Spare part management method and system of new energy station
By constructing a demand forecast model, the problem of spare parts demand forecasting in the management of spare parts in new energy stations has been solved, accurate prediction and reasonable management of spare parts are achieved, inventory costs are reduced, and spare parts shortages and backlogs are avoided.
Patent Information
- Application Number
- CN202410997842.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-19
AI Technical Summary
The equipment of new energy stations is complex, resulting in a wide variety of spare parts, different specifications, and uncertain fault prediction, making it difficult to accurately predict demand and demand time, making it difficult to avoid inventory backlog and waste of spare parts supply.
Build a demand forecast model, predict stocking demand by obtaining equipment data, operation and maintenance data and replacement records, and store the data in the management library. The management platform reasonably manages spare parts based on the database data to avoid shortages and backlogs.
Accurate prediction and reasonable management of spare parts for new energy stations, reduce inventory costs, and avoid unnecessary shortages and backlogs of spare parts.
Smart Images

Figure CN118982307A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of spare parts management for new energy stations, and in particular to a spare parts management method and management system for new energy stations. Background Art
[0002] The management of spare parts for new energy stations is of great significance to ensure the normal operation of the stations and reduce maintenance costs. Specifically, by replacing faulty spare parts in a timely manner, the normal operation of the equipment can be quickly restored, downtime can be reduced, and power generation efficiency can be improved. In addition, reasonable spare parts management can reduce unnecessary inventory backlogs and waste and reduce inventory costs. At the same time, by purchasing and reserving key spare parts in advance, emergency procurement and transportation costs caused by spare parts shortages can be avoided.
[0003] However, the equipment and systems used in new energy stations are complex and diverse, resulting in a wide variety of spare parts with different specifications. In addition, due to the uncertainty of the operating status and failure occurrence of new energy equipment, it is difficult to accurately predict the demand for spare parts and the time of demand.
[0004] Currently, it is a difficult problem to ensure the supply of spare parts while avoiding inventory backlogs and waste. Summary of the invention
[0005] The purpose of this application is to solve how to reasonably manage spare parts for new energy stations. Therefore, this application provides a spare parts management method and management system for new energy stations. By constructing a demand forecasting model, the future stocking demand of each demand side is accurately predicted, and the first information and stocking demand corresponding to the demand side are stored in the management library, so that the management platform can reasonably manage spare parts according to the data of the management library, avoid unnecessary inventory backlogs and waste while avoiding shortages of key spare parts, and reduce inventory costs.
[0006] The present application provides a method for managing spare parts of a new energy station, including:
[0007] S1. The management platform obtains first information of the demand side; the first information includes equipment data and operation and maintenance data of the demand side, and replacement records of spare parts;
[0008] S2. The demand forecasting model predicts the stocking demand of the demand side according to the first information; the stocking demand includes the type of spare parts, demand time and demand quantity;
[0009] S3, storing the first information and stocking demand corresponding to the demand side in the management library, and the management platform manages the spare parts according to the management library data;
[0010] S4. The demand side sends a replacement requirement for spare parts to the management platform; the management platform provides the demand side with corresponding spare parts according to the replacement requirement.
[0011] By adopting the above technical solution, through constructing a demand forecasting model, the future stocking demand of each demand side can be accurately predicted, and the first information and stocking demand corresponding to the demand side can be stored in the management library, so that the management platform can reasonably manage spare parts according to the data of the management library, avoid unnecessary inventory backlogs and waste while avoiding shortages of key spare parts, and reduce inventory costs.
[0012] In some embodiments, in step S2, the specific steps of constructing the demand forecasting model include:
[0013] S21, obtaining training data, where the training data includes historical data, test data, and simulation data;
[0014] in,
[0015] The historical data includes first information of multiple new energy stations; the equipment data of the first information includes equipment type, equipment quantity, equipment parameters, equipment operation strategy and equipment operation environment; the operation and maintenance data includes the number of operation and maintenance personnel, operation and maintenance costs, operation and maintenance methods and the geographical location of the demand side;
[0016] The test data includes the test environment data of the equipment in each new energy station during the test process and the corresponding test result data;
[0017] The simulation data includes simulation result data of a simulation performed in a simulation model based on simulation parameters; wherein the simulation parameters include actual parameters of equipment in the new energy station, an operating environment of the equipment, and an operating strategy of the equipment;
[0018] S22, extracting feature information from the training data;
[0019] S23. Using a learning model of a recurrent neural network and performing training based on the training data to construct the demand forecasting model.
[0020] By adopting the above technical solution, the acquired training data includes not only the historical data of the demand side, but also the test data and simulation data of the equipment; by acquiring the test data and simulation data, the sample size of the training data is greatly enriched, which can improve the prediction effect of the demand prediction model. At the same time, the test data and simulation data are highly generalized, which can greatly increase the diversity of the training data, avoid the overfitting of the demand prediction model that is only suitable for the historical data training of the demand side, and thus can greatly improve the accuracy of the demand prediction model for the new demand side prediction.
[0021] In some embodiments, step S22 specifically includes:
[0022] S22.1. Clean the training data to remove missing values and outliers.
[0023] S22.2, extracting feature data from historical data and test data in the training data, and based on the relationship between simulation parameters and simulation result data in the simulation data, extracting associated feature data between actual parameters of the device, operating environment, operating strategy and simulation result data through feature engineering methods;
[0024] S22.3. Construct new feature data, including the number of equipment maintained per capita and equipment operation efficiency;
[0025] S22.4. Use data transformation technology to normalize the feature data to obtain the feature information of the training data.
[0026] By adopting the above technical solution, the computational complexity of the demand forecasting model training process can be reduced through the processing of training data, and characteristic data of per capita maintenance equipment quantity and equipment operation efficiency can be constructed based on the original data, which may more accurately reflect the inherent laws and patterns between the demand-side stocking demand and the first information, thereby improving the classification or regression ability of the demand forecasting model and improving the accuracy of the demand forecasting model's predictions.
[0027] In some embodiments, step S23 specifically includes:
[0028] S23.1. The theoretical stocking requirements of each new energy station based on the equipment data in the first information of each new energy station, the test data and the simulation data;
[0029] S23.2, setting the theoretical stocking demand as the initial benchmark value of the learning model;
[0030] S23.3. Using the training data, training the learning model to construct the demand forecasting model;
[0031] The simulated data accounts for 20-40% of the training data.
[0032] By adopting the above technical solution, a benchmark value can be constructed for the demand forecasting model by calculating the theoretical stocking demand, so that the demand forecasting model can be adjusted and corrected on the basis of the benchmark value. While ensuring the training effect of the demand forecasting model, the amount of calculation for training the demand forecasting model is greatly reduced, thereby reducing the cost of building the demand forecasting model.
[0033] In some embodiments, in step S3, the management method of the management platform includes:
[0034] S31, determining the sales profit, daily reserve cost and importance coefficient of each type of spare parts, and calculating the first storage duration and the second storage duration of each type of spare parts according to the sales profit, the reserve cost and the importance coefficient; the first storage duration is less than the second storage duration;
[0035] S32. Classify and manage the stocking requirements in the management library according to the demand time, and reserve the corresponding types and quantities of spare parts for which the demand time is less than the first storage time; reserve the corresponding types and quantities of spare parts for which the importance coefficient is greater than the first threshold and the demand time is between the first storage time and the second storage time; and do not reserve spare parts for which the demand time is greater than the second storage time;
[0036] S33. Perform long-term reserve for spare parts whose importance coefficient is greater than a second threshold, wherein the second threshold is greater than the first threshold.
[0037] By adopting the above technical solution, by obtaining the first storage time and the second storage time of each spare part, the storage economic benefits of each spare part can be shown; that is, when the spare parts are stored between the first storage time and the second storage time, a balance between avoiding spare parts shortage and avoiding spare parts backlog can be achieved, and the longer the spare parts are stored, the lower the risk of shortage and the higher the risk of backlog.
[0038] Therefore, the first storage time and the second storage time of each spare part can be reasonably allocated according to the importance coefficient of each spare part, thereby reducing the cost of spare parts management.
[0039] In some embodiments, the first storage duration=(importance coefficient*the sales profit) / daily reserve cost;
[0040] Second storage duration = (safety factor * importance factor * sales profit) / daily reserve cost;
[0041] Wherein, the safety factor is determined based on the accuracy of the demand forecasting model, and the safety factor is 1.1 to 1.5.
[0042] In some embodiments, the importance coefficient is determined as follows:
[0043] Determine the types, stocking time and replacement time of the spare parts; the stocking time is the time from the time when the spare parts that need to be stockpiled are determined to the time when the spare parts are stockpiled; the replacement time is the time from when the spare parts are provided by the management platform to when the spare parts are replaced by the demand side;
[0044] Determining the critical value according to the type of the spare parts;
[0045] Normalizing the importance value, the stocking time, and the replacement time;
[0046] A weighted sum is performed on the normalized importance value, the stocking time, and the replacement time to obtain a weight value;
[0047] Finally, the weight values of all spare parts are normalized to obtain the importance coefficients corresponding to the spare parts, and the importance coefficients are made to be in the interval [0, 1].
[0048] In some embodiments, the step S5 is also included:
[0049] S51, the management platform records the time of obtaining the replacement demand, the type and quantity of the replacement parts, and stores them in the verification library;
[0050] S52, comparing the data of the verification library with the stocking demand of the demand forecasting model, and continuously optimizing the demand forecasting model;
[0051] S53: storing the replacement requirements with the comparison difference greater than the third threshold into an abnormal database, and manually analyzing the abnormal database.
[0052] By adopting the above technical solution, the training data of the demand forecasting model can be continuously optimized and enriched during long-term operation, thereby improving the prediction accuracy and rationality of the demand forecasting model.
[0053] The embodiment of the present application also provides a management system, including a warehouse, a management platform, and a plurality of demand terminals communicatively connected to the management platform;
[0054] The warehouse is used to store spare parts;
[0055] The management platform uses the management method described in any of the above embodiments to manage the spare parts in the warehouse.
[0056] By adopting the above technical solution, through constructing a demand forecasting model, the future stocking demand of each demand side can be accurately predicted, and the first information and stocking demand corresponding to the demand side can be stored in the management library, so that the management platform can reasonably manage the spare parts in the warehouse according to the data of the management library, avoid unnecessary inventory backlogs and waste while avoiding shortages of key spare parts, and reduce inventory costs.
[0057] Other features and corresponding beneficial effects of the present invention are described in the latter part of the specification, and it should be understood that at least part of the beneficial effects become obvious from the description in the specification of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1A schematic diagram of the steps of a spare parts management method for a new energy station provided in an embodiment of the present application. DETAILED DESCRIPTION
[0059] The following specific embodiments illustrate the implementation of the present invention, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Although the description of the present invention will be introduced in conjunction with the preferred embodiment, this does not mean that the features of this invention are limited to this implementation. On the contrary, the purpose of introducing the invention in conjunction with the implementation is to cover other options or modifications that may extend based on the claims of the present invention. In order to provide a deep understanding of the present invention, the following description will include many specific details. The present invention can also be implemented without using these details. In addition, in order to avoid confusion or blurring the focus of the present invention, some specific details will be omitted in the description. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.
[0060] It should be noted that in this specification, similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.
[0061] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0062] In the description of the present invention, it should be noted that the terms "first", "second", and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the invention of this application, unless otherwise specified, "plurality" means two or more.
[0063] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.
[0064] Figure 1 A schematic diagram of the steps of a spare parts management method for a new energy station provided in an embodiment of the present application.
[0065] like Figure 1 As shown, the embodiment of the present application provides a spare parts management method for a new energy station, including:
[0066] S1. The management platform obtains the first information of the demand side; the first information includes the equipment data and operation and maintenance data of the demand side, and the replacement record of spare parts;
[0067] S2. The demand forecasting model forecasts the stocking demand of the demand side according to the first information; the stocking demand includes the type of spare parts, demand time and demand quantity;
[0068] S3. The first information and stocking demand corresponding to the demand side are stored in the management library, and the management platform manages the spare parts according to the management library data;
[0069] S4. The demand side sends a replacement request for spare parts to the management platform; the management platform provides the corresponding spare parts to the demand side according to the replacement request.
[0070] In one embodiment, the management platform can directly obtain equipment data and operation and maintenance data, as well as replacement records of spare parts in the process of building cooperation with the demand side; and the replacement records of spare parts can be blank.
[0071] In other implementations, the device data on the demand side can be obtained from the sales manufacturer of the device itself, or the device model query.
[0072] It should be noted that the demand forecasting model requires a large amount of relevant training data to ensure the effect of the demand forecasting model after training; however, in the spare parts management needs, the historical data that the demand side can provide is often limited, and the historical data on the demand side can often only represent its own characteristics and laws;
[0073] Therefore, in one embodiment, in step S2, the specific steps of constructing the demand forecasting model include:
[0074] S21, obtaining training data, where the training data includes historical data, test data, and simulation data;
[0075] in,
[0076] The historical data includes the first information of multiple new energy stations; the equipment data of the first information includes the equipment type, equipment quantity, equipment parameters, equipment operation strategy and equipment operation environment; the operation and maintenance data includes the number of operation and maintenance personnel, operation and maintenance costs, operation and maintenance methods and the geographical location of the demand side;
[0077] The test data includes the test environment data of the equipment in each new energy station during the test process and the corresponding test result data;
[0078] The simulation data includes simulation result data of the simulation performed in the simulation model based on the simulation parameters; wherein the simulation parameters include actual parameters of the equipment in the new energy station, the operating environment of the equipment, and the operating strategy of the equipment;
[0079] S22, extracting feature information from training data;
[0080] S23. Use a recurrent neural network learning model and perform training based on training data to build a demand forecasting model.
[0081] Among them, the acquired training data includes not only the historical data of the demand side, but also the test data and simulation data of the equipment. Since the test data and simulation data are often universal and generalized, the data of various parameters of the equipment and the test process are often unified, so that the demand forecasting model can grasp the intrinsic relationship between the equipment and different demand sides. This improves the generalization ability of the training data and avoids the overfitting of the demand forecasting model that is only trained with the historical data of the demand side, thereby greatly improving the accuracy of the demand forecasting model's prediction of the new demand side.
[0082] Furthermore, by setting a large number of parameters and then obtaining simulation data through simulation software, the sample size of the training data can be guaranteed, so that the demand forecasting model can obtain the expected forecasting ability, increasing the accuracy and reliability of the demand forecasting model's predictions.
[0083] In one embodiment, step S22 specifically includes:
[0084] S22.1. Clean the training data to remove missing values and outliers.
[0085] S22.2. Extract feature data from historical data in training data and test data, and based on the relationship between simulation parameters and simulation result data in simulation data, extract associated feature data between actual parameters of equipment, operating environment, operating strategy and simulation result data through feature engineering method;
[0086] S22.3. Construct new feature data, including the number of equipment maintained per capita and equipment operation efficiency;
[0087] S22.4. Use data transformation technology to normalize the feature data to obtain the feature information of the training data.
[0088] Through feature extraction, the complex information in the original data is converted into features that are easier to understand and explain, thereby enhancing the interpretability of the model and reducing the amount of computation in the demand forecasting model training process; at the same time, based on the original data, feature data of per capita maintenance equipment and equipment operating efficiency are constructed, so that the demand forecasting model can more accurately reflect the inherent laws and patterns between the demand-side stocking needs and the first information, and when the amount of training data is not rich enough, the classification or regression ability of the demand forecasting model is improved, thereby improving the accuracy of the demand forecasting model's predictions.
[0089] In one embodiment, step S23 specifically includes:
[0090] S23.1. The theoretical stocking requirements of each new energy station based on the equipment data, test data and simulation data in the first information of each new energy station;
[0091] S23.2, setting the theoretical stocking demand as the initial benchmark value of the learning model;
[0092] S23.3, using training data to train the learning model and build a demand forecasting model;
[0093] The proportion of simulated data in training data is 20-40%.
[0094] Through the above method, based on equipment data, test data and simulation data, the theoretical stocking demand of each new energy station can be estimated or predicted by manual calculation (for example, based on the average service life of components in the equipment, the replacement cycle of components can be calculated, so as to deduce a rough stocking demand).
[0095] This can provide a general direction for the demand forecasting model in advance and build a benchmark value for the demand forecasting model, so that the demand forecasting model can be adjusted and corrected based on the benchmark value. While ensuring the training effect of the demand forecasting model, the amount of computational effort in training the demand forecasting model is greatly reduced, thereby reducing the cost of building the demand forecasting model.
[0096] In other alternative embodiments, the proportion of simulation data in the training data can be freely set according to different requirements for the model, the amount of historical data, etc.; for example, when the amount of historical data is sufficient, the simulation data can be reduced so that the proportion of simulation data is less than 20%; when the requirements for the prediction effect are low and the amount of historical data is insufficient, the proportion of simulation data in the training data can also be greater than 40%.
[0097] In one embodiment, in step S3, the management method of the management platform includes:
[0098] S31, determining the sales profit, daily reserve cost and importance coefficient of each type of spare parts, and calculating the first storage time and the second storage time of each type of spare parts according to the sales profit, the reserve cost and the importance coefficient; the first storage time is less than the second storage time;
[0099] S32. Classify and manage the stocking requirements in the management library according to the demand time, and reserve the corresponding types and quantities of spare parts for which the demand time is less than the first storage time; reserve the corresponding types and quantities of spare parts for which the importance coefficient is greater than the first threshold and the demand time is between the first storage time and the second storage time; and do not reserve spare parts for which the demand time is greater than the second storage time;
[0100] S33. Long-term reserve is performed for spare parts whose importance coefficient is greater than a second threshold, wherein the second threshold is greater than the first threshold.
[0101] It should be noted that the purpose of calculating the first storage time and the second storage time of various spare parts is to ensure that when a certain spare part is stored for the first storage time, it can be ensured that in the long-term management process, the spare part is not likely to be in short supply, and the storage cost is the lowest; when a certain spare part is stored for the second storage time, it can be ensured that in the long-term management process, the storage of the spare part begins to be backlogged.
[0102] Thus, by obtaining the first storage time and the second storage time of each spare part, the storage economic benefits of each spare part can be shown; that is, when the spare parts are stored between the first storage time and the second storage time, a balance between avoiding spare parts shortage and avoiding spare parts backlog can be achieved, and the longer the spare parts are stored, the lower the risk of shortage and the higher the risk of backlog.
[0103] Furthermore, according to the importance coefficient of each spare part, the first storage time and the second storage time of each spare part are reasonably allocated, thereby reducing the cost of spare part management.
[0104] It should be noted that the importance coefficient is not only the importance of the component relative to the equipment, but also the importance relative to the storage cost. When the storage cost of a component is very low (for example, screws and nuts of different models), and the replacement or demand frequency is very low, its importance coefficient is relatively high, that is, it can be greater than the second threshold, and it can be stored for a long time.
[0105] In one embodiment, the important coefficient is determined as follows:
[0106] Determine the types of spare parts, preparation time, and replacement time; the preparation time is the time from the time when the spare parts that need to be reserved are determined to the time when the spare parts are reserved; the replacement time is the time from the time when the spare parts are provided by the management platform to the time when the spare parts are replaced on the demand side;
[0107] Determine the importance value based on the type of spare parts. The type of spare parts can be used to determine the frequency of replacement and the importance to the equipment itself based on experience and test data, thereby determining its approximate importance value.
[0108] Normalize the importance value, preparation time, and replacement time;
[0109] The weight value is obtained by weighted summing the normalized importance value, stocking time and replacement time;
[0110] Finally, the weight values of all spare parts are normalized to obtain the importance coefficients corresponding to the spare parts, and the importance coefficients are made to be in the interval [0,1].
[0111] It should be noted that since the determination of important coefficients involves multiple dimensions, the accuracy requirement for determining a single dimension is relatively low, which can greatly reduce the complexity of management and reduce management costs.
[0112] In one embodiment, the first storage duration = (importance coefficient*sales profit) / daily reserve cost;
[0113] Second storage duration = (safety factor * importance factor * sales profit) / daily reserve cost;
[0114] Among them, the safety factor is determined based on the accuracy of the demand forecasting model, and the safety factor is 1.1 to 1.5.
[0115] It should be noted that the safety factor can be adjusted and changed according to the warehouse storage resources, the reliability of the model prediction capability, and the importance factor of the spare parts. For example, if the model prediction capability is reliable, the safety factor can be relatively low; and the safety factors of different spare parts can be the same or different.
[0116] In one embodiment, the step S5 is also included:
[0117] S51. The management platform records the time of obtaining the replacement demand, the type and quantity of the replacement spare parts, and stores them in the verification library;
[0118] S52, comparing the data of the verification library with the stocking demand of the demand forecasting model, and continuously optimizing the demand forecasting model;
[0119] S53: storing the replacement requirements with the comparison difference greater than the third threshold into an abnormal database, and manually analyzing the abnormal database.
[0120] By adopting the above technical solution, the training data of the demand forecasting model can be continuously optimized and enriched during long-term operation, thereby improving the prediction accuracy and rationality of the demand forecasting model.
[0121] The embodiment of the present application also provides a management system, including a warehouse, a management platform, and a plurality of demand terminals that are communicatively connected to the management platform;
[0122] The warehouse is used to store spare parts;
[0123] The management platform adopts the management method of any of the above embodiments for the spare parts in the warehouse.
[0124] By adopting the above technical solution, through constructing a demand forecasting model, the future stocking demand of each demand side can be accurately predicted, and the first information and stocking demand corresponding to the demand side can be stored in the management library, so that the management platform can reasonably manage the spare parts in the warehouse according to the data of the management library, avoid unnecessary inventory backlogs and waste while avoiding shortages of key spare parts, and reduce inventory costs.
[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for managing spare parts of a new energy station, characterized in that: include: S1. The management platform obtains the first information from the demand side; The first information includes equipment data and operation and maintenance data of the demand side, and replacement records of spare parts; S2. The demand forecasting model predicts the stocking demand of the demand side according to the first information; the stocking demand includes the type of spare parts, demand time and demand quantity; S3, storing the first information and stocking demand corresponding to the demand side in the management library, and the management platform manages the spare parts according to the management library data; S4. The demand side sends a replacement requirement for spare parts to the management platform; the management platform provides the demand side with corresponding spare parts according to the replacement requirement.
2. The management method according to claim 1, characterized in that: In step S2, the specific steps of constructing the demand forecasting model include: S21, obtaining training data, where the training data includes historical data, test data, and simulation data; in, The historical data includes first information of multiple new energy stations; the equipment data of the first information includes equipment type, equipment quantity, equipment parameters, equipment operation strategy and equipment operation environment; the operation and maintenance data includes the number of operation and maintenance personnel, operation and maintenance costs, operation and maintenance methods and the geographical location of the demand side; The test data includes the test environment data of the equipment in each new energy station during the test process and the corresponding test result data; The simulation data includes simulation result data of a simulation performed in a simulation model based on simulation parameters; wherein the simulation parameters include actual parameters of equipment in the new energy station, an operating environment of the equipment, and an operating strategy of the equipment; S22, extracting feature information from the training data; S23. Using a learning model of a recurrent neural network and performing training based on the training data to construct the demand forecasting model.
3. The management method according to claim 2, characterized in that: Step S22 specifically includes: S22.
1. Clean the training data to remove missing values and outliers. S22.2, extracting feature data from historical data and test data in the training data, and based on the relationship between simulation parameters and simulation result data in the simulation data, extracting associated feature data between actual parameters of the device, operating environment, operating strategy and simulation result data through feature engineering methods; S22.
3. Construct new feature data, including the number of equipment maintained per capita and equipment operation efficiency; S22.
4. Use data transformation technology to normalize the feature data to obtain the feature information of the training data.
4. The management method according to claim 2, characterized in that: The specific method of step S23 includes: S23.
1. The theoretical stocking requirements of each new energy station based on the equipment data in the first information of each new energy station, the test data and the simulation data; S23.2, setting the theoretical stocking demand as the initial benchmark value of the learning model; S23.
3. Using the training data, training the learning model to construct the demand forecasting model; The simulated data accounts for 20-40% of the training data.
5. The management method according to claim 1, characterized in that: In step S3, the management method of the management platform includes: S31, determining the sales profit, daily reserve cost and importance coefficient of each type of spare parts, and calculating the first storage duration and the second storage duration of each type of spare parts according to the sales profit, the reserve cost and the importance coefficient; the first storage duration is less than the second storage duration; S32. Classify and manage the stocking requirements in the management library according to the demand time, and reserve the corresponding types and quantities of spare parts for which the demand time is less than the first storage time; reserve the corresponding types and quantities of spare parts for which the importance coefficient is greater than the first threshold and the demand time is between the first storage time and the second storage time; and do not reserve spare parts for which the demand time is greater than the second storage time; S33. Perform long-term reserve for spare parts whose importance coefficient is greater than a second threshold, wherein the second threshold is greater than the first threshold.
6. The management method according to claim 5, characterized in that: First storage duration = (importance coefficient * sales profit) / daily storage cost; Second storage duration = (safety factor * importance factor * sales profit) / daily reserve cost; Wherein, the safety factor is determined based on the accuracy of the demand forecasting model, and the safety factor is 1.1 to 1.
5.
7. The management method according to claim 5, characterized in that: The important coefficient is determined as follows: Determine the types, stocking time and replacement time of the spare parts; the stocking time is the time from the time when the spare parts that need to be stockpiled are determined to the time when the spare parts are stockpiled; the replacement time is the time from when the spare parts are provided by the management platform to when the spare parts are replaced by the demand side; Determining the critical value according to the type of the spare parts; Normalizing the importance value, the stocking time, and the replacement time; A weighted sum is performed on the normalized importance value, the stocking time, and the replacement time to obtain a weight value; Finally, the weight values of all spare parts are normalized to obtain the importance coefficients corresponding to the spare parts, and the importance coefficients are made to be in the interval [0, 1].
8. The management method according to claim 1, characterized in that: The step S5 is also included: S51, the management platform records the time of obtaining the replacement demand, the type and quantity of the replacement parts, and stores them in the verification library; S52, comparing the data of the verification library with the stocking demand of the demand forecasting model, and continuously optimizing the demand forecasting model; S53: storing the replacement requirements with the comparison difference greater than the third threshold into an abnormal database, and manually analyzing the abnormal database.
9. A management system, characterized in that: It includes a warehouse, a management platform, and a plurality of demand terminals that are communicatively connected to the management platform; The warehouse is used to store spare parts; The management platform adopts the management method according to any one of claims 1 to 8 to manage the spare parts in the warehouse.
