(SEP 2021) Prof. Wai Lun LO, has research collaboration with Prof. Henry Shu Hung Chung (Chair Professor, Dean of Students, EE Dept. City University of Hong Kong) and Dr. Richard Tai Chiu HSUNG (Associate Professor, CS dept.). Prof. LO, Prof. Henry Chung and Dr. Richard HSUNG have successfully got research funding for the following project. It is expected that the project will start in Jan 2022.
Photovoltaic Panel Model Parameters Estimation and Monitoring by Using Artificial Neural Network
Prof. Wai Lun LO (PI), Head of CS dept., Associate Dean FSE, CHCHE
Prof. Henry Shu Hung Chung (Co-I), Chair Professor, Electrical Engineering Dept., Dean of Students, HKCityU

Dr. Richard Tai Chiu HSUNG (Co-I), Associate Professor, CS dept., CHCHE
Research Grant Committee, Hong Kong, Faculty Development Scheme (FDS)

36 months, HK$ 914,925, UGC/FDS13/E01/21

Project Summary
In this project, we will propose a Photovoltaic Panel (PV) model parameter estimation system by using the Artificial Neural Network (ANN) approach and the output voltage and current (VI) are monitored by Digital Data Acquisition Module (DDAM) from which the circuit’s model parameters are estimated. The VI approach for PV panel modelling has the advantages of convenience, real time and low cost. In this project, a Digital Data Acquisition Module and an Artificial Neural Network for PV panel modelling will be developed. The capture VI data will be used to estimate the PV panel parametric model parameters by using ANN. The non-linear mapping of VI characteristics to model parameters is curve fitted by the ANN. Furthermore, the structure and parameters of ANN will be optimized by Metaheuristic Computation algorithms in High Performance Computing (HPC) Clusters. The performance of the proposed algorithms will be evaluated by experimental studies on a PV panel setup in which PV panel is illuminated by light source with controlled intensity. Output VI characteristics under load fluctuation will be captured by the DAMM and the data series will be sent to the host Computer by using communication device. The ANN is trained for a non-linear mapping that can correlate the VI time series with the PV circuit model parameters. The ANN training and the structure optimization will be carried out in host computer or HPC, ANN parameters will be sent back to the local DAMM for PV panel model parameters monitoring purposes. The outcomes of this research project can be applied to PV panel model parameters estimation system and PV panel health monitoring system.