A Review on Machine Learning Strategies for Real-World Engineering Applications

Table 4

ML state-of-the-art systems in renewable energy domain.

Reference	Machine learning technique	Purpose
Wind power generation
[147]	Statistical machine learning techniques	Short and medium forecasting
[148]	Autoregressive integrated moving average and autoregressive moving average	Wind power forecasting and forecasting of wind speed
[149]	Kalman filter model is used	Wind-generated power and wind speed forecasting through online
[150]	Review on two machine learning techniques is done	Wind speed forecasting
[151]	ANN	TIME SERIES PREDICTION
[152]	ANN variant is used called recurrent multi-layer perception	Used for the prediction of long-term power generation
[153]	SVM is used	To measure the wind speed
[154]	Fuzzy models are used	For the prediction of wind power generation.
[155]	Numerical weather prediction model is used	Wind power consumption and generation forecast
[156]	Ensemble model is used	Wind power consumption and generation forecast
[157]	ANN and k-nearest neighbor approaches are used	Wind power generation forecast
[158]	Particle swarm optimization, k-NN and SVM are used	Wind power generation forecast
[159]	Techniques considered are; (i) Random forest; (ii) Regression trees; (iii) ANN; (iv) MLP; (v) SVM	Review on machine learning techniques for wind power generation forecasting
Solar energy generation
[160]	ANN is used	Solar energy generation
[161]	SVM is used	Power generation using solar
[162]	Ensemble method	Forecasting solar power generation
[163]	Statistical methods	Review on solar energy power generation
Hydro power generation
[164–166]	RNN, SVM	Rainfall prediction
[167]	RNN, SVM	Forecast values of rainfall depth
[168]	Ensemble learning	Forecast the hydro energy consumption
[169–171]	ANN	Hydropower plant management