maintenance program of the nuclear power plant, and determine the corresponding nuclear power plant maintenance plan based on the preventive maintenance program of the nuclear power plant; screen the overhaul rounds based on the nuclear power plant maintenance plan, and obtain the maintenance items corresponding to the current overhaul rounds; obtain the functional position of each item and the project text content corresponding to the maintenance items, and perform project equivalence processing based on the functional position of each item and the project text content corresponding to the maintenance items, and generate the corresponding preliminary maintenance item list.

[0253] In one embodiment, Fig.31 As shown, a nuclear power plant spare parts demand forecasting system is provided, referring to Fig.31 It can be seen that the nuclear power plant spare parts demand forecasting system includes a data query processing module 3102, a planned maintenance project processing module 3104, a planned spare parts demand forecasting benchmark database 3106, a spare parts demand quantity forecasting module 3108 and a spare parts procurement quantity forecasting module 3110. Among them:

[0254] The data query processing module 3102 is mainly used to receive and process user query requirements. Users can choose to sort out the planned spare parts demand types, and can query according to spare parts codes, maintenance outline items, standard package numbers and other conditions. They can also choose to query by spare parts packages, power plants, overhaul rounds and other conditions.

[0255] In one embodiment, Fig.32 As shown, a query interface of a nuclear power plant spare parts demand forecasting system is provided. Fig.32It can be seen that in the query interface, the user enters the relevant query conditions and clicks the spare parts annual demand list button, and the system can automatically output the demand quantity of each spare part and the recommended purchase quantity by year; click the spare parts detailed demand list, the system can output the demand data of each maintenance project and each spare part, including overhaul rounds, standard packages and other information. Users can also quickly query the maintenance outline list, standard package list, spare parts requisition information and other data through the system.

[0256] The data query processing module 3102 may convert the query information set by the user into standardized constraint conditions and convey them to the planned maintenance project processing module 3104 . The planned maintenance project processing module 3104 is used to sort out the corresponding maintenance projects.

[0257] The planned maintenance project processing module 3104 sorts out the corresponding planned maintenance projects according to the corresponding query conditions, and outputs the maintenance project list after project equivalence. The user can choose to directly adopt the maintenance project list pushed by the system, or further manually adjust the system-recommended projects to obtain a more accurate maintenance project list.

[0258] Among them, the spare parts demand quantity prediction module 3108 includes two sub-modules: planned Class A spare parts demand prediction and planned Class B spare parts demand prediction. By receiving the planned maintenance project list, reading the correspondence between maintenance projects-standard packages-spare parts in the planned spare parts demand prediction benchmark database 3106, as well as the benchmark list of maintenance projects-spare parts, the benchmark list of standard packages-spare parts, and the benchmark list of spare parts, based on the above data obtained, the demand for planned spare parts is predicted and the planned spare parts demand list is output.

[0259] The spare parts procurement quantity prediction module 3110 receives the planned spare parts demand list, reads the spare parts inventory quantity, ordered quantity, reserved quantity (including real reservations and virtual reservations) and inventory parameters, and further calculates the spare parts procurement application quantity based on the above data received, and outputs a planned spare parts procurement recommendation list.

[0260] In one embodiment, Fig.33 As shown, a spare parts historical use data analysis interface for a nuclear power plant spare parts demand forecasting system is provided. Fig.33 It can be seen that in the spare parts historical usage data analysis interface, users can query spare parts historical usage and guarantee classification data by entering spare parts codes, spare parts packages, power plants and other dimensions, which can be used as a reference for users when manually reviewing the planned spare parts requirements calculated by the system.

[0261] In the above-mentioned nuclear power plant spare parts demand prediction system, a complete nuclear power plant maintenance processing architecture is constructed by setting up a data query processing module, a planned maintenance project processing module, a planned spare parts demand prediction benchmark database, a spare parts demand quantity prediction module and a spare parts procurement quantity prediction module. At the same time, planned spare parts demand prediction and further procurement quantity prediction can be performed, and the different categories of spare parts requirements of the nuclear power plant during the maintenance process can be accurately determined. Procurement can be carried out in advance based on the obtained planned spare parts demand quantity list to avoid spare parts shortages during the maintenance process, so as to prepare sufficient maintenance spare parts in advance to meet maintenance needs and further ensure the stable operation of the nuclear power plant.

[0262] The specific definition of the nuclear power plant spare parts demand forecasting system can be found in the definition of the nuclear power plant spare parts demand forecasting method mentioned above, which will not be repeated here. Each module in the above-mentioned nuclear power plant spare parts demand forecasting system can be implemented in whole or in part through software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0263] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig.34 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as a preliminary maintenance project list, a first-category spare parts demand list, updated requisition record data, a first benchmark list, a second benchmark list, a target demand forecast model, a second-category spare parts demand list and a planned spare parts demand quantity list. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting spare parts demand for a nuclear power plant is implemented.

[0264] Those skilled in the art will understand that Fig.34 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0265] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0266] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0267] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0268] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0269] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for predicting demand for spare parts in a nuclear power plant, characterized in that: The method comprises: Determine the corresponding maintenance items according to the maintenance plan of the nuclear power plant, perform project equivalence processing based on the maintenance items, and generate a corresponding preliminary maintenance item list; Acquire a maintenance item change rate form corresponding to the preliminary maintenance item list, and select maintenance items from the preliminary maintenance item list according to the maintenance item change rate form; Acquire preset constraints corresponding to the maintenance project, and determine a first category spare parts requirement list corresponding to the maintenance project according to the preset constraints and maintenance standard package data; Acquire the collection records corresponding to the preliminary maintenance item list, and integrate the collection data based on the collection records to obtain updated collection record data; Establishing a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts according to the updated requisition record data; Obtaining maintenance item requisition data corresponding to the first benchmark list and standard package requisition data corresponding to the second benchmark list, and determining a target demand forecasting model corresponding to the maintenance item based on the maintenance item requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list; Based on the target demand forecasting model, forecast the demand for second-category spare parts for the maintenance project, and generate a corresponding demand list for second-category spare parts; Summarize and generate a planned spare parts demand quantity list based on the first category spare parts demand list and the second category spare parts demand list; Determining a target demand forecasting model corresponding to the maintenance project based on the maintenance project requisition data, the standard package requisition data, and the spare parts data corresponding to the spare parts list, including: When it is determined that the maintenance project requisition data meets the first requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is the maintenance project and spare parts model; when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data meets the second requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is the standard package and spare parts model; when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data does not meet the second requisition data threshold requirement, it is determined that the target demand forecasting model corresponding to the maintenance project is the spare parts model; or Determine the corresponding maintenance project category according to the maintenance project requisition data, determine the corresponding standard package category according to the standard package requisition data, and determine the corresponding spare parts category according to the spare parts data; match the corresponding maintenance project and spare parts model from the maintenance project and spare parts model library based on the combined data of the maintenance project category and the spare parts category, and determine it as the target demand forecast model; or match the corresponding standard package and spare parts model from the standard package and spare parts model library based on the combined data of the standard package category and the spare parts category, and determine it as the target demand forecast model; or obtain supplier data for providing spare parts of different categories, and determine the corresponding supplier category according to the supplier data; match the corresponding spare parts model from the spare parts model library based on the combined data of the spare parts category and the supplier category, and determine it as the target demand forecast model; The method also includes: obtaining actual on-site spare parts usage data, and comparing the planned spare parts demand quantity list with the actual on-site spare parts usage data to generate corresponding true positive data, false positive data, and false negative data; based on the true positive data, false positive data, and false negative data, determining the accuracy and recall rate corresponding to the target demand forecasting model; and evaluating the effect of the target demand forecasting model based on the accuracy and recall rate to generate corresponding effect evaluation data.

2. The method according to claim 1, characterized in that The preset constraints include: a total amount threshold of project spare parts and a long procurement cycle threshold of project spare parts; the obtaining of the preset constraints corresponding to the maintenance project, and determining a first category spare parts requirement list corresponding to the maintenance project according to the preset constraints and the maintenance standard package data, including: Obtaining a total amount threshold of project spare parts, a long procurement cycle threshold of project spare parts, and standard package data corresponding to the maintenance project; Establishing a first mapping table between the standard package data and the total amount threshold of project spare parts and the long procurement cycle threshold of project spare parts; A first category spare parts requirement list corresponding to the maintenance items is determined according to each of the maintenance items, the maintenance item change rate table and the first mapping table.

3. The method according to claim 1, characterized in that: The step of forecasting the demand for second-category spare parts for the maintenance project based on the target demand forecasting model and generating a corresponding demand list for second-category spare parts includes: Extract the total call times, total requisition quantity, outline change rate, spare parts importance, spare parts unit price and fluctuation coefficient in the first benchmark list, and obtain the first call times of this round; Obtaining first input data corresponding to the maintenance item and the spare parts model according to the total call times, total requisition quantity, outline change rate, spare parts importance, spare parts unit price and fluctuation coefficient in the first benchmark list, and the first call times of this round; According to the maintenance item and spare parts model, performing demand forecasting on the first input data to obtain corresponding first forecast demand data; Extract the total call times, total withdrawal quantity, spare parts importance, spare parts unit price and fluctuation coefficient in the second benchmark list, and obtain the second call times of this round; Obtain second input data corresponding to the standard package and spare parts model according to the total call times, total withdrawal quantity, spare parts importance, spare parts unit price, fluctuation coefficient, and the second call times of this round in the second benchmark list; According to the standard package and spare parts model, performing demand forecasting on the second input data to obtain corresponding second forecast demand data; Extract the total call times, total withdrawal quantity, spare parts importance, spare parts unit price and fluctuation coefficient in the spare parts list, and obtain the third call times in this round; Obtaining third input data corresponding to the spare parts model according to the total call times, the total withdrawal quantity, the importance of spare parts, the unit price of spare parts, the fluctuation coefficient, and the third call times in this round in the spare parts list; According to the spare parts model, performing demand forecasting on the third input data to obtain corresponding third forecast demand data; A corresponding second category spare parts demand list is generated based on the first predicted demand data, the second predicted demand data and the third predicted demand data.

4. The method according to claim 1 or 2, characterized in that: After the planned spare parts requirement quantity list is generated by summarizing the first category spare parts requirement list and the second category spare parts requirement list, the method further includes: Determine the corresponding virtual reserved quantity of spare parts according to the planned spare parts demand quantity list; Acquire the spare parts real reservation data corresponding to the planned spare parts demand quantity list, and obtain the corresponding actual spare parts reservation quantity according to the spare parts real reservation data and the spare parts virtual reservation quantity; Obtaining the inventory quantity, the regular order quantity and the inventory parameters corresponding to the spare parts list, and generating fourth input data corresponding to the trained procurement forecast data model according to the actual spare parts reserved quantity, the inventory quantity, the regular order quantity and the inventory parameters; According to the trained procurement forecast data model, the procurement quantity of the fourth input data is predicted to generate a corresponding planned spare parts procurement list.

5. The method according to claim 1 or 2, characterized in that: The obtaining of the collection records corresponding to the preliminary maintenance item list, and integrating the collection data based on the collection records to obtain updated collection record data, includes: Obtaining the requisition records corresponding to the preliminary maintenance item list, and filtering the requisition records to obtain duplicate spare parts records and replacement spare parts records; Based on the duplicate spare parts record, the replacement spare parts record and the collection record, collection data is integrated to obtain updated collection record data.

6. The method according to claim 1 or 2, characterized in that: Determining corresponding maintenance items according to the maintenance plan of the nuclear power plant, performing item equivalence processing based on the maintenance items, and generating a corresponding preliminary maintenance item list includes: Obtain the preventive maintenance program of the nuclear power plant, and determine the corresponding nuclear power plant maintenance plan based on the preventive maintenance program of the nuclear power plant; Performing overhaul round screening based on the nuclear power plant maintenance plan to obtain maintenance items corresponding to the current overhaul round; The functional positions and textual contents of each item corresponding to the maintenance item are obtained, and item equivalence processing is performed based on the functional positions and textual contents of each item corresponding to the maintenance item to generate a corresponding preliminary maintenance item list.

7. A nuclear power plant spare parts demand forecasting system, characterized in that: The system comprises: A preliminary maintenance item list generation module is used to determine corresponding maintenance items according to the maintenance plan of the nuclear power plant, and perform item equivalence processing based on the maintenance items to generate a corresponding preliminary maintenance item list; A maintenance item screening module, used to obtain a maintenance item change rate form corresponding to the preliminary maintenance item list, and screen maintenance items from the preliminary maintenance item list according to the maintenance item change rate form; A first-category spare parts requirement list determination module, used to obtain preset constraints corresponding to the maintenance project, and determine the first-category spare parts requirement list corresponding to the maintenance project according to the preset constraints and the maintenance standard package data; A collection record data updating module, used to obtain the collection record corresponding to the preliminary maintenance item list, and integrate the collection data based on the collection record to obtain updated collection record data; A reference list establishment module, used to establish a first reference list between maintenance items and spare parts, and a second reference list between standard packages and spare parts according to the updated requisition record data; a target demand forecast model determination module, configured to obtain maintenance item requisition data corresponding to the first benchmark list and standard package requisition data corresponding to the second benchmark list, and determine a target demand forecast model corresponding to the maintenance item based on the maintenance item requisition data, the standard package requisition data and the spare parts data corresponding to the spare parts list; A second-category spare parts demand list generation module, configured to perform a second-category spare parts demand forecast for the maintenance project based on the target demand forecast model, and generate a corresponding second-category spare parts demand list; A planned spare parts requirement quantity list generation module, used for summarizing and generating a planned spare parts requirement quantity list based on the first category spare parts requirement list and the second category spare parts requirement list; The target demand forecasting model determination module is further used to: when it is determined that the maintenance project requisition data meets the first requisition data threshold requirement, determine that the target demand forecasting model corresponding to the maintenance project is a maintenance project and spare parts model; when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data meets the second requisition data threshold requirement, determine that the target demand forecasting model corresponding to the maintenance project is a standard package and spare parts model; when it is determined that the maintenance project requisition data does not meet the first requisition data threshold requirement and the standard package requisition data does not meet the second requisition data threshold requirement, determine that the target demand forecasting model corresponding to the maintenance project is a spare parts model; The target demand forecasting model determination module is further used to: determine the corresponding maintenance project category according to the maintenance project requisition data, determine the corresponding standard package category according to the standard package requisition data, and determine the corresponding spare parts category according to the spare parts data; match the corresponding maintenance project and spare parts model from the maintenance project and spare parts model library based on the combined data of the maintenance project category and the spare parts category, and determine it as the target demand forecasting model; or match the corresponding standard package and spare parts model from the standard package and spare parts model library based on the combined data of the standard package category and the spare parts category, and determine it as the target demand forecasting model; or obtain supplier data for providing spare parts of different categories, and determine the corresponding supplier category according to the supplier data; match the corresponding spare parts model from the spare parts model library based on the combined data of the spare parts category and the supplier category, and determine it as the target demand forecasting model; The system also includes an effect evaluation data generation module, which is used to: obtain actual spare parts collection data on site, and compare the planned spare parts demand quantity list with the actual spare parts collection data on site to generate corresponding true positive data, false positive data and false negative data; based on the true positive data, false positive data and false negative data, determine the accuracy and recall rate corresponding to the target demand forecasting model; based on the accuracy and recall rate, evaluate the effect of the target demand forecasting model to generate corresponding effect evaluation data.

8. The system according to claim 7, characterized in that The preset constraint conditions include: a total amount threshold of project spare parts and a long procurement cycle threshold of project spare parts; the first type of spare parts requirement list determination module is also used to: Obtain the total amount threshold of project spare parts, the long procurement cycle threshold of project spare parts, and standard package data corresponding to the maintenance project; establish a first mapping form between the standard package data and the total amount threshold of project spare parts, and the long procurement cycle threshold of project spare parts; and determine a first category spare parts requirement list corresponding to the maintenance project based on each of the maintenance projects, the maintenance project change rate form, and the first mapping form.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.