Wind power generation has increased in the past twenty years due to the development of power electronic converters. Power generation through wind has advantages over other renewable sources, such as having more efficiency, being pollution-free, and its abundant availability. Power electronic converters play a vital role in the wind energy conversion system. This paper presents a wind-electric system with a permanent magnet synchronous generator, diode rectifier, DC-DC converter (buck-Boost or Cuk converter), and a three-phase five-level inverter. The five-level inverter is a modified form of a cascaded H-bridge inverter that uses a single DC source as an input irrespective of several levels and phases. As the wind speed changes, the Permanent Magnet Synchronous Generator (PMSG) voltage and frequency changes, but for practical applications, these changes should not be allowed; hence, a voltage controller is used that maintains the output voltage of a DC converter, andthus a constant AC output is obtained. The DClink voltage is maintained at the desired voltage by a Proportional plus Integral (PI)-based voltage controller. The DC link voltage fed to the multilevel inverter (MLI) is converted to AC to feed the load. The MLI is controlled with a new Selected Harmonic Elimination (SHE), which decreases the total harmonic distortion (THD). The system is simulated with an Resistive plus Inductive (RL) load and is tested experimentally with the same load;the results prove that the Cuk converter has a better efficiency compared to the Buck-Boost converter, and the system has less THD when compared with the conventional SHE Pulse Width Modulation (PWM) technique.
In wind energy systems, wind speed estimation plays an important role. For wind energy systems, accurate forecasting of wind direction is essential, but it is challenging because of its variability. In this paper, wind speed prediction is accomplished using a machine learning-based random forest (RF) method. For the production of wind energy, short-term wind speed prediction is a significant activity. However, it is difficult only to obtain deterministic estimation since wind supplies are erratic and unpredictable. It increases learning that helps to project future values. Average wind speed is a major feature that affects the atmosphere. This paper explains the estimation of wind speed with ML algorithm. It is valuable for assessing the prospects for abnormal climate events and wind energy in the future.