CC BY
5КОМПЬЮТЕРНЫЕ И ИНФОРМАЦИОННЫЕ ТЕХНОЛОГИИ
(COMPUTER & INFORMATION TECHNOLOGIES)
УДК 004.8
Mammadov N.S.
laboratory assistant Azerbaijan State Oil and Industry University (Baku, Azerbaijan)
Aliyeva G.A.
laboratory assistant Azerbaijan State Oil and Industry University (Baku, Azerbaijan)
Kerimova S.M.
laboratory assistant Azerbaijan State Oil and Industry University (Baku, Azerbaijan)
APPLICATION OF ARTIFICIAL INTELLIGENCE IN WIND ELECTRIC INSTALLATIONS
Abstract: with the development of artificial intelligence (AI) technologies, new prospects are opening up for optimizing the operation of wind turbines. Wind energy is becoming an increasingly popular source of renewable energy, and the application of AI can significantly improve the efficiency, reliability and controllability of wind turbines. In this article, we will look at various aspects of the application of artificial intelligence in wind energy, including the optimization and management of wind turbines.
Keywords: artificial intelligence, wind turbine, wind energy, optimization, efficiency, reliability.
Artificial intelligence (AI) is a collection of technologies that enable computer systems to learn from data, analyze inferences, make decisions, and perform tasks without first attempting human intervention. In recent cases, AI has become more relevant in various aspects, and wind power is not excluded. The use of AI in wind turbines can increase productivity, improve energy efficiency and improve process control.
Forecasting wind conditions. Forecasting wind conditions is a key task for optimizing the operation of wind turbines. Thanks to AI, it is possible to create predictive models that take into account complex relationships between various variables, such as wind speed and direction, temperature, pressure, and other meteorological parameters. Machine learning methods such as time series methods, regression models, or neural networks can be used to predict wind conditions. To predict wind conditions based on historical data, for example, the ARIMA (Autoregressive Integrated Moving Average) time series method or LSTM (Long Short-Term Memory) neural networks can be used.
An example of wind speed prediction using ARIMA:
python 3 Copy code
import pandas as pd from statsmodels.tsa.arima.model import ARIMA
# Загрузка временного ряда данных о скорости ветра data = pd.read_csv( wind_speed_data.csv )
# Создание модели ARIMA и обучение на данных model = ARIMA(data[ wind_speed ], order={3, , i)) model_fit = model.fit()
# Прогноз на следующий период forecast = model_fit.forecasttsteps= ) pri nt(fore cast)
Optimization of wind turbine operation. Optimization of wind turbine operation allows achieving maximum energy generation efficiency under various wind
conditions. With the help of AI, it is possible to develop control algorithms that adapt the parameters of a wind turbine based on current data on wind speed and direction, temperature, and other parameters. The use of genetic algorithms or particle swarm algorithms makes it possible to effectively search for the optimal values of wind turbine parameters [1-4].
Diagnostics and predictive maintenance. Diagnostics of the state of a wind turbine is an important aspect of ensuring its reliable operation and preventing possible breakdowns. With the help of artificial intelligence, it is possible to develop diagnostic algorithms that, based on data from sensors and sensors, can identify early signs of equipment malfunction or wear.
An example of rotor speed analysis using Fast Fourier Transform (FFT):
python Q Озру oode
import numpy as np inport scipy.signal as signal
# Загрузка данных с вращении ритора Totox_5peed_data - np.loadtxtC rotor_5peed_data.c5v )
# Вычисление спектра частот frequencies, spectrum = signal.periodogran(rotor_speed_data, fs=sampling _ ra1
# Определение частоты вращения ротора rDtoi_5peed = frequencies[пр.argmaxtspectrum)]
printC"Частота вращения ротора:", гai:or_speed, "об/иин"J
Adaptation and self-learning. Wind turbines operate in various conditions, and adaptation and self-learning algorithms can be used to optimize their performance in various situations. The use of such algorithms allows the wind turbine to continuously improve its algorithms and adapt to new situations based on experience and data.
An example of a self-learning algorithm for optimizing the operation of a wind turbine:
python Q Copy oode
import numpy as np
from sklearn.ensemble import RardomForestRegressor
# Загрузка исторических данных с работе ветрсустановки data ~ пр.loadtxt( historical.data,csv )
X = data[:, :- ] у = data[:, - ]
# Создание и обучение «одели случайного леса
model = RandonFoTestRegressor[n_e5timator5= .ОС , гаndom_state=41) model.fittX, у)
# Получение данный о текущих ветровых условиях и состоянии ветроустановки current_data = пр.array([[wind_speed, wind .direct ion, temperature, pressure!
# Прогнозирование мощности генерации энергии на основе текущих данных и иодели power.output - model.predict(current.data}
printf'nporHOSHpyefflaa лепноеть генерации:", power_output)
These are just examples of the application of artificial intelligence in wind turbines. In real systems, the application of AI can be much more complex and extensive. However, these examples demonstrate which methods and algorithms can be used to optimize the operation of wind turbines using artificial intelligence. It is important to note that the implementation of such systems requires the collection of data and the adaptation of methods to the specific characteristics of the wind turbine [5-8].
Conclusion. The application of artificial intelligence in wind turbines opens up new prospects for optimizing operation, increasing efficiency and improving control. The examples given demonstrate how various machine learning methods, optimization algorithms and data analysis can be used to solve problems related to forecasting wind conditions, optimizing the operation of a wind turbine, diagnosing the state and adapting to various conditions. However, it should be noted that the implementation of such systems requires the collection of large amounts of data, their processing, and the development of complex models and algorithms. In the future, the development of artificial intelligence and its application in wind energy promise to improve the operation of wind turbines and promote sustainable energy development based on renewable energy sources.
REFERENCES:
1. Mammadov N.S., Ganiyeva N.A., Aliyeva G.A. "Role of Renewable Energy Sources in the World". Journal of Renewable Energy, Electrical, and Computer Engineering, Vol. 2, №2, pp 63 -67, Indonesia, 30 September 2022
2. Mammadov Nijat, Mukhtarova Kubra. "ANALYSIS OF THE SMART GRID
3. SYSTEM FOR RENEWABLE ENERGY SOURCES", Journal «Universum:
4. technical sciences», Vol. 107, № 2, pp. 64-67, Moscow, Russia, 27.02.2023.
5. N.S. Mammadov, "Methods for improving the energy efficiency of wind turbines at low wind speeds", Vestnik nauki, Issue 2, Vol. 61, №4, April 202 3
6. Nijat, Mammadov. "Selection of the type of electric generators for a wind electric installation", Universum journal, Vol.102, №9 pp.65 -67, Russia, September 2022.
7. Nijat Mammadov, "Analysis of systems and methods of emergency braking of wind turbines". International Science Journal of Engineering & Agriculture Vol. 2, № 2, pp. 147-152, Ukraine, April 2023
8. Nijat Mammadov, Sona Rzayeva, Nigar Ganiyeva, "Analisys of sync hronized asynchronous generator for a wind electric installation", Przeglad Elektrotechniczny journal, Vol. 2, № 5, pp.37 -40, Poland, May 2023
9. I.N. Rahimli, S.V. Rzayeva, E.E. Umudov, "DIRECTION OF ALTERNATIVE ENERGY", Vestnik nauki, Issue 2, Vol. 61, №4, April 2023
10. N.S. Mammadov, G.A. Aliyeva, "Energy efficiency improving of a wind electric installation using a thyristor switching system for the stator winding of a two-speed asynchronous generator", IJTPE, Issue 55, Volume 55, Number 2, pp. 285 -290
