KR20240010803A - System and method for controlling artificial intelligence smart wind power capable of power prediction - Google Patents

Abstract

The artificial intelligence smart wind power control method capable of predicting power according to the embodiment includes the steps of (A) building a digital twin of the wind power generator installed in the analysis area where the wind power generator is installed; (B) conducting an experiment to estimate the average power production of a wind turbine in the constructed digital twin and collecting test result data; (C) predicting the amount of power generation according to the climatic conditions of the wind turbine installation spot based on the collected test result data; (D) determining the power generation prediction results and the power generation according to actual climate conditions and feeding back the power generation prediction results; and (E) performing wind power generation control according to predicted climate conditions; Includes.

Description

Artificial intelligence smart wind power system and control method capable of predicting power {SYSTEM AND METHOD FOR CONTROLLING ARTIFICIAL INTELLIGENCE SMART WIND POWER CAPABLE OF POWER PREDICTION}

This disclosure relates to an artificial intelligence smart wind power system and control method capable of predicting power. Specifically, it controls the wind power generator by predicting local weather near the wind power generator based on artificial intelligence, and provides a simulation for predicting the amount of power generated on a digital twin. It relates to systems and methods.

Unless otherwise indicated herein, the material described in this section is not prior art to the claims of this application, and is not admitted to be prior art by inclusion in this section.

Wind power generation is a power generation method that uses the power of the wind to turn a generator to produce electrical energy. Unlike thermal power generation, it does not emit greenhouse gases or is depleted, and is also economically efficient, so it is widely used as clean energy. It is expected that it will continue to play a significant role as an alternative energy solution to address energy depletion and abnormal climate.

Since small wind power generators produce less power per generator, a large amount of wind power generators must be installed to satisfy the required power demand. Because many wind turbines are difficult to maintain, protection of wind turbines through a control system is essential. In addition, small wind power generators installed on medians and shoulders are difficult to manage and maintain, so preemptive protection using a control system is essential.

To protect wind turbines, a system that can accurately predict wind speeds at the location where wind turbines are installed is needed to build a control system that prevents wind turbines from being damaged by winds exceeding critical wind speeds. However, because it is difficult to predict the amount of power generation through actual experiments, the demand for technology to predict power generation through simulation on the digital twin by building a digital twin is increasing.

1. Korean Patent Registration No. 10-2143757 (2020.08.06) 2. Korean Patent Registration No. 10-1844044 (2018.03.26)

The artificial intelligence smart wind power system and control method capable of predicting power according to the embodiment builds a weather dataset and builds a GRU and two-way RNN-based climate prediction model to provide an artificial intelligence-based climate prediction system.

Additionally, the embodiment provides a wind turbine control system and an interlocking function between the wind turbine control system and the climate prediction system to develop and build a wind turbine control system.

In the embodiment, a power generation prediction simulation is provided on a digital twin, and the wind power generator is controlled according to predicted climate information to improve the efficiency and stability of the wind power generator. For this purpose, the embodiment provides an RNN-based deep learning model that predicts the future weather environment, mainly wind direction, wind speed, precipitation probability, and precipitation amount, in a local area using weather data such as current temperature, humidity, wind direction, and wind speed. do.

In addition, the embodiment provides a power prediction simulation on a digital twin through a wind power generator's average power production estimation experiment.

In the embodiment, artificial intelligence input data is generated by integrating regional weather data from the Korea Meteorological Administration data and local weather data measured from wind power generators, and missing data from the Korea Meteorological Administration is restored through non-linear interpolation.

In the embodiment, the wind direction and wind speed around the location where the generator is installed are periodically predicted, and the blade rotation direction of the generator is controlled according to the predicted wind direction and wind speed data to perform pitch and yaw control. In addition, in the embodiment, local weather data is measured and collected through a small sensor attached to a wind power generator, and local climate data along with regional weather data are additionally input into a neural network to predict local climate data around the wind power generator. make it possible

The artificial intelligence smart wind power control method capable of predicting power according to the embodiment includes the steps of (A) building a digital twin of the wind power generator installed in the analysis area where the wind power generator is installed; (B) conducting an experiment to estimate the average power production of a wind turbine in the constructed digital twin and collecting test result data; (C) predicting the amount of power generation according to the climatic conditions of the wind turbine installation spot based on the collected test result data; (D) determining the power generation prediction results and the power generation according to actual climate conditions and feeding back the power generation prediction results; and (E) performing wind power generation control according to predicted climate conditions; Includes.

An artificial intelligence smart wind power system capable of predicting power according to another embodiment includes a test data collection module that performs a test to estimate the average power production of a wind power generator in a digital twin and collects test result data; A power generation calculation module that predicts power generation according to climate conditions based on collected test result data and feeds back the power generation prediction results according to power generation figures according to actual climate conditions; A control module that individually controls wind power generators according to predicted weather conditions and power generation; Includes.

The artificial intelligence smart wind power system and control method capable of predicting power as described above enables more accurate climate information prediction through local weather data around wind power generators. In addition, the efficiency and stability of the wind power generator can be improved by optimally controlling the wind power generator according to accurately predicted climate information.

In addition, through simulation on the digital twin, small and medium-sized wind power generators can be controlled more accurately according to the weather environment, thereby improving wind power generation efficiency.

The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention.

Figure 1 is a diagram showing the configuration of an artificial intelligence smart wind power system capable of power prediction according to an embodiment
Figure 2 is a diagram showing the data processing configuration of the server 200 according to an embodiment.
Figure 3 is a diagram showing the simulation process of an artificial intelligence smart wind power system capable of power prediction according to an embodiment.
Figure 4 is a diagram showing the wind power generator control flow of an artificial intelligence smart wind power system capable of power prediction according to an embodiment.
Figure 5 is a diagram showing the wind power generator control process of an artificial intelligence smart wind power system capable of power prediction according to an embodiment.
Figures 6 and 7 are diagrams showing an example of climate prediction data calculation of an artificial intelligence smart wind power system capable of predicting power according to an embodiment.
8 to 10 are diagrams showing display screens visualizing climate prediction data according to an embodiment.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete and to provide common knowledge in the technical field to which the present invention pertains. It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

In describing embodiments of the present invention, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description will be omitted. The terms described below are terms defined in consideration of functions in embodiments of the present invention, and may vary depending on the intention or custom of the user or operator. Therefore, the definition should be made based on the contents throughout this specification.

Figure 1 is a diagram showing the configuration of an artificial intelligence smart wind power system capable of power prediction according to an embodiment.

Referring to FIG. 1, an artificial intelligence smart wind power system according to an embodiment may be configured to include a wind power generator 100, a server 200, and an administrator terminal 300. The wind power generator 100 senses weather data in the local area where it is installed and transmits it to the server 200 and the manager terminal 300, and performs pitch and yaw control according to the control signal received from the server 200 or the manager terminal 300. This can be done. In an embodiment, weather data may include temperature, humidity, wind direction, wind speed, amount of sunlight, etc. in the area where the wind power generator is installed.

In the embodiment, at least one manager terminal 300 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a laptop equipped with a navigation system and a web browser, a desktop, a laptop, etc. At this time, at least one manager terminal 300 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one manager terminal 300 is, for example, a wireless communication device that guarantees portability and mobility, and includes navigation, personal communication system (PCS), global system for mobile communications (GSM), personal digital cellular (PDC), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) ) It may include all types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs.

The artificial intelligence smart wind power system capable of predicting power according to the embodiment builds a weather dataset and uses a climate prediction model based on GRU (Gated Recurrent Unit) and bi-directional RNN (Recurrent Neural Network) to provide an artificial intelligence-based climate prediction system. Build it. In addition, in order to develop and build a wind turbine control system, it provides a wind turbine control system and an interconnection function between the wind turbine control system and the climate prediction system. In addition, the embodiment provides a power generation prediction simulation on a digital twin, and controls the wind power generator according to predicted climate information to improve the efficiency and stability of the wind power generator. For this purpose, the embodiment provides an RNN-based deep learning model that predicts the future weather environment, mainly wind direction, wind speed, precipitation probability, and precipitation amount, in a local area using weather data such as current temperature, humidity, wind direction, and wind speed. do.

In addition, the embodiment provides a power prediction simulation on a digital twin through a wind power generator's average power production estimation experiment. In the embodiment, artificial intelligence input data is generated by integrating regional weather data from the Korea Meteorological Administration data and local weather data measured from wind power generators, and missing data from the Korea Meteorological Administration is restored through non-linear interpolation. Additionally, artificial intelligence according to the embodiment may use GRU and bidirectional RNN-based climate prediction models. In the embodiment, the wind direction and wind speed around the location where the generator is installed are periodically predicted through artificial intelligence, and the blade rotation direction of the generator is controlled according to the predicted wind direction and wind speed data to perform pitch and yaw control. do. In addition, in the embodiment, local weather data is measured and collected through a small sensor attached to a wind power generator, and local climate data along with regional weather data are additionally input into a neural network to predict local climate data around the wind power generator. make it possible

Figure 2 is a diagram showing the data processing configuration of the server 200 according to an embodiment.

Referring to FIG. 2, the server 200 according to the embodiment may be configured to include a test data collection module 210, a power generation calculation module 220, a control module 230, and a display module 240. The term 'module' used in this specification should be interpreted to include software, hardware, or a combination thereof, depending on the context in which the term is used. For example, software may be machine language, firmware, embedded code, and application software. As another example, hardware may be a circuit, processor, computer, integrated circuit, integrated circuit core, sensor, Micro-Electro-Mechanical System (MEMS), passive device, or a combination thereof.

The data collection module 210 performs an experiment to estimate the average power production of a wind turbine in the digital twin and collects test result data. In the embodiment, the data collection module 210 sets control conditions for the wind power generator according to the wind direction and wind speed at the point where the wind power generator is located, performs an average power production estimation experiment, and collects test result data.

The power generation calculation module 220 predicts power generation according to climate conditions based on the collected test result data, and compares the power generation prediction results with the power generation figures according to actual climate conditions. In the embodiment, the power generation prediction results are fed back according to the calculated power generation figures according to actual climate conditions, enabling more accurate power generation prediction as data is accumulated. In the embodiment, the power generation calculation module 220 builds a weather dataset including temperature, pressure, and wind speed according to the local area where the wind generator is installed and time for climate prediction, and builds a climate prediction model based on GRU and bidirectional RNN. . In addition, in the embodiment, a climate prediction model based on a deep learning method including an Artificial Neural Network (ANN), a Deep Neural Network (DNN), a Convolution Neural Network (CNN), and a Recurrent Neural Network (RNN) is used. can be built

In an embodiment, the power generation calculation module 220 may perform climate prediction and power generation prediction through analysis of collected test result data and cumulative prediction data analysis using artificial intelligence machine learning. In the embodiment, the power generation calculation module 220 performs an out of distribution detection process other than learning to process unlearned patterns in addition to noise response. Non-learning distribution data detection is to identify whether the data input to artificial intelligence is learned probability distribution data. In the embodiment, stability and reliability can be increased by filtering out or exception processing images that are difficult for the artificial neural network to judge through detection of distribution data other than learning. In the embodiment, in order to detect non-learning distribution data, a probability value indicating how confident the deep learning decision is is calibrated or non-learning distribution data is generated and learned using a generative adversarial network (GAN). This helps improve detection accuracy.

Additionally, in the embodiment, in order to reduce the size of the model while maintaining prediction accuracy, a lightweight deep learning technology that simplifies calculations is used to finalize climate and power generation predictions. In the embodiment, for image recognition, a convolutional filter is modified in a convolutional neural network (CNN) to reduce the operation dimension, or pruning and weight values to delete weights of a neural network that do not have a significant impact are performed. A quantization process is performed to simplify calculations by reducing the number of floating point numbers, enabling data lightweighting. Additionally, in the embodiment, the output of a pre-trained large neural network is imitated and learned by a small neural network to simplify computation and maintain accuracy.

The control module 230 individually performs wind power generation control according to predicted weather conditions and power generation amount. In an embodiment, when the predicted wind speed exceeds a set threshold, the control module 230 performs pitch control to rotate the blades of the wind turbine by controlling the angle of the blades so that they have the least resistance to the wind. . In addition, when the predicted wind speed is less than the set threshold, yaw control is performed to produce maximum wind power generation efficiency using the predicted wind direction information,

When a wind turbine is generating power, the current wind speed can be inversely calculated using the instantaneous power generation amount.

The display module 240 matches a visual object to each of the weather data including temperature, atmospheric pressure, and wind speed, adjusts the direction and size of the matched visual object according to the scalar amount of the collected weather data, and displays the collected weather data. Displays visual objects and power generation in real time.

Figure 3 is a diagram showing the simulation process of an artificial intelligence smart wind power system capable of power prediction according to an embodiment.

Referring to Figure 3, in step S100, a digital twin of the wind power generator is built in the analysis area where the wind power generator is installed, an experiment to estimate the average power production of the wind power generator is performed on the constructed digital twin, and test result data is collected.

In the S200 stage, power generation is predicted according to climate conditions based on the collected test result data, and power generation figures are calculated according to the power generation prediction results and actual climate conditions. In the embodiment, in step S200, control conditions for the wind power generator can be set according to the wind direction and wind speed at the point where the wind power generator is located. In the embodiment, a deep learning method-based climate prediction model is built including ANN (Artificial Neural Network), DNN (Deep Neural Network), CNN (Convolution Neural Network), and Recurrent Neural Network (RNN). . In addition, for climate prediction, we build a weather dataset that includes temperature, pressure, and wind speed over time and the local area where the wind generator is installed, and build a climate prediction model based on GRU and two-way RNN.

In the S300 step, the power generation prediction results and the power generation figures calculated according to actual climate conditions are accumulated and the power generation prediction results are fed back. In step S400, wind power generators are individually controlled according to predicted climate conditions. In the embodiment, when the predicted wind speed exceeds a set threshold, pitch control is performed to rotate the blades of the wind turbine by controlling the angle of the blades so that they have the least resistance to the wind. Additionally, when the predicted wind speed is less than a set threshold, yaw control can be performed to produce maximum wind power generation efficiency using the predicted wind direction information. In the embodiment, when the wind power generator is generating power, the current wind speed is inversely calculated through the instantaneous power generation amount.

In step S500, a visual object is matched to each meteorological data including temperature, air pressure, and wind speed, the direction and size of the matched visual object are adjusted according to the scalar amount of the collected meteorological data, and a visual object representing the collected meteorological data is adjusted. and power generation amount are displayed in real time.

Figure 4 is a diagram showing the wind power generator control flow of an artificial intelligence smart wind power system capable of power prediction according to an embodiment.

Referring to FIG. 4, in step S10, the test data collection module collects wide-area data including temperature, humidity, pressure, wind direction, and wind speed from the Korea Meteorological Administration server, and determines whether missing data exists in step S13. If missing data exists, step S15 is entered and regional weather data is supplemented through non-linear interpolation. In step S17, the test data collection module collects the supplemented wide-area weather data and the wide-area weather data collected from the Korea Meteorological Administration server. In step S12, local weather data including temperature and pressure is collected from the weather data collection module. Afterwards, in step S14, it is determined whether the wind power generator is generating power. If so, step S16 is entered and the prediction module traces back the current wind speed from the power generation amount and calculates it. If the wind power generator that collected local weather data in step S14 is not generating power, step S26 is entered to determine whether missing data exists among the collected local weather data. If missing data exists, proceed to step S28 to supplement the local weather data through non-linear interpolation, and collect the supplemented local weather data in step S28. In step S26, if it is determined that there is no missing data in the weather data supplementation module, step S30 is entered to collect local weather data corresponding to temperature and pressure in the weather data collection module.

In step S16, if the current wind speed is calculated in the power generation calculation module, in step S18

Collects temperature, pressure, and wind speed as local weather data in the test data collection module, and determines whether missing data exists in the power generation calculation module in step S20. If missing data exists, step S22 is entered and local weather data is supplemented through non-linear interpolation in the power generation calculation module. In step S24, the test data collection module collects local weather data such as temperature, pressure, and wind speed. In step S19, the power generation calculation module uses regional weather data collected in step S17, local weather data such as temperature, pressure, and wind speed collected in step S24 when the wind power generator is generating power, and in step S30 when the wind power generator is not generating power. The climate of the area where the wind turbine is installed is predicted using artificial intelligence through a GRU and bidirectional RNN-based climate prediction model using collected local weather data such as temperature and pressure data. In step S21, predicted wind direction and predicted wind speed data, which are prediction result values, are acquired. Afterwards, step S23 is entered to control the wind power generator through predicted wind direction and predicted wind speed data.

Figure 5 is a diagram showing the wind power generator control process of an artificial intelligence smart wind power system capable of power prediction according to an embodiment.

Referring to FIG. 5, in step S23, the control module determines whether the predicted wind speed value is greater than or equal to the critical wind speed. If the wind speed is above the critical wind speed, step S33 is entered to perform pitch control, and if the wind speed is below the critical wind speed, step S25 is entered. In step S25, the control module determines whether pitch control is being performed, and if pitch control is being performed, step S31 is performed to release pitch control. If pitch control is not being performed in step S25, step S27 is entered to determine whether the power generation efficiency of the wind generator is maximum. If the power generation efficiency of the wind generator is maximum, step S23 is re-entered, and if it is determined in step S27 that the power generation efficiency of the wind generator is not maximum, the yaw control can be performed by entering step S29.

Figures 6 and 7 are diagrams showing an example of climate prediction data calculation by an artificial intelligence smart wind power system capable of power prediction according to an embodiment.

Referring to Figures 6 and 7, in the embodiment, monthly climate prediction data and power generation data including wind volume, wind speed, temperature, power generation, etc. of the local area (spot) where the wind generator is installed are predicted and visualized in a chart, graph, It can be converted into a 3D image, etc.

8 to 10 are diagrams showing a display screen visualizing climate prediction data according to an embodiment.

Referring to FIGS. 8 and 9, the display module according to the embodiment matches each predicted climate data to each visual object such as an arrow, color, or three-dimensional figure, and changes the climate data to the size and color of the matched visual object. , so that it can be displayed by changing its shape. As shown in FIG. 8, climate change and climate change prediction amount can be more intuitively understood through visual objects through the artificial intelligence-based climate prediction wind power generator control system according to the embodiment. For example, wind speed is displayed by the size and length of an arrow, so that the wind speed in a local area where a wind power generator is installed can be identified by the size and color of the arrow. Also, in the embodiment, as shown in FIG. 10, the wind rose for each period of the wind power generator can be visualized and displayed so that it can be recognized over time.

The artificial intelligence smart wind power system capable of predicting power as described above enables more accurate climate information prediction through local weather data around wind power generators. In addition, the efficiency and stability of the wind power generator can be improved by optimally controlling the wind power generator according to accurately predicted climate information.

In addition, through simulation on the digital twin, small and medium-sized wind power generators can be controlled more accurately according to the weather environment, thereby improving wind power generation efficiency.

The disclosed content is merely an example, and various modifications and implementations may be made by those skilled in the art without departing from the gist of the claims, so the scope of protection of the disclosed content is limited to the above-mentioned specific scope. It is not limited to the examples.

Claims (11)

In the artificial intelligence smart wind power control method capable of predicting power,
(A) Building a digital twin of the wind turbine installed in the analysis area where the wind turbine is installed;
(B) conducting an experiment to estimate the average power production of a wind turbine in the constructed digital twin and collecting test result data;
(C) predicting the amount of power generation according to the climatic conditions of the wind turbine installation spot based on the collected test result data;
(D) determining the power generation prediction results and the power generation according to actual climate conditions and feeding back the power generation prediction results; and
(E) performing wind power generation control according to predicted climate conditions; An artificial intelligence smart wind power control method capable of predicting power including.
The method of claim 1, wherein step (C); Is
Setting wind power generator control conditions according to the wind direction and wind speed of the spot where the wind power generator is located; An artificial intelligence smart wind power control method capable of predicting power, comprising:
The method of claim 1, wherein step (B); Is
(B-1) Deep learning method-based climate prediction model including ANN (Artificial Neural Network), DNN (Deep Neural Network), CNN (Convolution Neural Network), and RNN (Recurrent Neural Network) Building steps; An artificial intelligence smart wind power control method capable of predicting power, comprising:
The method of claim 3, wherein the step (B-1); Is
For climate prediction, constructing a meteorological dataset including temperature, pressure, and wind speed over time and the local area where the wind generator is installed, and constructing a climate prediction model based on GRU and bidirectional RNN; An artificial intelligence smart wind power control method capable of predicting power, comprising:
The method of claim 1, wherein step (C); Is
Matching a visual object to each piece of weather data including temperature, air pressure, and wind speed, and adjusting the direction and size of the matched visual object according to the scalar amount of the collected weather data; and
Displaying visual objects representing collected weather data and power generation in real time; An artificial intelligence smart wind power control method capable of predicting power, comprising:
The method of claim 1, wherein step (E); Is
If the predicted wind speed exceeds a set threshold, performing pitch control to rotate the blades of the wind turbine by controlling the angle of the blades so that they have the least resistance to the wind;
If the predicted wind speed is less than a set threshold, performing yaw control to produce maximum wind power generation efficiency using predicted wind direction information; and
An artificial intelligence smart wind power control method capable of predicting power, characterized by inversely calculating the current wind speed through the instantaneous power generation amount when the wind power generator is generating power.
In an artificial intelligence smart wind power system capable of predicting power,
A test data collection module that conducts a test to estimate the average power production of a wind turbine in the digital twin and collects test result data;
A power generation calculation module that predicts power generation according to climate conditions based on collected test result data and feeds back the power generation prediction results according to power generation figures according to actual climate conditions;
A control module that individually controls wind power generators according to predicted weather conditions and power generation amount; An artificial intelligence smart wind power system capable of predicting power including.
The test data collection module of claim 7; silver
An artificial intelligence smart wind power system capable of predicting power, which sets control conditions for the wind power generator according to the wind direction and speed at the point where the wind power generator is located, performs an average power production estimation test, and collects test result data.
According to claim 7, wherein the power generation calculation module; silver
For climate prediction, an artificial intelligence capable of predicting power is characterized by building a weather dataset including temperature, pressure, and wind speed according to the local area and time where the wind generator is installed, and building a climate prediction model based on GRU and bidirectional RNN. Smart wind power system.
According to claim 7, an artificial intelligence smart wind power system capable of predicting power; silver
Match visual objects to each meteorological data including temperature, pressure, and wind speed, and adjust the direction and size of the matched visual objects according to the scalar amount of the collected meteorological data,
A display module that displays visual objects representing collected weather data and power generation in real time; An artificial intelligence smart wind power system capable of predicting power, further comprising:
According to claim 7, wherein the control module; silver
If the predicted wind speed exceeds the set threshold, pitch control is performed to rotate the blades of the wind turbine by controlling the angle of the blades so that they have the least resistance to the wind.
If the predicted wind speed is less than the set threshold, yaw control is performed to produce maximum wind power generation efficiency using the predicted wind direction information,
An artificial intelligence smart wind power system capable of power prediction that reversely calculates the current wind speed through the instantaneous power generation amount when the wind power generator is generating power.