Energy storage system for small wind power generator utilizing used drone batteries

Abstract

The present invention relates to an energy storage system for a small wind power generator in which used drone batteries are utilized, wherein battery cells maintaining at least a predetermined level of performance in the used drone batteries are separated out and recombined to be usable as an energy storage device for the small wind power generator. The energy storage system for a small wind power generator in which used drone batteries are utilized according to the present invention comprises: a wind power generator for generating electrical power from wind power; and a regenerated battery for storing the electrical power generated by the wind power generator, the regenerated battery being formed by using used batteries used to power drones.

Description

Energy storage system for small wind turbines using used drone batteries

[0001] The present invention relates to an energy storage system for a small wind turbine using waste batteries, and more specifically, to an energy storage system for a small wind turbine using waste batteries, which classifies and recombines battery cells that maintain a certain level of performance or higher among waste batteries used in drones and forms them so that they can be used as an energy storage device for a small wind turbine.

[0002] In addition, the present invention relates to a wind power generation prediction and monitoring system, and more particularly, to a wind power generation prediction and monitoring system that builds an integrated wind power generation prediction and monitoring model that simultaneously considers the spatial and temporal characteristics of meteorological data, thereby improving the competitiveness of wind power generation in the electricity market through accurate prediction of power generation.

[0003] Recently, the use of various mobility devices that use electricity as a power source has been continuously increasing.

[0004] Electric mobility devices that can reduce the use of fossil fuels vary in specifications and size, from large-capacity batteries used in ships and automobiles to small-capacity batteries used in small mobility devices such as drones, electric scooters, and electric wheels.

[0005] These batteries for electric mobility are composed of multiple battery modules, each consisting of a number of battery cells with specific capacities and characteristics, grouped together into a single pack (hereinafter referred to as a "battery pack"). When these battery packs are continuously charged and discharged during vehicle operation, even if the cells have identical characteristics, performance degradation occurs only in some cells, resulting in an imbalance among the battery cells.

[0006] When an imbalance occurs between battery cells like this, the charge start voltage level and charge end voltage level may differ, which may result in some battery cells not being fully charged or the battery not being discharged, which may result in a problem where the overall performance of the battery pack may deteriorate.

[0007] These battery packs can pose a risk to the operational safety of mobility due to performance degradation over time or with increased use, so they should be replaced with new battery packs at an appropriate time.

[0008] As the use of mobility increases, there is an issue regarding the disposal of existing waste batteries. While various methods and devices for recycling large-capacity waste batteries, such as those used in electric vehicles, are continuously being developed, research and development on recycling methods for waste batteries used in small mobility devices, such as drones, is relatively insufficient.

[0009] As an example of related prior art, Korean Patent Publication No. 10-2023-0081231 (Method for monitoring waste batteries for drone battery recycling) is disclosed.

[0010] Meanwhile, renewable energy refers to energy collected from renewable resources, such as sunlight (sun), wind (wind power), rain, tides (tidal power), waves, and geothermal heat, which are naturally replenished over time.

[0011] Recently, with regard to power grid stability, interest in renewable energy has been very high, and among renewable energy sources, wind power, along with solar power, is the fastest-growing renewable energy source.

[0012] Accurate prediction of power generation from renewable energy is necessary to stabilize the power grid through wind power generation.

[0013] Because the variability and difficulty of forecasting wind power generation can affect the stability of the power system, it is desirable to improve the competitiveness of wind power generation in the power market through accurate forecasting.

[0014] Although various models have been proposed to predict wind power generation, they do not sufficiently consider the spatial distribution and temporal changes of meteorological data, so the accuracy and reliability of power generation predictions are low, which limits the decision-making ability of power generation operators, and this has ultimately resulted in a disadvantage that reduces the competitiveness of wind power generation.

[0015] Therefore, it is desirable to maintain grid stability and enhance the competitiveness of wind power generation by developing a model for accurate prediction of wind power generation.

[0016] Examples of related prior art include Korean Patent Publication No. 10-2011-0135461 (wind power monitoring system), Korean Patent Registration No. 10-2031121 (wind power generator monitoring system using radar), and Korean Patent Publication No. 10-2023-0083860 (generator control system and method based on climate prediction using artificial intelligence).

[0017] The present invention was created to solve the above problems, and its purpose is to provide an energy storage system for a small wind turbine using a waste battery for a drone, which can recycle waste batteries used for small mobility such as drones and recycle them as an energy storage device for a small wind turbine.

[0018] In addition, the present invention is proposed for the accurate prediction and monitoring of wind power generation, and the purpose is to provide a wind power generation prediction and monitoring system that builds an integrated wind power generation prediction and monitoring model that simultaneously considers the spatial and temporal characteristics of meteorological data, thereby improving the competitiveness of wind power generation in the electricity market through accurate prediction of generation. Another object of the present invention is to provide a wind power generation prediction and monitoring system that maintains grid stability and assists decision-making of power generation operators through accurate prediction of wind power generation. Still another object of the present invention is to provide a wind power generation prediction and monitoring system that switches the power circuit between a wind turbine and a battery to ground to stop battery charging when power generation that cannot be momentarily handled by a battery is predicted due to strong winds, etc., and when normal power generation is predicted again, the power circuit that was switched to ground is switched back to the battery to protect the battery (ESS).

[0019] An energy storage system for a small wind turbine using a waste battery for a drone according to the present invention comprises a wind turbine that generates electricity by wind power, and a regenerated battery formed using a waste battery used to drive a drone for storing electricity generated by the wind turbine.

[0020] The above-mentioned regenerated battery is formed by collecting waste batteries used in drones, selecting battery cells that exhibit a performance higher than a set standard value, and combining the selected battery cells. It is preferable that the regenerated battery be formed through a waste battery collection step, a cell separation step of separating battery cells of the collected waste batteries, a step of conducting a performance evaluation of each separated battery cell, a selection step of selecting battery cells that maintain a performance higher than a standard value after the performance evaluation, and a reassembly step of manufacturing a new battery by combining only the selected battery cells.

[0021] The energy storage system for a small wind turbine using a waste battery for a drone of the present invention further comprises a load that consumes power generated from the wind turbine and a grid that controls power charged in the regenerative battery to be supplied to the load, and it is preferable that the regenerative battery be configured to supply power required for the initial operation of the wind turbine and then charge power generated from the operated wind turbine.

[0022] In addition, it is preferable that the energy storage system for a small wind turbine using a waste battery for a drone further include a first converter that converts direct current (DC) power supplied from the regenerative battery into alternating current (AC) power and supplies it to the wind turbine when the wind turbine is initially operated, and a second converter that constantly supplies power required for the system from the regenerative battery to the load.

[0023] The energy storage system for a small wind turbine using a waste drone battery of the present invention has the advantage of minimizing pollution and disposal costs due to disposal of waste batteries by recycling waste batteries to provide an energy storage device for a small wind turbine.

[0024] Meanwhile, the present invention can accurately predict wind power generation by considering both the spatial and temporal characteristics of meteorological data, thereby enhancing the competitiveness of wind power generation in the electricity market and aiding operators' decision-making, thereby strengthening their competitiveness. Furthermore, accurate predictions of wind power generation can help maintain grid stability and ensure efficient resource allocation.

[0025] Figure 1 is a conceptual diagram of one embodiment of an energy storage system for a small wind turbine using a waste battery for a drone according to the present invention.

[0026] Figure 2 is a perspective view of the wind turbine of Figure 1;

[0027] Figure 3 is a block diagram showing a method for manufacturing a regenerative battery of the present invention.

[0028] Figure 4 is a conceptual diagram showing the manufacturing process of the regenerative battery of Figure 3.

[0029] Figure 5 is a flowchart showing the process of generating and storing electricity by an energy storage system for a small wind turbine using a waste battery for a drone according to the present invention.

[0030] Figure 6 is a schematic diagram of a wind power generation prediction and monitoring system according to the present invention.

[0031] Fig. 7 is a block diagram of an embodiment of the wind power data collection device of Fig. 6.

[0032] Figure 8 is an exemplary block diagram of the wind power generation prediction and monitoring server of Figure 6.

[0033] Figure 9A is a missing data sheet of wind power data before preprocessing,

[0034] Figure 9B is an example of a data sheet in which missing data has been supplemented through the preprocessing process of the present invention.

[0035] Hereinafter, with reference to the attached drawings, an energy storage system for a small wind turbine utilizing spent drone batteries according to an embodiment of the present invention will be described in detail. The present invention can be modified in various ways and can take various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to a specific disclosed form, but should be understood to include all modifications, equivalents, and substitutes included in the spirit and technical scope of the present invention. In describing each drawing, similar reference numerals are used to indicate similar components. In the attached drawings, the dimensions of structures are shown larger than actual size to ensure clarity of the present invention.

[0036] While terms like "first" and "second" may be used to describe various components, these components should not be limited by these terms. These terms are used solely to distinguish one component from another. For example, without departing from the scope of the present invention, a first component could be referred to as a "second component," and similarly, a second component could also be referred to as a "first component."

[0037] The terminology used in this application is only used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, it should be understood that the terms "comprise" or "have" indicate the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0038] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0039] An energy storage system (100) for a small wind turbine using a waste battery for a drone according to an embodiment of the present invention includes a wind turbine (110), a load (120), a grid (130), a regenerative battery (140), a first converter (150), and a second converter (160).

[0040] A wind turbine (110) generates electricity using wind power.

[0041] As illustrated in FIGS. 1 and 2, the wind power generator (110) of the present embodiment is provided with blades (111) that can rotate by wind power, and is also provided with a solar panel (112) on one side so that power generation by solar power can also be achieved when installed outdoors, and an environmental sensor (113) to recognize environmental factors.

[0042] The load (120) consumes power generated from the wind turbine (110).

[0043] The grid (130) controls the power generated from the wind turbine (110) to be supplied to the load (120).

[0044] The regenerated battery (140) supplies the power required for the initial operation of the wind turbine (110) and then charges the power generated from the operated wind turbine (110), and is formed by recycling batteries used for drones. In other words, the regenerated battery is formed by collecting waste batteries used for drones, selecting battery cells that exhibit performance exceeding a set standard value, and combining the selected battery cells.

[0045] The process of forming the above-mentioned regenerated battery (140) includes, as illustrated in FIG. 3, a waste battery collection step, a cell separation step of separating battery cells of the collected waste batteries, a performance evaluation step of each separated battery cell, a selection step of selecting battery cells that maintain a performance higher than a reference value after the performance evaluation, and a reassembly step of manufacturing a new battery by combining only the selected battery cells.

[0046] As shown in Figure 4, when a battery pack used for a drone reaches the end of its service life and is replaced, the replaced battery pack is collected, the performance of each battery cell is measured, and battery cells that exceed a certain standard are selected.

[0047] It is used as a storage device that stores the generated power of a small wind turbine (110) by combining selected battery cells.

[0048] The above-mentioned regenerative battery (140) not only charges the generated power of the small wind turbine (110), but can also be used as a supply device to supply the power required for the initial operation of the wind turbine (110). If the state of charge (SOC: State Of Charge) of the above-mentioned regenerative battery (140) is low and the power required for the initial operation of the wind turbine (110) is insufficient, a certain amount of power can be charged by receiving power from the grid until the power required for the initial operation of the wind turbine (110) is filled.

[0049] The first converter (150) converts direct current (DC) power supplied from the regenerative battery (140) into alternating current (AC) power during the initial operation of the wind turbine (110) and supplies the power to the wind turbine (110), and may include a high-output inverter that operates with a capacity capable of handling the driving power of the wind turbine (110).

[0050] The second converter (160) supplies power required for the system from the regenerative battery (140) to the load (120) at a constant rate. In addition, the second converter (160) is connected between the regenerative battery (140) and the wind turbine (110), executes CVCF (Constant Voltage Constant Frequency) control operation, stores power generated from the wind turbine (110) in the regenerative battery (140), and supplies power required for the system from the regenerative battery (140) to the load (120) at a constant rate according to a power command from a power management system (PMS) or an energy management system (EMS).

[0051] FIG. 5 is a flowchart for explaining a process in which electric energy generated from a wind turbine is stored in a regenerative battery through an energy storage system (100) for a small wind turbine using a waste battery for a drone according to an embodiment of the present invention.

[0052] Referring to FIG. 5, in the energy storage system (100) for a small wind turbine using a spent battery for a drone according to an embodiment of the present invention, power required for initial operation is first supplied from a regenerative battery (140) to a first converter (150).

[0053] The first converter (150) converts the direct current power required for initial operation into alternating current power and supplies it to the wind turbine (110). At this time, the first converter (150) operates with a capacity capable of handling the driving power of the wind turbine (110) through a high-output inverter, and sequentially supplies driving power to multiple wind turbines.

[0054] After the wind turbine (110) is driven and power is generated by wind power, the first converter (150) converts the power generated by the wind turbine (110) into direct current power and charges the regenerative battery (140).

[0055] As described above, when the regenerative battery (140) has a low state of charge (SOC: State Of Charge) and thus lacks the power required for the initial operation of the wind turbine (110), it receives power from the grid and charges a certain amount of power until the power required for the initial operation of the wind turbine (110) is filled, and the first converter (150) is connected between the regenerative battery (140) and the wind turbine (110), executes CVCF (Constant Voltage Constant Frequency) control operation, and stores the power generated from the wind turbine (110) in the regenerative battery (140).

[0056] The second converter (160) supplies the power required for the system from the regenerative battery (140) to the load (120) at a constant rate. At this time, the second converter (160) is connected between the grid (130) and the regenerative battery (140), and executes a power quality (PQ) control operation so that power is not supplied from the grid (130) to the wind turbine (110) during operation. In addition, the second converter (160) supplies the power required for the system from the regenerative battery (140) at a constant rate to the load (120) according to a power command from a power management system (PMS) or an energy management system (EMS).

[0057] The energy storage system for a small wind turbine using a waste drone battery according to the present invention described above selects battery cells having a certain level of performance or higher from a waste battery pack used for a drone, and then reassembles the selected battery cells to form a regenerated battery that can be used as an energy storage device that stores the generated power of a small wind turbine, thereby reducing environmental pollution caused by the disposal of waste batteries to be disposed of and lowering the manufacturing cost of the energy storage device.

[0058]

[0059] FIG. 6 is a schematic diagram of a wind power generation prediction and monitoring system according to another preferred embodiment of the present invention, which may include a wind power data collection device (1100), a network (1200), a wind power generation prediction and monitoring server (1300), a wind power database (DB) (1400), and a cloud platform (1500).

[0060] The wind power data collection device (1100) collects wind power data for predicting wind power generation.

[0061] Here, the wind power data collection device (1100) can collect wind power data by using at least one of a wind power plant server, a weather station, a past wind power data database storing past wind power data, and Lidar, as illustrated in FIG. 7.

[0062] In particular, the present invention can improve prediction accuracy and reliability by collecting high-resolution wind speed and wind direction data using lidar.

[0063] The network (1200) serves to communicate data between the wind power data collection device (1100) and the wind power generation prediction and monitoring server (1300), and may use a wired or wireless communication network.

[0064] The wind power generation prediction and monitoring server (1300) preprocesses wind power data collected from the wind power data collection device (1100), learns a wind power generation prediction model generated based on artificial intelligence based on the preprocessed data, generates wind power generation prediction data based on the learning results, and compares actual wind power generation data with the wind power generation prediction data to optimize and distribute the wind power generation prediction model based on the error.

[0065] This wind power generation prediction and monitoring server (1300) generates learning wind power generation data based on preprocessed wind power data, extracts features by capturing spatial characteristics from the generated wind power generation data, and learns the extracted features to capture temporal dependencies for predicting wind power generation, thereby predicting wind power generation.

[0066] The above wind power generation prediction and monitoring server (1300), as illustrated in FIG. 8, may include a data collection module (1301) that collects wind power data for predicting wind power generation through the wind power data collection device (1100), a preprocessing module (1302) that preprocesses the wind power data collected by the data collection module (1301) to create a learning data set, a learning module (1303) that learns a wind power generation prediction model generated based on artificial intelligence based on the learning data set generated by the preprocessing module (1302) and creates wind power generation prediction data based on the learning result, and a control module (1304) that compares the wind power generation prediction data generated through the learning module (1303) with actual wind power generation data, optimizes the wind power generation prediction model according to an error, and controls distribution of the wind power generation prediction model.

[0067] The above learning module (1303) can predict wind power generation using a CGAN-CNN-LSTM wind power generation prediction model that combines a CGAN (Conditional Generative Adversarial Network) algorithm that generates learning wind power generation data by correcting missing values ​​and outliers based on preprocessed wind power data, a CNN (Convolutional Neural Network) algorithm that extracts features by capturing spatial characteristics from the generated wind power generation data, and an LSTM (Long Short-Term Memory) algorithm that learns the extracted features to capture temporal dependencies for predicting wind power generation and predicts wind power generation.

[0068] The above control module (1304) compares the actual wind power generation data with the wind power generation prediction data to calculate an error, compares the calculated error with a threshold set for updating the wind power generation prediction model, and if the calculated error is less than the threshold, distributes the wind power generation prediction model as a prediction model, and if the calculated error is greater than the threshold, optimizes the wind power generation prediction model through learning.

[0069] The above error calculation can be performed by calculating the error using the mean absolute error (MAE), which calculates the average value of the absolute difference between the wind power generation prediction data and the actual wind power generation data as the error, or by calculating the error using the root mean square deviation (RMSD) or the root mean square error (RMSE).

[0070] In addition, the control module (1304) switches the power circuit between the wind turbine and the battery to ground when power generation that the battery cannot handle is predicted to occur momentarily due to strong winds, etc., thereby stopping battery charging, and when power generation is predicted to be normal again, the power circuit switched to ground is switched back to the battery to protect the battery (ESS).

[0071] In addition, the wind power generation prediction and monitoring server (1300) may include an evaluation module (1305) that verifies the wind power generation prediction module using a verification data set and, if the performance of the wind power generation prediction module is below a set value, fine-tunes the hyper parameters of the wind power generation prediction model to improve the wind power generation prediction module, and a model distribution module that distributes an optimized wind power generation prediction model through a cloud platform (1500) according to the control of the control module (1304).

[0072] The above control module (1304) can update the wind power generation prediction model distributed to the cloud platform (1500) through the model distribution module (1306) whenever the wind power generation prediction module is improved.

[0073] The wind power database (1400) serves to store optimized wind power generation prediction models, actual wind power generation data, wind power generation prediction data, etc.

[0074] The cloud platform (1500) serves to provide an interface so that power generation operators can access the wind power generation prediction model distributed from the wind power generation prediction and monitoring server (1300).

[0075] These cloud platforms (1500) can provide API settings, data security, privacy protection, and a user-friendly interface that allows users to easily interact with the system.

[0076] Although not shown in the drawing, the power generation operator's operator terminal can be further connected to enable overall system operation.

[0077] A power generation business operator can request a wind power generation forecast by accessing the cloud platform (1500) using a business terminal, or can download a wind power generation forecast model provided by the cloud platform (50) and forecast the wind power generation directly on his / her terminal. Here, the business terminal may use a terminal such as a PC, a smartphone, a laptop, a tablet PC, etc., or may use a wearable device. The business terminal may be a terminal owned by a member who has registered as a member on the site that operates the wind power generation forecasting and monitoring server (1300). Here, the business terminal can download and use an application for wind power generation forecasting through the cloud platform (1500), and when the wind power generation application is downloaded, installed, and operated on the terminal, the forecasted wind power generation can be confirmed through the cloud platform (1500).

[0078] The operation of the wind power generation prediction and monitoring system according to a preferred embodiment of the present invention configured as described above is described in detail as follows.

[0079] First, the wind power data collection device (1100) collects wind power data for wind power generation prediction in various ways and provides it to the wind power generation prediction and monitoring server (1300).

[0080] That is, wind power data can be collected from wind power farm servers, weather data can be collected as wind power data through weather observation stations, wind power data can be collected through a historical wind power data database where historical wind power data is stored, and wind power data can be collected using equipment such as Lidar.

[0081] While Lidar is typically used to collect distance data, it can also be used to collect high-resolution wind speed and direction data. Lidar emits continuous laser (continuous wave) waves and measures the Doppler effect to obtain wind direction and speed data. Using Lidar, high-precision wind direction and speed data can be obtained with just 50 samples at a specific altitude, ultimately improving the accuracy and reliability of wind power generation forecasts.

[0082] The above wind power generation prediction and monitoring server (1300) collects wind power data from the wind power data collection device (1100), learns a preset wind power generation prediction model using artificial intelligence and machine learning technology, determines an optimal wind power generation prediction model for wind power generation prediction, and distributes it to business operators for use through a cloud platform (1500).

[0083] For example, the data collection module (1301) of the wind power generation prediction and monitoring server (1300) collects wind power data by linking with the wind power data collection device (1100) via a network (1200) and transmits the data to the preprocessing module (1302).

[0084] The above preprocessing module (1302) preprocesses the collected wind power data to create a learning data set and transmits it to the learning module (1303).

[0085] Here, the preprocessing module (1302) preprocesses the collected wind power data to enable learning using a pre-generated wind power generation prediction model, thereby generating a learning data set. This preprocessing process can ensure quality and consistency by tidying up and transforming the collected data. In particular, the preprocessing process can perform missing value processing, outlier removal, and data normalization or scaling as needed. It also serves to divide the preprocessed data into learning, verification, and test data sets. Figure 9a is a wind power generation data sheet including missing data before preprocessing, and Figure 9b is a wind power generation data sheet after the missing data has been supplemented through preprocessing.

[0086] The data generated through this preprocessing process is input into a learning module (1303), which learns a wind power generation prediction model generated based on artificial intelligence, and generates wind power generation prediction data based on the learning results.

[0087] For example, the learning module (1303) predicts wind power generation using a CGAN-CNN-LSTM wind power generation prediction model that combines a Conditional Generative Adversarial Network (CGAN) algorithm that generates learning wind power generation data through missing value and outlier correction based on preprocessed wind power data, a CNN (Convolutional Neural Network) algorithm that extracts features by capturing spatial characteristics from the generated wind power generation data, and an LSTM (Long Short-Term Memory) algorithm that learns the extracted features to capture temporal dependencies for predicting wind power generation, thereby predicting wind power generation.

[0088] Preferably, the learning module (1303) predicts wind power generation using a CGAN-CNN-LSTM wind power generation prediction model comprised of artificial intelligence and machine learning technologies.

[0089] First, the architecture is designed, hyperparameters are defined, and appropriate activation functions and loss metrics are selected. In the first step of the wind power generation prediction model, the CGAN algorithm is used to preprocess wind power data and generate training wind power generation data, which is actual wind data, through missing value and outlier correction. Next, the CNN algorithm is used to capture spatial characteristics from the wind power generation data generated by the CGAN algorithm and extract features. The feature data reflecting the extracted spatial characteristics are input into the time series of the LSTM algorithm. Finally, the LSTM algorithm learns the extracted features to capture temporal dependencies (patterns) for wind power generation prediction and predicts wind power generation. By using the LSTM neural network algorithm, the model's convergence speed can be improved, thereby reducing the wind power generation prediction time when predicting wind power generation using the computing resources of power generation users.

[0090] In this way, by predicting wind power generation by sufficiently considering both spatial and temporal characteristics of wind power data, it is possible to generate accurate and reliable wind power generation prediction data.

[0091] In addition, the evaluation module (1305) verifies the wind power generation prediction module using the verification data set generated separately from the preprocessing module (1302), and if the performance of the wind power generation prediction module is below a set value, the evaluation module improves the wind power generation prediction module by fine-tuning the hyper parameters of the wind power generation prediction model. After verifying the wind power generation prediction module using the verification data set and achieving satisfactory performance, the wind power generation prediction model is tested on the test data set separated from the preprocessing module (1302) to evaluate the prediction accuracy, reliability, and robustness in an actual scenario.

[0092] Meanwhile, when wind power generation prediction data is generated through the above process, the control module (1304) compares the actual wind power generation data acquired through the data collection module (1301) with the wind power generation prediction data, optimizes the wind power generation prediction model according to the error, and distributes it.

[0093] For example, the control module (1304) compares the actual wind power generation data with the predicted wind power generation data to calculate an error.

[0094] Here, the error can be calculated using the mean absolute error (MAE), which calculates the average value of the absolute difference between the wind power generation prediction data and the actual wind power generation data as the error, or using the root mean square deviation (RMSD) or the root mean square error (RMSE).

[0095] The error calculated in this way is compared with a threshold set in advance for updating the wind power generation prediction model. If the calculated error is less than the threshold, the wind power generation prediction model is distributed as a prediction model, and the wind power generation prediction data and actual wind power generation data are mapped and stored in a wind power database (1400).

[0096] Conversely, if the calculated error exceeds the threshold, the wind power generation prediction model is optimized through learning. Optimization of the wind power generation prediction model can be achieved through hyperparameter adjustment using the evaluation module (1305) and learning using more wind power data. Optimization of the wind power generation prediction model is performed to minimize the difference between the wind power generation prediction data and the actual wind power generation data.

[0097] Once the wind power generation prediction model is optimized through this process, the optimized wind power generation prediction model is distributed through a cloud platform (1500) using a model distribution module (1306).

[0098] In addition, the control module (1304) updates the wind power generation prediction model distributed to the cloud platform (1500) through the model distribution module (1306) whenever the wind power generation prediction module is improved.

[0099] The cloud platform (1500) provides an interface for power generation operators to access the wind power generation prediction model deployed from the wind power generation prediction and monitoring server (1300). Specifically, it provides API settings, data security, privacy protection, and a user-friendly interface that allows easy interaction between the system and users, allowing power generation operators to easily access the cloud platform (1500) using their operator terminals, download a wind power generation prediction application, or update a wind power generation prediction model.

[0100] In addition, the control module (1304) switches the power circuit between the wind turbine and the battery to ground when power generation that the battery cannot handle is predicted to occur momentarily due to strong winds, etc., thereby stopping battery charging, and when power generation is predicted to be normal again, the power circuit switched to ground is switched back to the battery to protect the battery (ESS).

[0101] According to the present invention described above, by accurately predicting wind power generation by considering both spatial and temporal characteristics of meteorological data, it is possible to improve the competitiveness of wind power generation in the electricity market and to help business operators make decisions, thereby enhancing their competitiveness.

[0102] Additionally, accurate prediction of wind power generation can help maintain grid stability and ensure efficient resource allocation.

[0103] The description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the present invention. Therefore, the present invention is not intended to be limited to the disclosed embodiments, but is to be construed in the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A wind turbine that generates electricity by wind power; A recycled battery formed using waste batteries used to drive a drone, for storing power generated from the above-mentioned wind turbine; Energy storage system for small wind turbines utilizing waste drone batteries.

2. In Paragraph 1, The above-mentioned regenerated battery is Characterized by collecting waste batteries used in drones, selecting battery cells that exhibit performance exceeding a set standard value, and combining the selected battery cells. Energy storage system for small wind turbines utilizing waste drone batteries.

3. In Paragraph 2, The above-mentioned regenerated battery is Waste battery collection stage; A cell separation step for separating battery cells of collected waste batteries; A step of conducting a performance evaluation of each separated battery cell; A selection step for selecting battery cells that maintain performance above a standard value after performance evaluation; Characterized by being formed through a reassembly step of manufacturing a new battery by combining only selected battery cells. Energy storage system for small wind turbines utilizing waste drone batteries.

4. In Paragraph 1, A load that consumes power generated from the above wind turbine; Further comprising a grid that controls the power charged in the above-mentioned regenerative battery to be supplied to the above-mentioned load, The above-described regenerative battery is characterized by being formed to supply power required for the initial operation of the wind turbine and to charge power generated from the subsequently operated wind turbine. Energy storage system for small wind turbines utilizing waste drone batteries.

5. In Paragraph 4, A first converter that converts direct current (DC) power supplied from the regenerated battery into alternating current (AC) power and supplies it to the wind turbine during the initial operation of the wind turbine, and Characterized by further including a second converter that consistently supplies power required by the grid from the above-mentioned regenerated battery to the above-mentioned load. Energy storage system for small wind turbines utilizing waste drone batteries.

Images