This paper proposes a method to reduce the steadystate oscillation and to mitigate the probability of losing the tracking direction of the perturb and observed (P&O)-based maximum power point tracking (MPPT) for PV system. The modified scheme retains the conventional P&O structure, but with a unique technique to dynamically alter the perturbation size. At the same time, a dynamic boundary condition is introduced to ensure that the algorithm will not diverge from its tracking locus. The modified P&O is simulated in MATLAB Simulink and its performance is benchmarked using the standard MPPT efficiency ηM P P T calculation. Furthermore, the proposed concept is validated experimentally using a buck-boost converter, fed by a solar PV array simulator (PVAS). Based on the EN 50530 dynamic irradiance tests, the proposed method achieved an average ηMP P T almost 1.1% higher than the conventional P&O when irradiance changes slowly and about 12% higher under fast change of irradiance.

This paper proposes an enhanced adaptive perturb and observe (EA-P&O) maximum power point tracking (MPPT) algorithm for the photovoltaic system. The objective is to mitigate the limitations of the conventional P&O namely, the steady-state oscillation, diverged tracking direction, and inability to detect the global peak during partial shading. A smart oscillation detection scheme and a dynamic boundary condition resolve the first two problems, respectively. Meanwhile, an intelligent prediction method is designed to ensure that the global peak is always correctly tracked. Another feature is the open-circuit voltage is determined without using sensors. The proposed idea is verified using MATLAB simulations by imposing stringent dynamic irradiance and partial shading tests. Moreover, an experimental validation is carried out using a buck-boost converter in conjunction with dSpace DS1104 DSP board. The performance of the algorithm is compared with four prominent MPPT techniques: first, the artificial bee colony; second, modified incremental conduction; third, cuckoo search; and fourth, the hybrid ant colony optimization-P&O. The results show that the proposed method tracks the global peak successfully under distinctive patterns of partial shading, when other algorithms fail occasionally. On top of that, it improves the tracking speed by two to three times, while efficiency is maintained over 99%.

This paper proposes an improved maximum power point tracker (MPPT) for photovoltaic (PV) system. The scheme is a hybrid between the adaptive perturb and observe (P&O) and particle swarm optimization (PSO). The algorithm incorporates the search-skip-judge (SSJ) mechanism to minimize the region within the P-V curve to be searched by the PSO. Furthermore, the PSO performance is enhanced by ensuring that the regions that have been previously explored (by other particle) will not be searched again by (another particle). Thus, the unnecessary movement of particles is minimized—leading to faster convergence. The proposed method is evaluated against four well-known MPPT techniques, namely the modified incremental conductance, the original version of SSJ, the modified cuckoo search and the hybrid PSO. In addition, an experimental prototype, which is based on PV array simulator is used to verify the simulation. The competing algorithms are tested with a buck-boost converter, driven by the TMS320F240 DSP on the dSPACE DS1104 platform. It was found that the proposed scheme converges to the global maximum power point (GMPP) most rapidly and the GMPP tracking is guaranteed even under complex partial shading conditions.

This paper proposes a fast and efficient maximum power point tracker (MPPT) for photovoltaic (PV) system under rapidly changing partial shading condition. An intelligent mechanism is adopted to systematically schedule the search for the global maximum power point (GMPP) on the P-V curve. As a result, the voltage track, i.e. the path-length that the operating point traverses along the voltage axis of the curve (until it converges to GMPP), is reduced. The search region is further minimized using a novel skipping scheme, where the voltage section that does not contain GMPP is discarded. The superiority of the proposed scheme is evaluated against two recent algorithms, namely the maximum power trapezium and the flower pollination MPPT. The performance is analyzed in terms of convergence time, voltage track and transient efficiency. The MATLAB simulation is verified experimentally using a PV array simulator, in conjunction with a buck-boost converter. The competing MPPT algorithms are implemented using the TMS320F240 DSP on the dSPACE DS1104 platform. The results indicate that, under the same operating and shading conditions, the proposed scheme is the fastest, most reliable and exhibits the highest overall transient efficiency.