By Efe Onodjae

A team led by pioneering mechanical engineer Victor Eniola and Professor Kafayat Adeyemi, along with distinguished professors of mechanical and electrical engineering, has developed an artificial intelligence-based method for high-precision solar photovoltaic energy forecasting.

The advanced model accurately predicts energy generation for the 1.19 MW Lower Usuma Dam solar photovoltaic system. These forecasts enable the solar power plant to optimize operations and maintain reliable scheduling, ensuring a consistent energy supply.

The Lower Usuma Dam photovoltaic power plant, located in Abuja, Nigeria, was fully funded by the Japan International Cooperation Agency through the Nigerian Federal Ministry of Power. The photovoltaic modules were installed in tandem with the national grid. In addition to supplying power to the grid, the plant also powers the pumps that drive the water treatment processes at the Lower Usuma Dam.

Harnessing solar power at Lower Usuma Dam represents a significant green energy milestone in Nigeria. Engr. Eniola and his team have significantly advanced the forecasting of solar photovoltaic energy generation at the site through the development, testing, and validation of a novel method.

Their approach integrates a data preprocessing algorithm (DPA) with a multilayer perceptron neural network model to predict daily energy output from the 1.19 MW solar power plant during both the rainy and dry seasons. The DPA systematically identifies and corrects missing or irrational data points caused by human bias or equipment malfunctions, ensuring the model’s robustness. This enhancement has rendered the forecasting model highly accurate and stable, with a mean absolute percentage error consistently below 6% across all seasons, making it an effective tool for optimizing solar photovoltaic system performance.

Engr. Eniola and his team’s ingenuity and dedication to enhancing grid stability and reliability are truly commendable. This forecasting tool demonstrates outstanding aptitude and a deep understanding of the scientific community’s needs. Accurate forecasts reduce the need for expensive and carbon-intensive backup power sources, minimizing the use of costly and less efficient peaker plants.

Forecasting solar photovoltaic power is essential for ensuring the efficient, reliable, and economical operation of the power grid while supporting the broader adoption and integration of renewable energy sources. Recognizing the crucial role of solar power forecasting in enhancing efficient energy management, Engr. Eniola and his team are collaborating with power producers and policymakers to ensure a consistent power supply and develop strategies for the long-term growth of renewable energy.

Reflecting on his research, Engr. Eniola said, “My research focuses on forecasting the power output of solar photovoltaic systems, with a particular emphasis on enhancing accuracy through AI-based techniques and optimization methods. I have successfully predicted the power output of a 1.2 kW photovoltaic system using a genetic algorithm-optimized hidden Markov model. Currently, I am deploying advanced deep learning methods to improve predictions for both solar and wind power.” Engr. Eniola’s work on power forecasting has received several citations, showcasing his scholarly contributions and recognition by peers for the high quality and broad interest of his research.

Solar energy is a rapidly evolving field crucial for achieving a sustainable and carbon-neutral future. Renewable energy sources currently account for nearly 30% of global electricity generation, with significant contributions from both solar and wind power. Solar photovoltaic energy alone accounts for about 4.5% of the world’s total electricity generation. This marks a substantial increase from previous years, reflecting the rapid growth and deployment of solar photovoltaic technology. The expansion of solar energy has been driven by decreasing costs, technological advancements, improved efficiency, supportive policies, and the urgent need to reduce greenhouse gas emissions, making it one of the fastest-growing sources of renewable energy globally.

Engr. Eniola’s work in renewable energy is vital in this context, as his research and innovations in solar power forecasting are instrumental in enhancing the efficiency and reliability of solar technologies. His work is especially significant in advancing the share of renewable energy in the energy mix of nations committed to energy independence and environmental sustainability.

One of the team members, Victor Eniola earned a national diploma with distinction, a first-class bachelor’s degree in mechanical engineering from Obafemi Awolowo University, and a master’s degree in renewable energy with perfect distinction.