Electric power material demand prediction method and system based on distributed warehouse, medium and processor

Abstract

The invention discloses an electric power material demand prediction method and system based on a distributed warehouse, a medium and a processor, and relates to the technical field of warehouse management, and the method comprises the steps: obtaining the historical data of electric power materials in the distributed warehouse, and classifying the electric power materials according to the historical data; performing trend analysis on the classified various electric power materials to obtain trend analysis results of the various electric power materials; according to a trend analysis result, constructing a first demand prediction model of various electric power materials meeting the trend, and constructing a second demand prediction model of various electric power materials not meeting the trend; and selecting the first demand prediction model or the second demand prediction model to perform demand prediction of the corresponding electric power materials according to the electric power materials selected by the distributed warehouse. According to the method, the electric power materials are classified according to the electric power materials, the trend result is obtained after classification, the corresponding prediction model is selected according to the trend result for prediction, and accurate demand prediction for trended and non-trended demands is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to a method, system, medium and processor for predicting power material demand based on distributed warehouses. Background Art

[0002] At present, in order to build a unified, strong and intelligent power grid and realize intensive management of power materials, modern means, optimal configuration and maximum benefits, power grid companies are required to establish a unified material management system. The core of the system construction is to improve the level of engineering and operation and maintenance material management and reduce the existing inventory costs while meeting a certain inventory service level.

[0003] The power grid proposes to realize the intensive management of power materials, establish a unified material management system, and reorganize the logistics service department to form a logistics service center, which is responsible for the logistics distribution services of power materials, including multiple power material warehouses and multiple power material distribution transit. There are difficulties in the management of the above-mentioned distributed warehouses, but the essential difficulty lies in the difficulty of demand forecasting. The demand has obvious seasonal characteristics, and the forecast is relatively accurate and reliable, but for random demand, it is difficult to get accurate results by relying on traditional forecasting methods. This is because the consumption regularity of most power materials is not significant, and the occurrence of demand is discrete and intermittent. There are few historical data available, and there are many factors that affect the occurrence of demand.

[0004] In view of this, a method, system, medium and processor for predicting power material demand based on distributed warehouses are needed. Summary of the invention

[0005] In view of the problem that the demand for materials is irregular and difficult to predict in the prior art, the present invention provides a method, system, medium and processor for predicting the demand for power materials based on distributed warehouses, which can select the corresponding prediction model for prediction according to the trend results of power materials, and realize accurate demand prediction for those with and without trends. The specific technical scheme is as follows:

[0006] A method for predicting power material demand based on distributed warehouses, comprising:

[0007] Obtain historical data of power materials in distributed warehouses and classify power materials based on the historical data;

[0008] Conduct trend analysis on various types of classified power materials to obtain trend analysis results of various types of power materials;

[0009] According to the trend analysis results, a first demand forecasting model for various types of power materials that meet the trend is constructed, and a second demand forecasting model for various types of power materials that do not meet the trend is constructed;

[0010] According to the electric power materials selected by the distributed warehouse, the first demand forecasting model or the second demand forecasting model is selected to perform demand forecasting for the corresponding electric power materials.

[0011] Preferably, the properties of the power materials in the distributed warehouse are obtained, and the power materials are classified according to the properties;

[0012] The demand quantity characteristics in the historical data of the electric power materials in the distributed warehouse are obtained, and the electric power materials are divided into large-category electric power materials and detailed-category electric power materials according to the demand quantity characteristics.

[0013] Preferably, the trend analysis of the classified various types of electric power materials is performed to obtain trend analysis results of the various types of electric power materials, including:

[0014] Obtain the historical demand data of various types of power materials sorted by time after classification;

[0015] Obtain demand data distribution curves for various types of power materials based on the historical demand data;

[0016] The exponential smoothing method is used to analyze the long-term trend, seasonal trend and cycle trend of various types of power materials on the demand data distribution curve to obtain trend analysis results of various types of power materials.

[0017] Preferably, the step of constructing a first demand forecasting model for various types of electric power materials that conform to the trend according to the trend analysis results includes:

[0018] According to the trend analysis results, the first demand forecasting model for various types of power materials is constructed by exponential smoothing method for the power materials that meet the trend; the first demand forecasting model is expressed as:

[0019] Y t =f(S t , T t , C t )

[0020] Among them, Y t is the demand forecast value when the time series is t; t is time; S t is the seasonal trend factor; C t It is a cyclical trend factor.

[0021] Preferably, the constructing of a second demand forecasting model for various types of electric power materials that do not conform to the trend according to the trend analysis results includes:

[0022] According to the trend analysis results, the power materials that do not conform to the trend are used through grey prediction or neural network to build the second demand forecast model for various types of power materials.

[0023] Preferably, it also includes:

[0024] Use seasonal factors to modify the second demand forecasting model.

[0025] Preferably, the forecasting of power material demand by using the first demand forecasting model and the second demand forecasting model includes:

[0026] Obtain the trend analysis results of the electric power materials. If the electric power materials are in line with the trend, use the first demand forecasting model to forecast the demand for the electric power materials. Otherwise, use the second demand forecasting model to forecast the demand for the electric power materials.

[0027] A distributed warehouse-based power material demand forecasting system, applied to the aforementioned distributed warehouse-based power material demand forecasting method, comprises:

[0028] A classification unit, used to obtain historical data of power materials in the distributed warehouse and classify the power materials according to the historical data;

[0029] The trend analysis unit is used to perform trend analysis on various types of classified power materials and obtain trend analysis results of various types of power materials;

[0030] A forecasting model building unit, used to build a first demand forecasting model for various types of power materials that meet the trend and a second demand forecasting model for various types of power materials that do not meet the trend according to the trend analysis result;

[0031] The forecast model selection unit is used to select the first demand forecast model or the second demand forecast model to perform demand forecasting for the corresponding power materials according to the power materials selected by the distributed warehouse.

[0032] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the aforementioned method for predicting power material demand based on distributed warehouses.

[0033] A processor is used to run a program, wherein the program executes the aforementioned method for predicting power material demand based on distributed warehouses when running.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] The invention discloses a method for predicting demand for electric power materials based on distributed warehouses, which obtains historical data of electric power materials in distributed warehouses and classifies electric power materials according to the historical data; performs trend analysis on various types of classified electric power materials to obtain trend analysis results for various types of electric power materials; constructs a first demand prediction model for various types of electric power materials that meet the trend, and constructs a second demand prediction model for various types of electric power materials that do not meet the trend according to the trend analysis results; selects a first demand prediction model or a second demand prediction model to predict demand for corresponding electric power materials according to the electric power materials selected by the distributed warehouse. The invention classifies electric power materials according to the electric power materials, obtains trend results after classification, selects corresponding prediction models to predict according to the trend results, and realizes accurate demand prediction for those with and without trends. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0037] Figure 1 This is a flow chart of a method for predicting power material demand based on distributed warehouses according to the present invention.

[0038] Figure 2 The present invention is a flow chart of another embodiment of a method for predicting power material demand based on distributed warehouses.

[0039] Figure 3 This is a schematic diagram of a distributed warehouse-based power material demand forecasting system of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0041] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0042] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0043] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0044] The first characteristic of the material demand of power grid enterprises is that the demand types are diverse: the company's materials are mainly managed based on the material categories, and the material categories have expanded from power grid construction materials to comprehensive materials covering infrastructure, overhaul, technical transformation, operation and maintenance, and auxiliary materials. The material demand types are numerous and complex. The second characteristic is that the demand trend is changeable: the demand for the same material category is different in different time periods, and the demand trends of different categories of materials in the same time period are also ever-changing. This application should select different models for prediction based on different trends in the demand for power materials to ensure the accuracy of key material predictions.

[0045] See the following examples Figures 1 to 3 .

[0046] The present application embodiment provides a method for predicting power material demand based on a distributed warehouse, including:

[0047] Step S01: Obtain historical data of power materials in the distributed warehouse, and classify the power materials according to the historical data;

[0048] The historical data of electric power materials in the distributed warehouse includes demand statistics of various electric power materials. The demand data can obtain demand quantity characteristics in the historical data of electric power materials in the distributed warehouse, and divide the electric power materials into major categories of electric power materials and minor categories of electric power materials according to the demand quantity characteristics.

[0049] For example, primary equipment, secondary equipment, communication equipment, building materials, accessories and low-voltage electrical appliances can be classified into major categories of power materials; AC transformers, AC circuit breakers, reactors and combined electrical appliances are classified into minor categories of power materials.

[0050] Step S02: performing trend analysis on the classified various types of electric power materials to obtain trend analysis results of the various types of electric power materials;

[0051] By analyzing the historical data of major categories of materials, the demand trend of major categories of materials is determined:

[0052] The trends of primary equipment, secondary equipment, communication equipment, instruments and meters, installation materials, auxiliary equipment and facilities, and hardware materials are in line with the trend line chart of seasonal index changes; the annual material demand for materials such as building materials, accessories, and low-voltage electrical appliances does not conform to the trend changes.

[0053] Step S03: constructing a first demand forecasting model for various types of electric power materials that meet the trend, and constructing a second demand forecasting model for various types of electric power materials that do not meet the trend, according to the trend analysis result;

[0054] If the trend line graph of seasonal index changes is in line with the trend, seasonal predictive prediction methods can be used; if it does not conform to the seasonal change trend or long-term trend, it can be predicted through gray model prediction or application of network model.

[0055] Specifically, the first demand forecasting model for various types of power materials that conform to the trend is constructed according to the trend analysis result, including:

[0056] (1) According to the trend analysis results, the first demand forecasting model of various types of power materials is constructed by exponential smoothing method for the power materials that meet the trend; the first demand forecasting model is expressed as:

[0057] Y t =f(S t , T t , C t )

[0058] Among them, Y t is the demand forecast value when the time series is t; t is time; S t is the seasonal trend factor; C t It is a cyclical trend factor.

[0059] There are usually two types of relationships between the variable factors in the seasonal forecasting method time series: the additive model and the multiplicative model. The additive model assumes that the time series is composed of the addition of three components, and each component is independent of each other and has no interaction. The multiplicative model assumes that the time series is composed of the multiplication of three components, and each component is interdependent. In this embodiment, the multiplicative model is selected for prediction. Therefore, the first demand forecasting model is expressed as:

[0060] Y t =S t *T t *C t

[0061] (2) According to the trend analysis results, the power materials that do not conform to the trend are used to construct the second demand forecasting model for various types of power materials through grey prediction or neural network.

[0062] The modeling process of the grey prediction model is essentially to accumulate and produce irregular original data (r-AGO), and then re-model the generated series with strong regularity. The data obtained from the generated model is then used to obtain the restored model through cumulative subtraction (r-IAGO), and then the restored model is used as the prediction model.

[0063] BP network is a multi-layer network composed of an input layer, several hidden layers and an output layer. The basic idea is: the input signal reaches the output layer after being acted on by the function of the hidden layer. If the actual output signal does not match the expected value, the network will perform back propagation and repeatedly correct the weights and closed values ​​between layers until the expected output is minimized, that is, the global error of the network is minimized. The network training uses the improved BP algorithm function traingdx, based on MATLAB's BP neural network training, parameter setting and data input.

[0064] Step S04: According to the electric power materials selected by the distributed warehouse, the first demand forecasting model or the second demand forecasting model is selected to perform demand forecasting for the corresponding electric power materials.

[0065] Specifically, the power material demand forecasting by using the first demand forecasting model and the second demand forecasting model includes:

[0066] Obtain the trend analysis results of the electric power materials. If the electric power materials are in line with the trend, use the first demand forecasting model to forecast the demand for the electric power materials. Otherwise, use the second demand forecasting model to forecast the demand for the electric power materials.

[0067] The invention discloses a method for predicting demand for electric power materials based on distributed warehouses, which obtains historical data of electric power materials in distributed warehouses and classifies electric power materials according to the historical data; performs trend analysis on various types of classified electric power materials to obtain trend analysis results for various types of electric power materials; constructs a first demand prediction model for various types of electric power materials that meet the trend, and constructs a second demand prediction model for various types of electric power materials that do not meet the trend according to the trend analysis results; selects a first demand prediction model or a second demand prediction model to predict demand for corresponding electric power materials according to the electric power materials selected by the distributed warehouse. The invention classifies electric power materials according to the electric power materials, obtains trend results after classification, selects corresponding prediction models to predict according to the trend results, and realizes accurate demand prediction for those with and without trends.

[0068] Specifically, in a preferred implementation of the present application, the trend analysis of the classified various types of electric power materials is performed to obtain trend analysis results of the various types of electric power materials, including:

[0069] Step S21, obtaining the historical demand data of various types of power materials sorted by time after classification;

[0070] According to the major and minor categories of power materials, the historical demand data is sorted in chronological order (such as monthly, quarterly or annual) to ensure the time series of the data. The data comes from various channels such as the power company's material management system, financial statements, and procurement records.

[0071] Step S22, obtaining demand data distribution curves of various types of power materials according to the historical demand data;

[0072] By drawing the demand data distribution curve, the historical changes in the demand for power materials can be intuitively displayed, and by observing the shape and trend of the curve, the overall pattern of changes in material demand over time can be understood.

[0073] Step S23: using the exponential smoothing method to analyze the long-term trend, seasonal trend and periodic trend of various types of electric power materials on the demand data distribution curve to obtain trend analysis results of various types of electric power materials.

[0074] In this embodiment, in response to the uncertainty of demand, the future demand is predicted with the help of historical data and smoothing data methods, which can reduce the losses caused by demand fluctuations and improve the flexibility of the entire supply chain. In addition, the demand for power materials is affected by seasonality, has certain seasonal factors, and presents a certain fluctuation trend. The exponential smoothing method can be fitted based on the historical material demand data, thereby discovering the changing trend of material demand, and has good adaptability to the prediction of power materials. The exponential smoothing method has several different forms:

[0075] Single exponential smoothing is for series without trend and seasonality;

[0076] The quadratic exponential smoothing rule is for series with trend but no seasonality;

[0077] Triple exponential smoothing is used for series that have both trend and seasonality.

[0078] The trend analysis results of various types of power materials are obtained through exponential smoothing method.

[0079] Specifically, in a preferred implementation of the present application, constructing the first demand forecasting model further includes:

[0080] Use seasonal factors to modify the second demand forecasting model.

[0081] In the specific implementation, the quarter is used as the time observation unit, the seasonal factors in the time series are targeted, the predictions of the gray model and the neural network model are improved and supplemented, and the seasonal factors of the power materials in the power grid company in the demand forecast are extracted. Through the extraction of seasonal factors, it is checked that the gray model and the neural network model adjusted by the seasonal factors can improve the fitting accuracy of the forecast model, so as to correct the first demand forecast model. The following formula is used to adjust the seasonal coefficient:

[0082]

[0083] In the formula, S i is the seasonal coefficient of the i-th quarter, is the average value of the i-th quarter, It is the average of all quarters.

[0084] In order to address the shortcomings of grey forecasting and neural network forecasting that seasonal factors are not fully considered, the material demand forecasting model with a deviation of more than 15% and an annual cyclical change in the demand trend curve is corrected by taking quarter as the time observation unit.

[0085] The embodiment of the present application further provides a distributed warehouse-based power material demand forecasting system, which is applied to the aforementioned distributed warehouse-based power material demand forecasting method, including:

[0086] A classification unit, used to obtain historical data of power materials in the distributed warehouse and classify the power materials according to the historical data;

[0087] The trend analysis unit is used to perform trend analysis on various types of classified power materials and obtain trend analysis results of various types of power materials;

[0088] A forecasting model building unit, used to build a first demand forecasting model for various types of power materials that meet the trend and a second demand forecasting model for various types of power materials that do not meet the trend according to the trend analysis result;

[0089] The forecast model selection unit is used to select the first demand forecast model or the second demand forecast model to perform demand forecasting for the corresponding power materials according to the power materials selected by the distributed warehouse.

[0090] The functions of each unit in the system are the same as those explained in the embodiment of a method for predicting power material demand based on a distributed warehouse, and the technical effects are also the same and will not be repeated here.

[0091] An embodiment of the present application also provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the aforementioned distributed warehouse-based power material demand forecasting method.

[0092] The technical effect of this embodiment is the same as the technical effect of an embodiment of a method for predicting power material demand based on a distributed warehouse, and will not be repeated here.

[0093] The present invention can be used in many general or special computer system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like.

[0094] An embodiment of the present application also provides a processor, which is used to run a program, wherein the program executes the aforementioned distributed warehouse-based power material demand forecasting method when running.

[0095] The technical effect of this embodiment is the same as the technical effect of an embodiment of a method for predicting power material demand based on a distributed warehouse, and will not be repeated here.

[0096] The processor in this embodiment may be a central processing unit (CPU), a controller, a microcontroller, or other data processing chips.

[0097] The technical effect of this embodiment is the same as the technical effect of the method in embodiment xxxxx, and will not be repeated here.

[0098] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0099] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0100] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0101] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0102] 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 by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for predicting power material demand based on distributed warehouses, characterized in that: include: Obtain historical data of power materials in distributed warehouses and classify power materials based on the historical data; Conduct trend analysis on various types of classified power materials to obtain trend analysis results of various types of power materials; According to the trend analysis results, a first demand forecasting model for various types of power materials that meet the trend is constructed, and a second demand forecasting model for various types of power materials that do not meet the trend is constructed; According to the electric power materials selected by the distributed warehouse, the first demand forecasting model or the second demand forecasting model is selected to perform demand forecasting for the corresponding electric power materials.

2. The method for predicting power material demand based on distributed warehouses according to claim 1 is characterized in that: Obtain the attributes of power materials in the distributed warehouse and classify the power materials according to the attributes; The demand quantity characteristics in the historical data of the electric power materials in the distributed warehouse are obtained, and the electric power materials are divided into large-category electric power materials and detailed-category electric power materials according to the demand quantity characteristics.

3. The method for predicting power material demand based on distributed warehouses according to claim 2 is characterized in that: The trend analysis of the classified various types of power materials is performed to obtain trend analysis results of the various types of power materials, including: Obtain the historical demand data of various types of power materials sorted by time after classification; Obtain demand data distribution curves for various types of power materials based on the historical demand data; The exponential smoothing method is used to analyze the long-term trend, seasonal trend and cycle trend of various types of power materials on the demand data distribution curve to obtain trend analysis results of various types of power materials.

4. The method for predicting power material demand based on distributed warehouses according to claim 3 is characterized in that: The first demand forecasting model for various types of power materials that conform to the trend is constructed according to the trend analysis result, including: According to the trend analysis results, the first demand forecasting model for various types of power materials is constructed by exponential smoothing method for the power materials that meet the trend; the first demand forecasting model is expressed as: Y t =f(S t ,T t ,C t ) Among them, Y t is the demand forecast value when the time series is t; t is time; S t is the seasonal trend factor; C t It is a cyclical trend factor.

5. The method for predicting power material demand based on distributed warehouses according to claim 1 is characterized in that: The second demand forecasting model for various types of power materials that do not conform to the trend is constructed according to the trend analysis results, including: According to the trend analysis results, the power materials that do not conform to the trend are used through grey prediction or neural network to build the second demand forecast model for various types of power materials.

6. The method for predicting power material demand based on distributed warehouses according to claim 5 is characterized in that: Also includes: The second demand forecasting model is modified using seasonal factors.

7. The method for predicting power material demand based on distributed warehouses according to claim 5 is characterized in that: The power material demand forecasting by using the first demand forecasting model and the second demand forecasting model includes: Obtain the trend analysis results of the electric power materials. If the electric power materials are in line with the trend, use the first demand forecasting model to forecast the demand for the electric power materials. Otherwise, use the second demand forecasting model to forecast the demand for the electric power materials.

8. A distributed warehouse-based power material demand forecasting system, characterized in that: The method for predicting power material demand based on a distributed warehouse as applied to any one of claims 1 to 7 comprises: A classification unit, used to obtain historical data of power materials in the distributed warehouse and classify the power materials according to the historical data; The trend analysis unit is used to perform trend analysis on various types of classified power materials and obtain trend analysis results of various types of power materials; A forecasting model building unit, used to build a first demand forecasting model for various types of power materials that meet the trend and a second demand forecasting model for various types of power materials that do not meet the trend according to the trend analysis result; The forecast model selection unit is used to select the first demand forecast model or the second demand forecast model to perform demand forecasting for the corresponding power materials according to the power materials selected by the distributed warehouse.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for predicting power material demand based on distributed warehouses as described in any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the method for predicting power material demand based on distributed warehouses as described in any one of claims 1 to 7.

Images