الفهرس 
AbstractOnly 14 pages are availabe for public view
 |  | from | 174 |  |  |
| | | from | 174 | | |
Abstract
PV panels are still generally suffering from low efficiency and high cost. To overcome these drawbacks, Maximum Power Point (MPP) techniques must be used, either by using physical sun tracking or Maximum Power Point Tracking (MPPT). Performing these two approaches simultaneously using single processing platform can produce unexpected complications in the control system due to increasing the processing load.In this study, a new design of PV panel control system using the FPGA-Based Multi-Processor System on a Chip (MPSoC) is presented. The FPGA based subsystems is used for controlling a single axis sun tracker and MPPT using three soft-core processors. The sun tracking control scheme is based on a Fuzzy Logic Control (FLC) algorithm guided by a set of time-based solar angle equations. The MPPT is employed using the Variable Step-Size Incremental Conductance (INC) algorithm. The proposed MPSoC architecture improves the sampling rate of the PV control system. Also, the user can easily add soft-core Intellectual Properties (IPs) blocks and processors to reduce the software complexity and increase the processing speed. The proposed architecture is also suitable for control algorithms, which require large memory size and floating point computations. Additionally, the proposed PV system uses a feedback sensor to provide a minimum tracking error and can be installed using low cost components. A practical implementation of a 50 W PV is implemented, constructed and instrumented to measure solar radiation, temperature of the panel surface and the power drainage from it.All experiments were carried out in clear sky days, in the Faculty of Engineering, Mansoura University, Egypt. Experimental results have shown that the performance of fuzzy-based sun tracking is better than that of the PID. Also, the sun tracking achieves a small settling time with zero overshoot. Moreover, increasing the number of processors reduces the settling time of the sun tracking and MPPT. The tracking system performances are heavily dependent on the processing speed. The combination of the sun tracking and the MPPT increases the yielded energy with an average of about 10.1%. The efficiency of the variable step-size INC with the proposed MPSoC architecture is approximately equal to that of MPPT-based on complex artificial intelligent (AI) algorithms. Finally, each processor increases the usage of FPGA resources by about 10 %.