Short-Term Wind Speed Forecasting Using Support Vector Regression Optimized by Cuckoo Optimization Algorithm

Abstract

This paper develops an effectively intelligent model to forecast short-term wind speed series. A hybrid forecasting technique is proposed based on recurrence plot (RP) and optimized support vector regression (SVR). Wind caused by the interaction of meteorological systems makes itself extremely unsteady and difficult to forecast. To understand the wind system, the wind speed series is analyzed using RP. Then, the SVR model is employed to forecast wind speed, in which the input variables are selected by RP, and two crucial parameters, including the penalties factor and gamma of the kernel function RBF, are optimized by various optimization algorithms. Those optimized algorithms are genetic algorithm (GA), particle swarm optimization algorithm (PSO), and cuckoo optimization algorithm (COA). Finally, the optimized SVR models, including COA-SVR, PSO-SVR, and GA-SVR, are evaluated based on some criteria and a hypothesis test. The experimental results show that (1) analysis of RP reveals that wind speed has short-term predictability on a short-term time scale, (2) the performance of the COA-SVR model is superior to that of the PSO-SVR and GA-SVR methods, especially for the jumping samplings, and (3) the COA-SVR method is statistically robust in multi-step-ahead prediction and can be applied to practical wind farm applications.

Current gust forecasting techniques, developments and challenges

“…Mohandes et al, 2004;Sreelakshmi and Kumar, 2008;Zeng and Qiao, 2011;Giorgi et al, 2014;Wang et al, 2015;Dunstan et al, 2016;Schicker et al, 2017). The agility of machine learning techniques in dealing with non-linear behaviour of predictand variables as a function of predictors makes them well suited to modelling extreme winds, which are frequently associated with highly non-linear atmospheric phenomena (e.g.…”

Real-time prognosis of flowing bottom-hole pressure in a vertical well for a multiphase flow using computational intelligence techniques

“…PSO is a stochastic population-based evolutionary algorithm, motivated by the societal attitude of fish schooling and birds clustering (Kennedy 1997;Shi and Eberhart 1998;Abido 2002). ANN coupled with PSO has proved to be better and faster predictive tool in comparison against the conventional ANN technique (Catalao et al 2010;Vasumathi and Moorthi 2012;Wang et al 2015;Tariq et al 2016;Jahed Armaghani et al 2017;Chatterjee et al 2017;Tariq et al 2018a;Ethaib et al 2018). PSO represents population of random solutions in the search space as particles assigning random velocities to them and iteratively tuning the fitness of the particles until the best solution called global best is achieved.…”

Design of Arduino-based small wind power generation system

“…In this study, it was used a wind turbine blade with a radius of 0.5 m and assumed the value of ρ at 1.37kg / m3, Cp of 0.3, and T 0.1 Nm. By entering the known variable value, accordingly the value of wind, v wind , and ω were calculated by combining equation 1with equation 3into the equation (4). Figure 8 shows the graph of the output voltage of the generator.…”

“…These alternative energy uses renewable recourses not only to reduce the dependence on fossil fuels but also the carbon emission and improve the air quality. Wind energy naturally has become the resource of the fastest growing renewable energy in the worldwide [3,4].…”