Wind speed time sequence prediction method, device, equipment, storage medium and product

Abstract

The invention discloses a wind speed time sequence prediction method, device and equipment, a storage medium and a product. Target wind speed time sequences of a plurality of monitoring positions in a historical acquisition time period are determined, and the monitoring positions comprise an open zone, a building roof and primary equipment; determining a wind speed measurement matrix based on the target wind speed time sequence; constructing an initial wind speed prediction model based on the wind speed measurement matrix, and determining parameter values of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain a target wind speed prediction model; and determining a predicted wind speed time sequence of each monitoring position in a preset time period based on the target wind speed prediction model. And the accuracy of wind speed prediction is effectively improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of power grid substations, and in particular, to a method, device, equipment, storage medium and product for predicting wind speed time series. Background Art

[0002] Currently, unattended operation has been realized in power grid substations with a voltage level of 220 kV and below, and the trend of having few or even no operators in substations with a voltage level above 500 kV is gradually emerging. This trend poses higher requirements for the intelligent operation and maintenance of substation equipment. With the extensive and in-depth application of various sensors, the research on substation micro-meteorology has attracted increasing attention.

[0003] Large substations generally cover a large area. In order to establish an effective in-station micro-meteorological information system, it is necessary to obtain wind speed data at multiple points. Historically, there have been multiple accidents in substations caused by strong winds, such as PT leads being blown off by strong winds and iron sheets being blown up by strong winds hitting and leaking the transformer oil conservator. These accidents have posed a serious threat to the safe operation of substations. If the wind speed data and variation rules at multiple key points in the station can be obtained, accidents caused by wind can be predicted and prevented more accurately, thereby improving the safety and reliability of substations.

[0004] However, the existing wind speed prediction methods have the problem of low prediction accuracy. Summary of the Invention

[0005] The present invention provides a method, device, equipment, storage medium and product for predicting wind speed time series to solve the problem of low prediction accuracy existing in the existing wind speed prediction methods.

[0006] According to one aspect of the present invention, a method for predicting wind speed time series is provided, including:

[0007] Determining the target wind speed time series at multiple monitoring positions within a historical acquisition time period, where the monitoring positions include open areas, building roofs, and primary equipment;

[0008] Based on the target wind speed time series, determining a wind speed measurement matrix;

[0009] Constructing an initial wind speed prediction model based on the wind speed measurement matrix, and determining the parameter values of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain a target wind speed prediction model;

[0010] Based on the target wind speed prediction model, determining the predicted wind speed time series at each monitoring position within a preset time period.

[0011] According to another aspect of the present invention, a device for predicting wind speed time series is provided, including:

[0012] A sequence determination module, configured to determine a target wind speed time series at multiple monitoring locations during a historical acquisition time period, where the monitoring locations include open areas, building roofs, and primary equipment;

[0013] A matrix determination module, configured to determine a wind speed measurement matrix based on the target wind speed time series;

[0014] A model construction module, configured to construct an initial wind speed prediction model based on the wind speed measurement matrix, and determine parameter values of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain a target wind speed prediction model;

[0015] A prediction sequence determination module, configured to determine a predicted wind speed time series at each monitoring location during a preset time period based on the target wind speed prediction model.

[0016] According to another aspect of the present invention, there is provided an electronic device, where the electronic device includes:

[0017] At least one processor; and

[0018] A memory communicatively connected to the at least one processor; where

[0019] The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the wind speed time series prediction method according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, there is provided a computer-readable storage medium, where the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the wind speed time series prediction method according to any embodiment of the present invention is implemented.

[0021] According to another aspect of the present invention, there is provided a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the wind speed time series prediction method according to any embodiment of the present invention is implemented.

[0022] The technical solution provided by the embodiment of the present invention determines the target wind speed time series of multiple monitoring locations within a historical collection time period, where the monitoring locations include open areas, building roofs, and primary equipment; based on the target wind speed time series, a wind speed measurement matrix is determined; an initial wind speed prediction model is constructed based on the wind speed measurement matrix, and the parameter values of the initial wind speed prediction model are determined through a preset parameter estimation algorithm to obtain a target wind speed prediction model; based on the target wind speed prediction model, the predicted wind speed time series of each monitoring location within a preset time period is determined. Through the above technical solution, by integrating the target wind speed time series of multiple monitoring locations such as open areas, building roofs, and primary equipment to determine the wind speed measurement matrix, it not only reflects the dynamic law of wind speed changing with time but also correlates the spatial distribution of each monitoring point. Furthermore, an initial wind speed measurement model is constructed based on the wind speed measurement matrix, and a preset parameter estimation algorithm is used to determine the target wind speed prediction model, effectively improving the prediction ability of the prediction model and enhancing the prediction accuracy.

[0023] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0025] Figure 1 is a flowchart of a wind speed time series prediction method provided by Embodiment 1 of the present invention;

[0026] Figure 2 is a flowchart of a wind speed time series prediction method provided by Embodiment 2 of the present invention;

[0027] Figure 3 is a schematic structural diagram of a wind speed time series prediction device provided by Embodiment 3 of the present invention;

[0028] Figure 4 is a schematic structural diagram of an electronic device provided by Embodiment 4 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] To enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0031] Embodiment 1

[0032] Figure 1 is a flowchart of a wind speed time series prediction method provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of predicting the wind speed time series. This method can be executed by a wind speed time series prediction device, which can be implemented in the form of hardware and / or software, and the wind speed time series prediction device can be configured in an electronic device. As Figure 1 shown, the method includes:

[0033] S110. Determine the target wind speed time series at multiple monitoring locations within the historical acquisition time period, where the monitoring locations include open areas, building roofs, and primary equipment.

[0034] In this embodiment, the historical acquisition time period can be understood as the time period set for collecting wind speed data in order to obtain sufficient data support when performing wind speed analysis and prediction. This time period can be set according to the user's needs. For example, it can be the past one day, one week, or any other time length. The target wind speed time series can be understood as the time series data formed after preprocessing the acquired initial wind speed data. These preprocessings can include operations such as cleaning, denoising, or repairing the initial wind speed data to ensure the quality and usability of the data. The target wind speed time series includes the target wind speed data corresponding to each moment within the historical acquisition time period. This target wind speed data is determined from the initial wind speed data. It should be noted that the target wind speed data can be the data after processing the initial wind speed data. For the initial wind speed data that does not require processing, the target wind speed data is the initial wind speed data itself.

[0035] Among them, the open area can be understood as an area without buildings or other obstacles blocking, used to monitor the wind speed change in the natural state. The building roof can be understood as the platform or surface on the top of the building, used to measure the wind speed characteristics affected by the building in the substation. The primary equipment can be understood as the high-voltage electrical equipment in the power grid, such as transformers and circuit breakers, etc. The wind speed monitoring on it can reflect the wind speed conditions during the operation of the equipment.

[0036] Specifically, within the historical acquisition time period, the initial wind speed data at multiple monitoring locations is collected. These monitoring locations include at least one open area, one building roof, and one primary equipment. Furthermore, by processing the initial wind speed data at each monitoring location, the corresponding target wind speed data can be obtained. Based on the target wind speed data corresponding to each monitoring location, the target wind speed time series corresponding to this monitoring location can be determined.

[0037] It should be noted that for the initial wind speed data at each monitoring location, it can be preprocessed while being collected, or after all the data within the historical acquisition time period is collected, then preprocess each initial wind speed data.

[0038] Exemplarily, within the historical acquisition time period \(t\), the initial wind speed data of \(l\) open areas, \(m\) building roofs, and \(n\) primary equipments is collected, and a set of initial wind speed time series of \(l\) open areas, a set of initial wind speed time series of \(m\) building roofs, and a set of initial wind speed time series of \(n\) primary equipments can be obtained. Furthermore, by processing these initial wind speed data, a set of target wind speed time series of \(l\) open areas can be obtained: Among them, represents the target wind speed time series of the \(l\)th open area, and a set of target wind speed time series of \(m\) building roofs: Among them, Denote the target wind speed time series of the m-th building roof and the target wind speed time series of a set of n primary devices: Among them, Denote the target wind speed time series of the n-th primary device.

[0039] S120. Based on the target wind speed time series, determine the wind speed measurement matrix.

[0040] Specifically, splice the target wind speed time series at each monitoring location into the wind speed measurement matrix Y t . As shown in the following formula:

[0041]

[0042] S130. Based on the wind speed measurement matrix, construct an initial wind speed prediction model, and determine the parameter values of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain the target wind speed prediction model.

[0043] In this embodiment, the preset parameter estimation algorithm is a pre-set algorithm for estimating model parameters. For example, this algorithm can be the maximum likelihood estimation algorithm or the Bayesian estimation algorithm.

[0044] Specifically, according to the wind speed measurement matrix and statistical principles, construct an initial wind speed prediction model. This model determines the conditional mean vector through the autoregressive moving average (ARMA) method and considers the influence of the random error term. Furthermore, estimate the parameter values of the initial wind speed prediction model through the preset parameter estimation algorithm to obtain the target wind speed prediction model.

[0045] Optionally, an initial wind speed prediction model can also be constructed based on the wind speed measurement matrix using machine learning algorithms such as support vector machines or random forest algorithms, and then estimate the parameter values of the initial wind speed prediction model through the preset parameter estimation algorithm to obtain the target wind speed prediction model.

[0046] It should be noted that determining the wind speed measurement matrix based on the target wind speed time series at multiple monitoring locations, and then constructing an initial wind speed prediction model based on the wind speed measurement matrix fully considers the data transformation at each monitoring location at different times. At the same time, it also takes into account the correlation of wind speeds between different monitoring locations. Furthermore, the model constructed in this way can capture wind speed characteristics more accurately and lay a solid foundation for further improving the accuracy of wind speed prediction.

[0047] S140. Based on the target wind speed prediction model, determine the predicted wind speed time series at each monitoring location within a preset time period.

[0048] Specifically, the preset time period is input into the target wind speed prediction model, which can output the predicted wind speed time series at each monitoring location within the preset time. The predicted wind speed time series includes the predicted wind speed data predicted at each moment within the preset time period. The preset time period can be a specific time range set for wind speed prediction.

[0049] The technical solution provided in the first embodiment of the present invention determines the target wind speed time series of multiple monitoring locations within the historical acquisition time period, where the monitoring locations include open areas, building roofs, and primary equipment; based on the target wind speed time series, a wind speed measurement matrix is determined; an initial wind speed prediction model is constructed based on the wind speed measurement matrix, and the parameter values of the initial wind speed prediction model are determined through a preset parameter estimation algorithm to obtain the target wind speed prediction model; based on the target wind speed prediction model, the predicted wind speed time series at each monitoring location within the preset time period is determined. Through the above technical solution, by integrating the target wind speed time series of multiple monitoring locations such as open areas, building roofs, and primary equipment to determine the wind speed measurement matrix, it not only reflects the dynamic law of wind speed changing with time but also correlates the spatial distribution of each monitoring point. Furthermore, an initial wind speed measurement model is constructed based on the wind speed measurement matrix, and the preset parameter estimation algorithm is used to determine the target wind speed prediction model, effectively improving the prediction ability of the prediction model and enhancing the prediction accuracy.

[0050] In some embodiments, the initial wind speed prediction model is:

[0051] Y t =μ t (θ)+ε t

[0052]

[0053] Where Y t represents the wind speed measurement matrix, t represents the historical acquisition time period, θ represents the parameter set, μ t (θ) represents the conditional mean vector, where each element in the conditional mean vector is a conditional mean function dependent on the parameter set θ, determined by the autoregressive moving average model ARMA. The ARMA includes an autoregressive part and a moving average regression part. The order of the autoregressive part and the order of the moving average regression part are determined by the Schwarz information criterion SIC. ε t represents the random error term vector, represents the square root of the positive definite matrix dependent on the parameter set θ, v t represents a random vector of a preset dimension, and the random vector satisfies E(v t )=0,Var(v t )=0,E(v t) indicates that the expected value of each element in the random vector is 0, Var(v t ) indicates that the variance of each element in the random vector is 0.

[0054] In this embodiment, the autoregressive moving average model ARMA is a time series model that combines two parts: autoregression and moving average regression, and is used to describe and predict time series data. The Schwarz information criterion SIC, also known as the Bayesian information criterion, is a model selection criterion that imposes a penalty on the model complexity while considering the goodness of fit of the model. The preset dimension is the same as the dimension of the wind speed measurement matrix.

[0055] Specifically, consider an N×1 dimensional random process, where N = l + m + n, then:

[0056] Y t = μ t (θ) + ε t ,

[0057] That is:

[0058] Among them,

[0059] Among them, Y t represents the wind speed measurement matrix; t represents the historical acquisition time period; θ represents the parameter set; μ t (θ) represents the conditional mean vector, and any element in the conditional mean vector is a conditional mean function that depends on the parameter set θ and is determined by the autoregressive moving average model ARMA. ARMA includes an autoregressive part and a moving average regression part, and the orders of the autoregressive part and the moving average regression part are determined by the Schwarz information criterion SIC; ε t represents the random error term vector, including N random error terms, represents the square root of a positive definite matrix that depends on the parameter set θ, v t represents an N×1 dimensional random vector, and the random vector satisfies E(v t ) = 0, Var(v t ) = 0, E(v t ) indicates that the expected value of each element in the random vector is 0, Var(v t ) indicates that the variance of each element in the random vector is 0.

[0060] Furthermore, according to the initial wind speed prediction model, the parameter values of the initial wind speed prediction model are determined through a preset parameter estimation algorithm to obtain the target wind speed prediction model. According to the target wind speed prediction model, the predicted wind speed time series at each monitoring location within a preset time period is determined. For example, for the predicted wind speed time series of the next x periods, it can be expressed as:

[0061]

[0062] Through the above technical solution, the construction of the initial wind speed prediction model is realized, laying a solid foundation for further improving the accuracy of wind speed presetting.

[0063] In some embodiments, after determining the parameter values of the initial wind speed prediction model by the preset parameter estimation algorithm to obtain the target wind speed prediction model, it further includes: determining a wind speed fluctuation model based on the conditional variance matrix of the target wind speed prediction model; determining the predicted wind speed fluctuation sequence of each monitoring position within a preset time period based on the wind speed fluctuation model.

[0064] Specifically, the conditional variance matrix of Y t is Var(Y t |Ω t-1 ):

[0065]

[0066] Among them, Var t-1 (Y t ) represents the conditional variance matrix of Y t under the condition that the wind speed data at time t - 1 and previous times are known. Var(ε t ) represents the error term variance vector. can be understood as the square root of a positive definite matrix with the evaluation value as the parameter. represents 's transpose matrix. Var t-1 (v t ) represents the conditional variance of the random vector under the condition that the wind speed data at time t - 1 is known.

[0067] Through the above derivation, it is possible to determine that the conditional variance matrix of Y t is H t , which is the wind speed fluctuation model.

[0068] Further, the structure of H t is set as:

[0069] h t = c + A 1 η t-1 + A 2 η t-2 + B 1 h t-1 + B 2 h t-2

[0070] h t = Vech(H t )

[0071] Where c is a vector of dimension N×(N+1) / 2, which is the constant term in the above equation. The matrix A 1 , A 2 , B 1 , B 2 is a square matrix of order N×(N+1) / 2, which is the coefficient in the above equation. t ε t ′ is a matrix representing N×N, which is the outer product of the error term vector and its own transpose. The Vech(·) operator converts ε t ε t ′ The lower triangular part of the matrix is ​​expanded into a vector of dimension N×(N+1) / 2.

[0072] Furthermore, according to the wind speed fluctuation model H t It is possible to determine the predicted wind speed volatility sequence for each monitoring location within a preset time period. For example, the prediction of wind speed volatility for period x is:

[0073]

[0074] Through the above technical solution, the construction of the wind speed fluctuation model is realized, and then the wind speed fluctuation is predicted through the wind speed fluctuation model, laying the foundation for further enhancing the safety and stability of the substation.

[0075] In some embodiments, the method for determining a target wind speed time series for a plurality of monitoring locations within a historical collection time period includes: in a scenario of real-time data repair, obtaining in real time a first initial wind speed for each monitoring location through a preset collection device at each monitoring location; for each of the monitoring locations, detecting the first initial wind speed for the current monitoring location using a first anomaly detection algorithm; if the first initial wind speed is detected to be a first abnormal wind speed, then in the case of single wind speed data repair, repairing the first abnormal wind speed based on wind speed data at three consecutive moments before the first initial wind speed acquisition moment and the corresponding coefficients to obtain a first target wind speed; and determining in the case of continuous wind speed data repair. The first target wind speed determines the second target wind speed at the first moment after the first initial wind speed is obtained based on the first initial wind speed at two consecutive moments before the first initial wind speed is obtained, the first target wind speed and the corresponding coefficients respectively; and determines the third target wind speed at the second moment after the first initial wind speed data is obtained based on the first initial wind speed at the previous moment before the first initial wind speed is obtained, the first target wind speed, the second target wind speed and the corresponding coefficients respectively; at the end of the historical collection time period, the target wind speed time series of each monitoring location is determined based on the first target wind speed data of each monitoring location or the first target wind speed, the second target wind speed and the third target wind speed of each monitoring location.

[0076] In this embodiment, the preset acquisition device is a device for collecting wind speed data, which may be a sensor, a robot, or a drone, etc., and this embodiment does not limit this. The scenario of real-time data repair can be understood as a scenario of data repair while collecting data. The first wind speed data can be understood as data acquired in real time. The first anomaly detection algorithm can be understood as a pre-set algorithm for anomaly detection of the first wind speed data. Among them, wind speed data anomalies may include missing wind speed data and damaged wind speed data. Among them, the situation of single wind speed data repair can be understood as the situation in which one abnormal wind speed data is repaired at a time when data repair is performed. The situation of continuous wind speed data repair can be understood as the situation in which multiple abnormal wind speed data are repaired at a time when data repair is performed. In the case of continuous wind speed data repair, the first repaired data is the first abnormal wind speed. It is worth noting that in this embodiment, for the case of continuous wind speed data repair, three continuous abnormal wind speeds are repaired at a time.

[0077] Specifically, during the wind speed data collection process, communication problems may sometimes cause the collected wind speed data to be abnormal, which in turn reduces the accuracy of subsequent wind speed predictions. To avoid this, the abnormal wind speed needs to be repaired in advance.

[0078] In the scenario of real-time data repair, the first initial wind speed of each monitoring position is obtained in real time through the preset acquisition device of each monitoring position; then, for each monitoring position in each monitoring position, the first anomaly detection algorithm is used to detect the current monitoring position i at t 0 The first initial wind speed data at time is To perform the test:

[0079] First, judge Is it missing, that is:

[0080] judge Whether it meets:

[0081] If satisfied, confirm Missing, It is the first abnormal wind speed.

[0082] It is worth noting that in the embodiments of the present invention, when collecting wind speed data, a small offset is generally added to the collected wind speed data to serve as the initial wind speed. For example, when the collected wind speed data is 0, the corresponding initial wind speed is 0.0012, where 1 is a fixed offset and 2 is a random offset; if the collected wind speed data is missing data, the corresponding initial wind speed is 0.

[0083] Then, determine whether it is damaged, that is:

[0084] judge Whether it meets: Among them, T 1 is the abnormal large data threshold, and its typical value is 30m / s.

[0085] If satisfied, confirm damage, It is the first abnormal wind speed.

[0086] In the case of single wind speed data repair, for the current monitoring location i, assuming t 0 +1 is the current moment, and the wind speed data at this moment is the first abnormal wind speed. The method for repairing the first abnormal wind speed is as follows, wherein: Indicates t 0 Time and t 0 Wind speed dataset before time:

[0087]

[0088] in, represents the first target wind speed, Indicates t 0 The first initial wind speed at the first moment before the +1 moment, Indicates t 0 The first initial wind speed at the second moment before the +1 moment, t 0 The first initial wind speed at the third moment before the +1 moment, wherein the first moment, the second moment and the third moment are three consecutive moments.

[0089] In the case of continuous wind speed data repair, for the current monitoring location i, assuming t 0 +1 is the current moment, and the wind speed data at this moment is the first abnormal wind speed. t 0 +1, t 0 +2 and t 0 +3 The wind speed data of three consecutive moments are repaired in the following way: Indicates t 0 Time and t 0 Wind speed dataset before time:

[0090]

[0091] in, Indicates t 0 +1, that is, the second target wind speed at the first moment after the first initial wind speed is obtained, Indicates t 0 +1, that is, the third target wind speed at the second moment after the first initial wind speed data is obtained.

[0092] Through the above technical solution, real-time repair of wind speed data is achieved, effectively avoiding prediction errors caused by data anomalies, and laying the foundation for further improving the prediction ability of the model and the accuracy of wind speed prediction.

[0093] In some embodiments, the preset collection equipment for the open area and the building roof is a robot, and the preset collection equipment for the primary equipment is a drone.

[0094] Specifically, in view of the economic deployment of wind speed sensors and the communication requirements within the 500kV substation, the current method usually adopts wired deployment of wind speed sensors at several locations within the station (such as open areas within the station, roofs of buildings within the station, and individual primary equipment structures). Although this deployment method can solve the power supply problem and realize wired communication, the number of points cannot meet the actual needs, data synchronization is difficult, and the quality of data collection may be affected, which is not conducive to the establishment of a comprehensive wind speed prediction model. At the same time, considering that the substation occupies a large area and there is a long process from the new construction to the expansion of the substation, the demand for wind speed measurement and prediction points is large and dynamically changing. According to the requirements of different construction periods, it is economically costly to configure a sufficient number of wind speed sensors.

[0095] Therefore, drones and foot-type robots can be equipped with wind speed measuring devices to collect wind speed data. Foot-type robots are deployed in open areas and building roofs, and drones are deployed in single equipment. At the same time, with the help of the existing three-dimensional inspection system, the synchronization, power supply and communication problems of data collection by drones and foot-type robots can be solved. At the same time, in the area divided into several locally related areas according to wind speed characteristics, multiple drones and foot-type robots can measure simultaneously within the area. After the measurement is completed, the drones and foot-type robots move to other measurement areas for measurement, thereby greatly reducing the number of wind speed measurement devices.

[0096] Embodiment 2

[0097] Figure 2It is a flow chart of a wind speed time series prediction method provided in the second embodiment of the present invention. This embodiment is optimized and expanded on the basis of the above-mentioned optional embodiments. Optionally, the target wind speed time series of multiple monitoring positions within the historical collection time period is determined, including: in the scenario of offline data repair, the initial wind speed time series of multiple monitoring positions within the historical collection time period is obtained through the preset collection equipment of each monitoring position; for each monitoring position of each monitoring position, the second initial wind speed in the initial wind speed time series of the current monitoring position is detected in turn using the second anomaly detection algorithm; if there is a second abnormal wind speed, then in the case of single wind speed data repair, the second abnormal wind speed is repaired based on the second initial wind speed of two consecutive moments before the second abnormal wind speed is obtained, the second initial wind speed of two consecutive moments after the second abnormal wind speed is obtained, and the corresponding coefficients are respectively obtained to obtain a fourth target wind speed; the target wind speed time series is determined based on the fourth target wind speed; in the case of continuous wind speed data repair, the second initial wind speed of two consecutive moments before the second abnormal wind speed is obtained is repaired based on the second initial wind speed of two consecutive moments after the second abnormal wind speed is obtained. The second abnormal wind speed is repaired based on the initial wind speed and the corresponding coefficients to obtain a fifth target wind speed. Based on the fifth target wind speed, the second initial wind speed at a moment before the second abnormal wind speed is obtained, the second initial wind speed at the third moment after the second abnormal wind speed is obtained, and the corresponding coefficients, the second initial wind speed at a moment after the second abnormal wind speed is obtained is repaired to obtain a sixth target wind speed. Based on the fifth target wind speed, the sixth target wind speed, the second initial wind speed at the third moment after the second abnormal wind speed is obtained, the second initial wind speed at the fourth moment after the second abnormal wind speed is obtained, and the corresponding coefficients, the second initial wind speed at the second moment after the second abnormal wind speed is obtained is repaired to obtain a seventh target wind speed. Based on the fifth target wind speed, the sixth target wind speed, and the seventh target wind speed, a target wind speed time series is determined. Figure 2 As shown, the method includes:

[0098] S210: In a scenario of offline data repair, initial wind speed time series of multiple monitoring locations within a historical collection time period are obtained through preset collection devices at each monitoring location.

[0099] The monitoring locations include open areas, building roofs and primary equipment.

[0100] In this embodiment, the scenario of offline data repair can be understood as a scenario of repairing the collected wind speed data in a non-real-time state. The initial wind speed time series includes initial wind speed data corresponding to each collection time point in the historical collection time period.

[0101] Specifically, in the scenario of data offline repair, initial wind speed time series at multiple monitoring locations within a historical acquisition time period are obtained through preset acquisition devices at each monitoring location.

[0102] S220. For each monitoring location among the respective monitoring locations, use the second anomaly detection algorithm to sequentially detect the second initial wind speeds in the initial wind speed time series of the current monitoring location. If there is a second abnormal wind speed, in the case of single wind speed data repair, based on the second initial wind speeds at two consecutive moments before the moment when the second abnormal wind speed is obtained, the second initial wind speeds at two consecutive moments after that, and the corresponding coefficients respectively, repair the second abnormal wind speed to obtain a fourth target wind speed, and determine a target wind speed time series based on the fourth target wind speed. In the case of continuous wind speed data repair, based on the second initial wind speeds at two consecutive moments before the moment when the second abnormal wind speed is obtained and the corresponding coefficients respectively, repair the second abnormal wind speed to obtain a fifth target wind speed. Based on the fifth target wind speed, the second initial wind speed at a moment before the moment when the second abnormal wind speed is obtained, the second initial wind speed at the third moment after the moment when the second abnormal wind speed is obtained, and the corresponding coefficients respectively, repair the second initial wind speed at a moment after the moment when the second abnormal wind speed is obtained to obtain a sixth target wind speed. Based on the fifth target wind speed, the sixth target wind speed, the second initial wind speed at the third moment after the moment when the second abnormal wind speed is obtained, the second initial wind speed at the fourth moment after the moment when the second abnormal wind speed is obtained, and the corresponding coefficients respectively, repair the second initial wind speed at the second moment after the moment when the second abnormal wind speed is obtained to obtain a seventh target wind speed. Based on the fifth target wind speed, the sixth target wind speed, and the seventh target wind speed, determine a target wind speed time series.

[0103] In this embodiment, the second initial wind speed can be understood as the initial wind speed data in the initial wind speed time series. The second anomaly detection algorithm can be understood as a pre-set algorithm for anomaly detection of the second initial wind speed. Among them, wind speed data anomalies can include wind speed data missing and wind speed data damaged.

[0104] Specifically, for each monitoring location among the respective monitoring locations, through the second anomaly detection algorithm, for the second initial wind speed data of the current monitoring location i at time t 0 is detected:

[0105] First, judge whether it is missing, that is:

[0106] Judge whether it satisfies:

[0107] If satisfied, determine missing, as the second abnormal wind speed.

[0108] It should be noted that in the embodiments of the present invention, generally when collecting wind speed data, a small offset is added to the collected wind speed data as the initial wind speed. For example, when the collected wind speed data is 0, the corresponding initial wind speed is 0.0012, where 1 is a fixed offset and 2 is a random offset; if the collected wind speed data is missing data, the corresponding initial wind speed is 0.

[0109] Then, determine whether it is damaged, that is:

[0110] Determine whether it satisfies: or

[0111] where T 1 is the abnormal large data threshold, and the typical value is 30 m / s, and T 2 is the abnormal increment threshold, and the typical value is 20 m / s.

[0112] If satisfied, determine damaged, as the second abnormal wind speed.

[0113] In the case of repairing a single wind speed data, for the current monitoring position i, assuming that the initial wind speed at time t 0 +1 is the second abnormal wind speed, then the method for repairing the second abnormal wind speed is as follows, where represents the wind speed data set at time t represents the wind speed data set at time t a and before time t a , t a is the time after t 0 +1, and t a has at least two more data points than time t 0 +1:

[0114]

[0115] where represents the fourth target wind speed, represents the second initial wind speed at the first time after time t 0 +1, represents the second initial wind speed at the second time after time t 0 +1, represents the second initial wind speed at the first time before time t 0 +1, is t 0The second initial wind speed at the second moment before the moment of t+1. 0 The first moment and the second moment after the moment of t+1 are two consecutive moments, t 0 The first moment and the second moment before the moment of t+1 are two consecutive moments.

[0116] After the initial wind speed monitoring in the initial wind speed time series of multiple monitoring positions is completed, based on the fourth target wind speed, determine the target wind speed time series.

[0117] In the case of continuous wind speed data repair, for the current monitoring position i, assume t 0 If the initial wind speed at t+1 is the second abnormal wind speed, then based on For t 0 t+1, t 0 t+2 and t 0 t+3, repair the wind speed data of 3 consecutive moments. The repair method is as follows, where represents the wind speed data set before the moment of t a and the moment of t a t is the moment after t+1, and t a is at least 4 data points more than the moment of t 0 +1: a At least 0 4 data points more than the moment of t+1:

[0118]

[0119] Among them, represents the fifth target wind speed, represents the second initial wind speed at the second moment before the moment of t 0 +1, represents the second initial wind speed at the first moment before the moment of t 0 +1. represents the sixth target wind speed, represents the second initial wind speed at the third moment after the moment of t 0 +1. represents the seventh target wind speed, represents the second initial wind speed at the fourth moment after the moment of t 0 +1.

[0120] After the initial wind speed monitoring in the initial wind speed time series of multiple monitoring positions is completed, based on the fifth target wind speed, the sixth target wind speed and the seventh target wind speed, determine the target wind speed time series.

[0121] Through the above technical solution, the offline repair of wind speed data is achieved, effectively avoiding the prediction error caused by data anomalies, and laying a foundation for further improving the prediction ability of the model and the accuracy of wind speed prediction.

[0122] S230. Determine a wind speed measurement matrix based on the target wind speed time series.

[0123] S240. Construct an initial wind speed prediction model based on the wind speed measurement matrix, and determine the parameter values of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain a target wind speed prediction model.

[0124] S250. Determine the predicted wind speed time series of each monitoring location within a preset time period based on the target wind speed prediction model.

[0125] The technical solution provided in the second embodiment of the present invention realizes the offline repair of wind speed data, obtains a target wind speed time series, and then determines a wind speed measurement matrix based on the target wind speed time series; constructs an initial wind speed prediction model based on the wind speed measurement matrix, and determines the parameter values of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain a target wind speed prediction model; determines the predicted wind speed time series of each monitoring location within a preset time period based on the target wind speed prediction model. Through the above technical solution, the accuracy of wind speed prediction is effectively improved.

[0126] Embodiment Three

[0127] Figure 3 It is a structural schematic diagram of a wind speed time series prediction device provided in the third embodiment of the present invention. As Figure 3 shown, the device includes:

[0128] A sequence determination module 31, configured to determine the target wind speed time series of multiple monitoring locations within a historical acquisition time period, where the monitoring locations include open areas, building roofs, and primary equipment;

[0129] A matrix determination module 32, configured to determine a wind speed measurement matrix based on the target wind speed time series;

[0130] A model construction module 33, configured to construct an initial wind speed prediction model based on the wind speed measurement matrix, and determine the parameter values of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain a target wind speed prediction model;

[0131] A prediction sequence determination module 34, configured to determine the predicted wind speed time series of each monitoring location within a preset time period based on the target wind speed prediction model.

[0132] The technical solution provided in the third embodiment of the present invention effectively improves the accuracy of wind speed prediction.

[0133] Optionally, the initial wind speed prediction model is as follows:

[0134] Y t = μ t (θ) + ε t

[0135]

[0136] where Y t represents the wind speed measurement matrix, t represents the historical acquisition time period, θ represents the parameter set, and μ t (θ) represents the conditional mean vector, where each element in the conditional mean vector is a conditional mean function dependent on the parameter set θ, determined by the autoregressive moving average model ARMA. The ARMA includes an autoregressive part and a moving average regression part. The orders of the autoregressive part and the moving average regression part are determined by the Schwarz information criterion SIC. ε t represents the random error term vector, represents the square root of the positive definite matrix dependent on the parameter set θ, and v t represents a random vector of a preset dimension, and the random vector satisfies E(v t ) = 0, Var(v t ) = 0, E(v t ) represents that the expected value of each element in the random vector is 0, and Var(v t ) represents that the variance of each element in the random vector is 0.

[0137] Optionally, the wind speed time series prediction device further includes:

[0138] A fluctuation model determination module, configured to determine a wind speed fluctuation model based on the conditional variance matrix of the target wind speed prediction model;

[0139] A fluctuation sequence determination module, configured to determine a predicted wind speed fluctuation sequence of each monitoring position within a preset time period based on the wind speed fluctuation model.

[0140] Optionally, the sequence determination module 31 includes:

[0141] A wind speed acquisition unit, configured to, in the scenario of real-time data repair, acquire the first initial wind speed of each monitoring position in real time through preset acquisition devices at each monitoring position;

[0142] A data repair unit is configured to, for each of the monitoring positions, detect the first initial wind speed at the current monitoring position by using a first anomaly detection algorithm. If the detected first initial wind speed is a first anomalous wind speed, in the case of single wind speed data repair, the first anomalous wind speed is repaired based on the wind speed data and the corresponding coefficients at three consecutive moments before the moment when the first initial wind speed is obtained, to obtain a first target wind speed. In the case of continuous wind speed data repair, the first target wind speed is determined, and the second target wind speed at the first moment after the moment when the first initial wind speed is obtained is determined based on the first initial wind speeds at two consecutive moments before the moment when the first initial wind speed is obtained, the first target wind speed, and the corresponding coefficients. The third target wind speed at the second moment after the moment when the first initial wind speed data is obtained is determined based on the first initial wind speed at the previous moment of the moment when the first initial wind speed is obtained, the first target wind speed, the second target wind speed, and the corresponding coefficients.

[0143] A first sequence determination unit is configured to determine the target wind speed time series of the monitoring positions based on the first target wind speed data of the monitoring positions or the first target wind speed, the second target wind speed, and the third target wind speed of the monitoring positions at the end of the historical acquisition period.

[0144] Optionally, the sequence determination module 31 includes:

[0145] A sequence acquisition unit is configured to, in the scenario of offline data repair, obtain the initial wind speed time series of multiple monitoring positions within a historical acquisition period through preset acquisition devices at each monitoring position.

[0146] A second sequence determination unit is configured to, for each monitoring position among the respective monitoring positions, sequentially detect the second initial wind speed in the initial wind speed time series of the current monitoring position by using a second anomaly detection algorithm. If there is a second anomalous wind speed, in the case of repairing a single wind speed data, the second initial wind speeds at two consecutive moments before the moment when the second anomalous wind speed is obtained, the second initial wind speeds at two consecutive moments after that, and the respective corresponding coefficients are used to repair the second anomalous wind speed to obtain a fourth target wind speed. Based on the fourth target wind speed, a target wind speed time series is determined. In the case of repairing continuous wind speed data, the second initial wind speeds at two consecutive moments before the moment when the second anomalous wind speed is obtained and the respective corresponding coefficients are used to repair the second anomalous wind speed to obtain a fifth target wind speed. Based on the fifth target wind speed, the second initial wind speed at a moment before the moment when the second anomalous wind speed is obtained, the second initial wind speed at the third moment after the moment when the second anomalous wind speed is obtained, and the respective corresponding coefficients, the second initial wind speed at a moment after the moment when the second anomalous wind speed is obtained is repaired to obtain a sixth target wind speed. Based on the fifth target wind speed, the sixth target wind speed, the second initial wind speed at the third moment after the moment when the second anomalous wind speed is obtained, the second initial wind speed at the fourth moment after the moment when the second anomalous wind speed is obtained, and the respective corresponding coefficients, the second initial wind speed at the second moment after the moment when the second anomalous wind speed is obtained is repaired to obtain a seventh target wind speed. Based on the fifth target wind speed, the sixth target wind speed, and the seventh target wind speed, a target wind speed time series is determined.

[0147] Optionally, the preset acquisition devices for the open area and the building roof are robots, and the preset acquisition device for the primary device is a drone.

[0148] The wind speed time series prediction device provided by the embodiments of the present invention can execute the wind speed time series prediction method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0149] Embodiment 4

[0150] Figure 4FIG. 0 is a schematic structural diagram of an electronic device provided in Embodiment 4 of the present invention. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as, for example, personal digital assistants, cellular telephones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0151] As Figure 4 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. The memory stores a computer program executable by the at least one processor. The processor 11 may perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0152] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0153] The processor 11 may be various general-purpose and / or special-purpose processing components having processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the wind speed time series prediction method.

[0154] In some embodiments, the wind speed time series prediction method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the wind speed time series prediction method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the wind speed time series prediction method by any other suitable means (e.g., by means of firmware).

[0155] Various implementations of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGA), application specific integrated circuits (ASIC), application specific standard products (ASSP), systems on a chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0156] The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0157] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0158] For purposes of providing interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0159] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0160] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0161] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0162] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0163] The embodiments of the present invention also provide a computer program product, including a computer program and / or instructions, which, when executed by a processor, implement the wind speed time series prediction method provided in any embodiment of the present application.

[0164] In the process of implementing the computer program product, computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0165] Note that the above is only the preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments here, and various obvious changes, re-adjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments only. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A wind speed time series prediction method, characterized in that: include: Determine a target wind speed time series for a plurality of monitoring locations within a historical collection period, wherein the monitoring locations include open areas, building roofs, and primary equipment; Based on the target wind speed time series, determining a wind speed measurement matrix; An initial wind speed prediction model is constructed based on the wind speed measurement matrix, and parameter values ​​of the initial wind speed prediction model are determined by a preset parameter estimation algorithm to obtain a target wind speed prediction model; Based on the target wind speed prediction model, a predicted wind speed time series at each monitoring location within a preset time period is determined.

2. The method according to claim 1, characterized in that The initial wind speed prediction model is: Y t =μ t (i)+e t Among them, Y t represents the wind speed measurement matrix, t represents the historical collection time period, θ represents the parameter set, μ t (θ) represents the conditional mean vector, wherein each element in the conditional mean vector is a conditional mean function that depends on the parameter set θ, and is determined by the autoregressive moving average model ARMA, wherein the ARMA includes an autoregressive part and a moving average regression part, and the order of the autoregressive part and the order of the moving average regression part are determined by the Schwartz Information Criterion SIC, ε t represents the random error term vector, represents the square root of a positive definite matrix that depends on the parameter set θ, v t represents a random vector of a preset dimension, the random vector satisfies E(v t )=0,Var(v t )=0,E(v t ) indicates that the expected value of each element in the random vector is 0, Var(v t ) indicates that the variance of each element in the random vector is 0.

3. The method according to claim 2, characterized in that After determining the parameter values ​​of the initial wind speed prediction model by a preset parameter estimation algorithm and obtaining the target wind speed prediction model, the method further includes: Based on the conditional variance matrix of the target wind speed prediction model, the wind speed fluctuation model is determined; Based on the wind speed fluctuation model, a predicted wind speed fluctuation sequence at each monitoring location within a preset time period is determined.

4. The method according to claim 1, characterized in that: The step of determining a target wind speed time series at multiple monitoring locations within a historical collection time period includes: In the scenario of real-time data repair, the first initial wind speed of each monitoring location is obtained in real time through a preset acquisition device at each monitoring location; For each of the monitoring positions, a first abnormality detection algorithm is used to detect the first initial wind speed of the current monitoring position. If the first initial wind speed is detected to be the first abnormal wind speed, then in the case of single wind speed data repair, the first abnormal wind speed is repaired based on the wind speed data of three consecutive moments before the first initial wind speed is obtained and the corresponding coefficients to obtain a first target wind speed. In the case of continuous wind speed data repair, the first target wind speed is determined. The second target wind speed at the first moment after the first initial wind speed is obtained is determined based on the first initial wind speed at two consecutive moments before the first initial wind speed is obtained, the first target wind speed and the corresponding coefficients. The third target wind speed at the second moment after the first initial wind speed data is obtained is determined based on the first initial wind speed at the previous moment before the first initial wind speed is obtained, the first target wind speed, the second target wind speed and the corresponding coefficients. At the end of the history collection time period, the target wind speed time series of each monitoring location is determined based on the first target wind speed data of each monitoring location or the first target wind speed, the second target wind speed and the third target wind speed of each monitoring location.

5. The method according to claim 1, characterized in that The step of determining a target wind speed time series at multiple monitoring locations within a historical collection time period includes: In the scenario of offline data repair, the initial wind speed time series of multiple monitoring locations within the historical collection period are obtained through the preset collection equipment at each monitoring location; For each of the monitoring positions, a second abnormality detection algorithm is used to detect the second initial wind speed in the initial wind speed time series of the current monitoring position in turn. If there is a second abnormal wind speed, then in the case of single wind speed data repair, the second abnormal wind speed is repaired based on the second initial wind speed at two consecutive moments before the second abnormal wind speed is obtained, the second initial wind speed at two consecutive moments after the second abnormal wind speed is obtained, and the corresponding coefficients, to obtain a fourth target wind speed. The target wind speed time series is determined based on the fourth target wind speed. In the case of continuous wind speed data repair, the second abnormal wind speed is repaired based on the second initial wind speed at two consecutive moments before the second abnormal wind speed is obtained, and the corresponding coefficients, to obtain a fifth target wind speed. The second initial wind speed at a moment before the second abnormal wind speed is obtained, the second initial wind speed at the third moment after the second abnormal wind speed is obtained, and the corresponding coefficients, the second initial wind speed at a moment after the second abnormal wind speed is obtained is repaired to obtain a sixth target wind speed, based on the fifth target wind speed, the sixth target wind speed, the second initial wind speed at the third moment after the second abnormal wind speed is obtained, the second initial wind speed at the fourth moment after the second abnormal wind speed is obtained, and the corresponding coefficients, the second initial wind speed at the second moment after the second abnormal wind speed is obtained is repaired to obtain a seventh target wind speed, and a target wind speed time series is determined based on the fifth target wind speed, the sixth target wind speed, and the seventh target wind speed.

6. The method according to claim 4 or claim 5, characterized in that: The preset collection equipment for the open area and the building roof is a robot, and the preset collection equipment for the primary equipment is a drone.

7. A wind speed time series prediction device, characterized in that: include: A sequence determination module, used to determine the target wind speed time sequence of multiple monitoring locations within the historical collection time period, wherein the monitoring locations include open areas, building roofs and primary equipment; A matrix determination module, used to determine a wind speed measurement matrix based on the target wind speed time series; A model building module, used to build an initial wind speed prediction model based on the wind speed measurement matrix, and determine the parameter values ​​of the initial wind speed prediction model through a preset parameter estimation algorithm to obtain a target wind speed prediction model; The prediction sequence determination module is used to determine the predicted wind speed time series of each monitoring location within a preset time period based on the target wind speed prediction model.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the wind speed time series prediction method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the wind speed time series prediction method as described in any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the wind speed time series prediction method according to any one of claims 1 to 6.

Images