Wind time series prediction method, device and equipment based on multiple features

Abstract

The present application relates to a method, device and equipment for predicting wind time series based on multiple features. The method comprises: analyzing the historical data of wind time series to obtain multiple attributes of wind; the attributes include trend, periodicity and sensitivity to major meteorological events; modeling is performed using an additive decomposition model based on trend, periodicity and sensitivity to major meteorological events to obtain an initial prediction model; the initial prediction model includes trend terms, periodic terms and climate terms; modeling is performed on the trend terms, periodic terms and climate terms respectively to obtain models corresponding to the trend terms, periodic terms and climate terms; substituting the models corresponding to the trend terms, periodic terms and climate terms into the initial model to obtain a wind time series prediction model; and using the wind time series prediction model to predict the wind time series to be predicted. The use of this method can improve the interpretability of the model.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, computer equipment and storage medium for predicting wind time series based on multiple features. Background Art

[0002] The uncertainty of wind resources has an adverse impact on the stable operation of the power grid. Wind speed prediction can effectively improve the safety and reliability of wind power generation systems.

[0003] The division of wind speed time scale is of great significance for wind speed and wind power prediction. At present, there is no strict standard for the division of prediction time scale, which can be adjusted appropriately according to the needs of the prediction model. Soman et al. divided the wind speed prediction time scale into ultra-short-term 0-30min, short-term 30-360min (6h), medium-term 6-24h (1d), long-term 24-168h (7d) or longer time scales. Short-term and ultra-short-term wind speed prediction plays an important role in stabilizing power quality and coordinating wind turbine control; medium-term wind speed prediction can guide the grid connection and off-grid status of wind turbines; long-term wind speed prediction can be used to guide wind farm operation and maintenance scheduling, evaluate the feasibility of construction, etc. Among them, long-term wind speed prediction needs to start from the perspective of weather and climate, comprehensively consider relevant factors such as temperature, atmospheric pressure, humidity, as well as the climate characteristics of the prediction location and the impact of global climate change.

[0004] The medium-term wind speed time series forecast can ignore the impact of global climate change and local climate characteristics on wind energy resources and regard the wind speed data as a pure time series.

[0005] However, currently, for the prediction of wind data, the commonly used models are time series analysis models, machine learning models and artificial neural network models. In recent years, due to the rise of artificial intelligence research, machine learning models and artificial neural network models have achieved good results in the prediction of general data sets, but these methods have poor interpretability and it is difficult to integrate some data features into the model. Summary of the invention

[0006] Based on this, it is necessary to provide a wind time series prediction method, device, computer equipment and storage medium based on multiple features that can improve the interpretability of the model in response to the above technical problems.

[0007] A wind time series prediction method based on multiple features, the method comprising:

[0008] Obtain the wind time series to be predicted and the historical data of the wind time series;

[0009] Analyze the historical data of wind time series to obtain multiple attributes of wind; the attributes include trend, periodicity and sensitivity to major meteorological events;

[0010] According to the trend, periodicity and sensitivity to major meteorological events, an additive decomposition model is used to build a model to obtain an initial prediction model; the initial prediction model includes trend terms, periodic terms and climate terms;

[0011] Model the trend term, cycle term and climate term respectively, and obtain the corresponding models of the trend term, cycle term and climate term;

[0012] Substitute the models corresponding to the trend term, cycle term and climate term into the initial model to obtain the wind time series prediction model;

[0013] The wind time series prediction model is used to predict the wind time series to be predicted.

[0014] In one embodiment, an additive decomposition model is used to model the trend, periodicity and sensitivity to major meteorological events to obtain an initial prediction model, including:

[0015] According to the trend, periodicity and sensitivity to major meteorological events, the additive decomposition model is used to build the initial forecast model:

[0016]

[0017] Among them, s(t) represents the period term, g(t) represents the trend term, c(t) represents the climate term, ε t Represents the residual term.

[0018] In one embodiment, the trend item, the cycle item and the climate item are modeled respectively to obtain the corresponding models of the trend item, the cycle item and the climate item, including:

[0019] The trend item is modeled using a linear model to obtain a model corresponding to the trend item;

[0020] The wind speed period and Fourier series are used to build a model, and the model corresponding to the periodic term is obtained;

[0021] The climate term is modeled according to the climate impact factor and the discount factor to obtain the model corresponding to the climate term.

[0022] In one embodiment, a linear model is used to model the trend item, and a model corresponding to the trend item is obtained, including:

[0023] The trend item is modeled using a linear model, and the model corresponding to the trend item is obtained as follows:

[0024] g1(t)=k g·(t-t0)+g0

[0025] Among them, t0 is the initial time of the wind field of interest, t is the final time of the wind field of interest, g0 is the wind speed at the initial time, k g Indicates the rate at which the trend item changes.

[0026] In one embodiment, the wind speed period and Fourier series are used to perform modeling to obtain a model corresponding to the periodic term, including:

[0027] Using the wind speed period and Fourier series to build a model, the model corresponding to the periodic term is:

[0028]

[0029] Among them, T i Indicates the period of wind speed, D k 、E k Represents different parameters to be fitted, and N1 is the number of truncated terms in the Fourier series.

[0030] In one embodiment, the climate term is modeled according to the climate impact factor and the discount factor to obtain a model corresponding to the climate term, including:

[0031] The climate term is modeled according to the climate impact factor and discount factor, and the model corresponding to the climate term is obtained as follows:

[0032]

[0033] Among them, q i and l i Represents different climate impact factors, γ i represents the discount factor, I j (t) represents the indicator function, N2 is the number of cutoff terms, k c represents the rate of change of the climate term, and b represents the residual term in the climate term.

[0034] In one embodiment, the models corresponding to the trend term, the cycle term and the climate term are substituted into the initial model to obtain a wind time series prediction model, including:

[0035] Substituting the models corresponding to the trend term, cycle term and climate term into the initial model, the wind time series prediction model is obtained as follows:

[0036]

[0037] Among them, α i represents the discount factor of the trend term, β i represents the discount factor for the climate term.

[0038] A wind time series prediction device based on multiple features, the device comprising:

[0039] The data preprocessing module is used to obtain the wind time series to be predicted and the historical data of the wind time series; analyze the historical data of the wind time series to obtain multiple attributes of the wind; the attributes include trend, periodicity and sensitivity to major meteorological events;

[0040] The initial forecast model modeling module is used to model the initial forecast model using an additive decomposition model based on the trend, periodicity and sensitivity to major meteorological events; the initial forecast model includes a trend term, a period term and a climate term;

[0041] The wind time series prediction model modeling module is used to model the trend item, the cycle item and the climate item respectively, and obtain the models corresponding to the trend item, the cycle item and the climate item; the models corresponding to the trend item, the cycle item and the climate item are substituted into the initial model to obtain the wind time series prediction model.

[0042] The prediction module is used to predict the wind time series to be predicted by using the wind time series prediction model.

[0043] 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:

[0044] Obtain the wind time series to be predicted and the historical data of the wind time series;

[0045] Analyze the historical data of wind time series to obtain multiple attributes of wind; the attributes include trend, periodicity and sensitivity to major meteorological events;

[0046] According to the trend, periodicity and sensitivity to major meteorological events, an additive decomposition model is used to build a model to obtain an initial prediction model; the initial prediction model includes trend terms, periodic terms and climate terms;

[0047] Model the trend term, cycle term and climate term respectively, and obtain the corresponding models of the trend term, cycle term and climate term;

[0048] Substitute the models corresponding to the trend term, cycle term and climate term into the initial model to obtain the wind time series prediction model;

[0049] The wind time series prediction model is used to predict the wind time series to be predicted.

[0050] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0051] Obtain the wind time series to be predicted and the historical data of the wind time series;

[0052] Analyze the historical data of wind time series to obtain multiple attributes of wind; the attributes include trend, periodicity and sensitivity to major meteorological events;

[0053] According to the trend, periodicity and sensitivity to major meteorological events, an additive decomposition model is used to build a model to obtain an initial prediction model; the initial prediction model includes trend terms, periodic terms and climate terms;

[0054] Model the trend term, cycle term and climate term respectively, and obtain the corresponding models of the trend term, cycle term and climate term;

[0055] Substitute the models corresponding to the trend term, cycle term and climate term into the initial model to obtain the wind time series prediction model;

[0056] The wind time series prediction model is used to predict the wind time series to be predicted.

[0057] The above-mentioned wind time series prediction method, device, computer equipment and storage medium based on multiple features first obtain the wind time series to be predicted and the historical data of the wind time series; analyze the historical data of the wind time series to obtain multiple attributes of the wind; the attributes include trend, periodicity and sensitivity to major meteorological events; based on the trend, periodicity and sensitivity to major meteorological events, an additive decomposition model is used to model to obtain an initial prediction model; the initial prediction model includes trend items, periodic items and climate items; the trend items, periodic items and climate items are modeled respectively to obtain models corresponding to the trend items, periodic items and climate items; the models corresponding to the trend items, periodic items and climate items are substituted into the initial model to obtain a wind time series prediction model; the wind time series prediction model is used to predict the wind time series to be predicted. The present invention analyzes the historical data of the wind time series, establishes an initial prediction model according to the obtained multiple attributes of the wind, and then successively establishes corresponding attribute models according to the attributes of the wind, and integrates the model containing the attributes into the initial prediction model. The obtained wind time series prediction model incorporates multiple attributes and has stronger interpretability, thereby improving the accuracy of wind prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic diagram of a flow chart of a wind time series prediction method based on multiple features in one embodiment;

[0059] Figure 2 is a structural block diagram of a wind time series prediction device based on multiple features in one embodiment;

[0060] Figure 3 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0061] 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.

[0062] In one embodiment, Figure 1 As shown, a wind time series prediction method based on multiple features is provided, comprising the following steps:

[0063] Step 102, obtaining the wind time series to be predicted and the historical data of the wind time series; analyzing the historical data of the wind time series to obtain multiple attributes of the wind; the attributes include trend, periodicity and sensitivity to major meteorological events.

[0064] The historical data of wind time series refers to the wind time series data that existed before the wind time series to be predicted. By analyzing the historical data of wind time series, we can know that although the wind speed fluctuates every year, there is no significant change in the wind speed data in the same period of different years, and the wind speed distribution in adjacent years is similar. From the perspective of periodicity, wind speed has a certain periodicity. From the perspective of climate influence, as one of the manifestations of climate, wind speed is affected by major climate events. For example, under the background of global warming, when a typhoon passes, if the weather is fine before the passage and the wind speed is small, the invasion of the outer cloud system of the typhoon will increase the wind speed. As the typhoon approaches, the wind speed will become larger and larger, and the closer to the center, the greater the wind speed. Therefore, the wind speed is greatly affected by major climate events.

[0065] Step 104, using an additive decomposition model to build a model based on trend, periodicity and sensitivity to major meteorological events to obtain an initial prediction model; the initial prediction model includes a trend term, a period term and a climate term.

[0066] For the prediction of wind power, we know that it has three characteristics: periodicity, trend, and sensitivity to major meteorological events. Therefore, through time series decomposition analysis, modeling can be carried out, and the item that affects the period can be made the periodic item, the item that affects the development trend can be made the trend item, and the item affected by major climate events can be made the climate item. The additive decomposition model is used to obtain the initial prediction model for predicting wind power.

[0067] Step 106, modeling the trend item, the cycle item and the climate item respectively to obtain the models corresponding to the trend item, the cycle item and the climate item; substituting the models corresponding to the trend item, the cycle item and the climate item into the initial model to obtain the wind time series prediction model.

[0068] Substituting the models corresponding to the trend term, cycle term, and climate term into the initial model, the resulting wind time prediction model incorporates the relevant attributes of wind, is more interpretable, and is more accurate when making wind predictions.

[0069] Step 108: forecasting the wind time series to be predicted using the wind time series forecasting model.

[0070] In the above-mentioned wind time series prediction method based on multiple features, first, the wind time series to be predicted and the historical data of the wind time series are obtained; the historical data of the wind time series are analyzed to obtain multiple attributes of the wind; the attributes include trend, periodicity and sensitivity to major meteorological events; an additive decomposition model is used to model the trend, periodicity and sensitivity to major meteorological events to obtain an initial prediction model; the initial prediction model includes trend items, periodic items and climate items; the trend items, periodic items and climate items are modeled respectively to obtain models corresponding to the trend items, periodic items and climate items; the models corresponding to the trend items, periodic items and climate items are substituted into the initial model to obtain a wind time series prediction model; the wind time series prediction model is used to predict the wind time series to be predicted. The present invention analyzes the historical data of the wind time series, establishes an initial prediction model according to the obtained multiple attributes of the wind, and then successively establishes corresponding attribute models according to the attributes of the wind, and integrates the model containing the attributes into the initial prediction model. The obtained wind time series prediction model incorporates multiple attributes of the wind, has stronger interpretability, and thus improves the accuracy of wind prediction.

[0071] In one embodiment, an additive decomposition model is used to model the trend, periodicity and sensitivity to major meteorological events to obtain an initial prediction model, including:

[0072] According to the trend, periodicity and sensitivity to major meteorological events, the additive decomposition model is used to build the initial forecast model:

[0073]

[0074] Among them, s(t) represents the period term, g(t) represents the trend term, c(t) represents the climate term, ε t,i Represents the residual term.

[0075] In a specific embodiment, the wind speed is obviously affected by major climate events, and its overall trends in different years are roughly similar. In summary, a time series model of wind speed f(t) is established as an initial prediction model.

[0076] In one embodiment, the trend item, the cycle item and the climate item are modeled respectively to obtain the corresponding models of the trend item, the cycle item and the climate item, including:

[0077] The trend item is modeled using a linear model to obtain a model corresponding to the trend item;

[0078] The wind speed period and Fourier series are used to build a model, and the model corresponding to the periodic term is obtained;

[0079] The climate term is modeled according to the climate impact factor and the discount factor to obtain the model corresponding to the climate term.

[0080] In one embodiment, a linear model is used to model the trend item, and a model corresponding to the trend item is obtained, including:

[0081] The trend item is modeled using a linear model, and the model corresponding to the trend item is obtained as follows:

[0082] g1(t)=k g ·(t-t0)+g0

[0083] Among them, t0 is the initial time of the wind field of interest, t is the final time of the wind field of interest, g0 is the wind speed at the initial time, k g Indicates the rate at which the trend item changes.

[0084] For wind speed, the variation trend in different years is relatively stable. When establishing its trend term, this application adopts a linear model for modeling.

[0085] In one embodiment, the wind speed period and Fourier series are used to perform modeling to obtain a model corresponding to the periodic term, including:

[0086] Using the wind speed period and Fourier series to build a model, the model corresponding to the periodic term is:

[0087]

[0088] Among them, T i Indicates the period of wind speed, D k 、E k Represents different parameters to be fitted, and N1 is the number of truncated terms in the Fourier series.

[0089] After knowing the period of wind speed, the present invention uses truncated Fourier series as the model of its periodic term. The truncated Fourier series has a strong ability to express the period and can enhance the interpretability of the model.

[0090] In one embodiment, the climate term is modeled according to the climate impact factor and the discount factor to obtain a model corresponding to the climate term, including:

[0091] The climate term is modeled according to the climate impact factor and discount factor, and the model corresponding to the climate term is obtained as follows:

[0092]

[0093] Among them, q i and l i Represents different climate impact factors, γ i represents the discount factor, I j (t) represents the indicator function, N2 is the number of cutoff terms, k c represents the rate of change of the climate term, and b represents the residual term in the climate term.

[0094] In a specific embodiment, the impact of climate on wind speed is often reflected in a large turning point in wind speed. For example, under the background of global warming, when a typhoon passes through, if the weather is fine before the passage and the wind speed is low, the invasion of the outer cloud system of the typhoon will increase the wind speed. As the typhoon approaches, the wind speed will increase, and the closer to the center, the greater the wind speed. Therefore, when establishing the climate term function in the present invention, the impact of climate can be established in the form of a turning point based on the trend term. For the wind speed prediction model, let the start time of climate impact be t i , the climate impact ends at t i+k , the indicator function is

[0095]

[0096] Let the climate impact factor be q i ,l i , which affect the slope and intercept of the line respectively. Climate impacts usually weaken over time, and can be discounted by adding a discount factor This feature is reflected in the way.

[0097] In summary, the following climate terms can be established:

[0098]

[0099] Then, by adding the climate impact factor and discount factor 0<γ<1 at the climate time node, the model can reflect the rule that the climate impact usually weakens over time.

[0100] In one embodiment, the models corresponding to the trend term, the cycle term and the climate term are substituted into the initial model to obtain a wind time series prediction model, including:

[0101] Substituting the models corresponding to the trend term, cycle term and climate term into the initial model, the wind time series prediction model is obtained as follows:

[0102]

[0103] Among them, α i represents the discount factor of the trend term, β irepresents the discount factor for the climate term.

[0104] In a specific embodiment, when the period parameter T i , climate impact factor q i ,l i is the climate event factor that affects the wind speed at the most recent time ti, and the discount factor γ i The introduction of makes this climate event decay over time. When the climate influence time period [t i ,t i+j ]When , Stan's L-BFGS can be used to find the maximum likelihood estimate of the model, and then the model fitting can be completed. The wind time series prediction model can be used to predict wind time series.

[0105] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0106] In one embodiment, Figure 2 As shown, a wind time series prediction device based on multiple features is provided, including: a data preprocessing module 202, an initial prediction model building module 204, a wind time series prediction model building module 206 and a prediction module, wherein:

[0107] The data preprocessing module 202 is used to obtain the wind time series to be predicted and the historical data of the wind time series; analyze the historical data of the wind time series to obtain multiple attributes of the wind; the attributes include trend, periodicity and sensitivity to major meteorological events;

[0108] The initial prediction model modeling module 204 is used to model the initial prediction model using an additive decomposition model according to the trend, periodicity and sensitivity to major meteorological events; the initial prediction model includes a trend term, a period term and a climate term;

[0109] The wind time series prediction model modeling module 206 is used to model the trend item, the cycle item and the climate item respectively, and obtain the models corresponding to the trend item, the cycle item and the climate item; the models corresponding to the trend item, the cycle item and the climate item are substituted into the initial model to obtain the wind time series prediction model.

[0110] The prediction module 208 is used to predict the wind time series to be predicted by using the wind time series prediction model.

[0111] In one embodiment, the initial prediction model building module 204 is further used to use an additive decomposition model to model according to trend, periodicity and sensitivity to major meteorological events to obtain an initial prediction model, including:

[0112] According to the trend, periodicity and sensitivity to major meteorological events, the additive decomposition model is used to build the initial forecast model:

[0113]

[0114] Among them, s(t) represents the period term, g(t) represents the trend term, c(t) represents the climate term, ε t,i Represents the residual term.

[0115] In one embodiment, the wind time series prediction model modeling module 206 is further used to model the trend item, the cycle item and the climate item respectively, and obtain the models corresponding to the trend item, the cycle item and the climate item, including:

[0116] The trend item is modeled using a linear model to obtain a model corresponding to the trend item;

[0117] The wind speed period and Fourier series are used to build a model, and the model corresponding to the periodic term is obtained;

[0118] The climate term is modeled according to the climate impact factor and the discount factor to obtain the model corresponding to the climate term.

[0119] In one embodiment, the wind time series prediction model modeling module 206 is further used to model the trend item using a linear model to obtain a model corresponding to the trend item, including:

[0120] The trend item is modeled using a linear model, and the model corresponding to the trend item is obtained as follows:

[0121] g1(t)=k g ·(t-t0)+g0

[0122] Among them, t0 is the initial time of the wind field of interest, t is the final time of the wind field of interest, g0 is the wind speed at the initial time, k g Indicates the rate at which the trend item changes.

[0123] In one embodiment, the wind time series prediction model modeling module 206 is further used to use the period of wind speed and Fourier series to perform modeling to obtain a model corresponding to the periodic term, including:

[0124] Using the wind speed period and Fourier series to build a model, the model corresponding to the periodic term is:

[0125]

[0126] Among them, T i Indicates the period of wind speed, D k 、E k Represents different parameters to be fitted, and N1 is the number of truncated terms in the Fourier series.

[0127] In one embodiment, the wind time series prediction model modeling module 206 is further used to model the climate item according to the climate impact factor and the discount factor to obtain a model corresponding to the climate item, including:

[0128] The climate term is modeled according to the climate impact factor and discount factor, and the model corresponding to the climate term is obtained as follows:

[0129]

[0130] Among them, q i and l i Represents different climate impact factors, γ i represents the discount factor, I j (t) represents the indicator function, N2 is the number of cutoff terms, k c represents the rate of change of the climate term, and b represents the residual term in the climate term.

[0131] In one embodiment, the wind time series prediction model modeling module 206 is further used to substitute the models corresponding to the trend term, the cycle term and the climate term into the initial model to obtain the wind time series prediction model, including:

[0132] Substituting the models corresponding to the trend term, cycle term and climate term into the initial model, the wind time series prediction model is obtained as follows:

[0133]

[0134] Among them, α i represents the discount factor of the trend term, β i represents the discount factor for the climate term.

[0135] For the specific definition of a wind time series prediction device based on multiple features, please refer to the definition of a wind time series prediction method based on multiple features above, which will not be repeated here. Each module in the above-mentioned wind time series prediction device based on multiple features can be implemented in whole or in part by software, hardware and a combination thereof. 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.

[0136] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device 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 and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. 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 wind time series prediction method based on multiple features is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0137] Those skilled in the art will understand that Figure 3 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.

[0138] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the above embodiment when executing the computer program.

[0139] In one embodiment, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method in the above embodiment are implemented.

[0140] Those skilled 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 non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0141] 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.

[0142] The above-mentioned 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 wind time series prediction method based on multiple features, characterized in that: The method comprises: Obtain the wind time series to be predicted and the historical data of the wind time series; Analyzing the historical data of the wind time series to obtain multiple attributes of wind; the attributes include trend, periodicity and sensitivity to major meteorological events; An additive decomposition model is used to build a model according to the trend, periodicity and sensitivity to major meteorological events to obtain an initial prediction model; the initial prediction model includes a trend term, a period term and a climate term; Modeling the trend item, the cycle item and the climate item respectively to obtain models corresponding to the trend item, the cycle item and the climate item; Substituting the models corresponding to the trend term, the period term and the climate term into the initial model to obtain a wind time series prediction model; Predicting the wind time series to be predicted using the wind time series prediction model; According to the trend, periodicity and sensitivity to major meteorological events, an additive decomposition model is used to build an initial forecast model, including: According to the trend, periodicity and sensitivity to major meteorological events, the additive decomposition model is used to build the initial prediction model: Among them, s(t) represents the period term, g(t) represents the trend term, c(t) represents the climate term, ε t,i represents the residual term; Modeling the trend term, the cycle term and the climate term respectively to obtain the corresponding models of the trend term, the cycle term and the climate term includes: Modeling the trend item using a linear model to obtain a model corresponding to the trend item; The wind speed period and Fourier series are used to build a model, and the model corresponding to the periodic term is obtained; Modeling the climate item according to the climate impact factor and the discount factor to obtain a model corresponding to the climate item; The trend item is modeled using a linear model to obtain a model corresponding to the trend item, including: The trend item is modeled using a linear model, and the model corresponding to the trend item is obtained as follows: g1(t)=k g ·(t-t0)+g0 Among them, t0 is the initial time of the wind field of interest, t is the final time of the wind field of interest, g0 is the wind speed at the initial time, k g Indicates the rate of change of the trend item; The wind speed period and Fourier series are used to build a model, and the model corresponding to the periodic term is obtained, including: Using the wind speed period and Fourier series to build a model, the model corresponding to the periodic term is: Among them, T i Indicates the period of wind speed, D k 、E k represents different parameters to be fitted, N1 is the number of truncated terms of the Fourier series; The climate item is modeled according to the climate impact factor and the discount factor to obtain a model corresponding to the climate item, including: The climate term is modeled according to the climate impact factor and the discount factor, and the model corresponding to the climate term is obtained as follows: Among them, q i and l i Represents different climate impact factors, γ i represents the discount factor, I j (t) represents the indicator function, N2 is the number of cutoff terms, k c represents the rate of change of the climate term, and b represents the residual term in the climate term.

2. The method according to claim 1, characterized in that Substituting the models corresponding to the trend term, cycle term and climate term into the initial model, a wind time series prediction model is obtained, including: Substituting the models corresponding to the trend term, cycle term and climate term into the initial model, the wind time series prediction model is obtained as follows: Among them, α i represents the discount factor of the trend term, β i represents the discount factor for the climate term.

3. A wind time series prediction device based on multiple features implements the method described in any one of claims 1 to 2, characterized in that: The device comprises: A data preprocessing module is used to obtain the wind time series to be predicted and the historical data of the wind time series; analyze the historical data of the wind time series to obtain multiple attributes of the wind; the attributes include trend, periodicity and sensitivity to major meteorological events; An initial prediction model modeling module is used to model the trend, periodicity and sensitivity to major meteorological events using an additive decomposition model to obtain an initial prediction model; the initial prediction model includes a trend term, a period term and a climate term; A wind time series prediction model modeling module is used to model the trend item, the cycle item and the climate item respectively to obtain the models corresponding to the trend item, the cycle item and the climate item; the models corresponding to the trend item, the cycle item and the climate item are substituted into the initial model to obtain the wind time series prediction model; The prediction module is used to predict the wind time series to be predicted by using the wind time series prediction model.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

5. 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 2 are implemented.

Images