To eradicate downsides of current perturb and observe(P&O) maximum power point tracking (MPPT) algorithms as well improvements on dynamic performances, this article proposes fast-hybrid P&O (FH-PO) and intelligent self-adaptive P&O (SA-PO) algorithms for wind generation systems. Both proposed algorithms concentrate the search area for the maximum power point (MPP) to 10% of optimal P-omega curve without dividing it into modular operating sectors and prior-knowledge of perturbation step-sizes. Below 90% of optimal power, the FH-PO algorithm perturbs the rotor speed with fixed step-sizes to enhance convergence speed without redundant calculations of step-sizes at each point. At the remaining 10%, an adaptive step-size is employed to ensure low oscillations around the MPP. However, FH-PO algorithm doesn't reflect the real required step-size on each point. The SA-PO algorithm utilizes the self-adaptive step-size routine which adeptly estimates the required step-size by applying the idea of optimal hypothetical circle. Although both proposed algorithms have smallest oscillations, the SA-PO algorithm yields smallest settling time and a 4.34% increase in system efficiency. A fair comparison among proposed algorithms and other current P&O algorithms is deliberated to confirm the SA-PO algorithm superiority. The performances of proposed algorithms are tested by real wind data (Hokkaido-Island, Japan) using MATLAB/SIMULINK. (C) 2020 Elsevier Ltd. All rights reserved.