![Sample our Engineering & Technology journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5933/TOC-Subject-Sample-2021-Engineering-Tech.jpg"})
ABSTRACT
A novel control strategy based on an effective combination of the fuzzy logic controller and the M5P decision tree algorithm (Fuzzy-M5P) for a grid-interfaced solar energy system is proposed in this paper. The aim is to benefit from fuzzy logic advantages using its resulting datasets along with the fast model tree functionality. This mixed control has the ability to cope effectively with the large computation time and complexity of fuzzy logic controllers especially when embedded systems or real-time implementation are considered. Using a photovoltaic (PV) inverter with active filtering capability, the primary objective is to provide a grid-solar topology with a maximum PV energy extraction and power quality enhancement. For the inverter control, the M5P model was tested in the indirect current technique. This harmonic generation method is very efficient and its performances are entirely dependent on the chosen dc-link controller. Moreover, a novel maximum power point tracking technique (MPPT) based on the M5P model tree is performed in which data inputs are collected from fuzzy logic-based MPPT. WEKA and MATLAB software are used to translate the model by formulating a simple hierarchical program structure. To validate the effectiveness of the proposed control, simulation, and experimental tests are done under steady and dynamic conditions. The final developed algorithm based on simple IF-ELSE rules shows promising results, it can reduce harmonic distortions to 2.77%, perfectly compensates reactive power with almost unity power factor, and enhances dynamic performances with fast transient response.
List of symbols and abbreviations
| Ts | = | Sampling-time |
| PV | = | Photovoltaic system |
| M5P | = | Pruned M5 model tree |
| MPPT | = | Maximum power point tracking |
| FLC | = | Fuzzy logic control |
| ANN | = | Artificial neural network |
| APF | = | Active power filter |
| ICC | = | Indirect current control |
| THD | = | Total harmonic distortion |
| PLL | = | Phase locked-loop |
| VSI | = | Voltage source inverter |
| idc-link | = | Output dc current from the regulator |
| iS | = | Source current |
| iF | = | Filter current |
| iL | = | Load current |
| Vs | = | Source voltage |
| Vdc | = | DC-Link voltage |
| P | = | Active power |
| Q | = | Reactive power |
| D | = | Harmonic power |
| C | = | DC bus capacitor |
| NL | = | Nonlinear load |
| PI | = | Proportional-Integral |
| Linv | = | Filter inductance |
| Lg | = | Source inductance |
| Rd | = | Resistance of NL |
| Ld | = | Inductance of NL |
| LMT | = | Logistic model tree |
| DT | = | Decision tree |
| LUT | = | lookup table |
| PO | = | Perturb & observe technique |
| INC | = | Incremental conductance technique |
| MAE | = | The mean absolute error |
| RMSE | = | The root mean square error |
| R | = | The correlation coefficient |
Disclosure statement
No potential conflict of interest was reported by the author(s).