Professor Andrew Amenaghawon Advances Biodiesel Innovation Through AI-Driven Feedstock Optimization

Professor Andrew Amenaghawon, a leading researcher in chemical engineering at the University of Benin, is advancing the frontiers of sustainable energy through innovative work that integrates artificial intelligence with biodiesel production. 

In a recent study where he served as the lead investigator, Amenaghawon and his team combined experimental methods with machine learning (ML) models to optimize blended oil feedstocks, offering a smarter and more efficient pathway toward high-quality biodiesel production.

The research goes beyond traditional experimental approaches by leveraging AI tools to uncover complex relationships between feedstock composition, process parameters, and biodiesel yield. By integrating data-driven modeling with laboratory experiments, the study demonstrates how intelligent systems can accelerate optimization and improve predictive accuracy in biodiesel production.

“We did not just rely on experimental trial-and-error,” Amenaghawon explains. “We integrated machine learning into the process to help us understand patterns that are not immediately obvious and to guide us toward optimal conditions more efficiently.”

Improving Biodiesel Through Strategic Blending and AI Insight

Biodiesel production has long faced challenges related to feedstock variability, cost, and fuel quality. Conventional approaches often rely on a single oil source, limiting flexibility and efficiency. In this study, Amenaghawon’s team explored blended oil systems while simultaneously applying machine learning algorithms to model and predict process outcomes.

The ML models were trained on experimental datasets incorporating physicochemical properties of the feedstocks and operating conditions. These models enabled the team to predict biodiesel yield and key fuel properties with high accuracy, significantly reducing the need for extensive experimental iterations.

“Artificial intelligence allowed us to move faster and make better decisions,” he says. “Instead of testing every possible combination experimentally, the models helped us identify the most promising conditions to focus on.”

The study demonstrates that combining oils with complementary properties, when guided by ML-based optimization, can lead to improved reaction efficiency and enhanced fuel characteristics such as viscosity, density, and overall yield.

Using AI to Accelerate Discovery and Optimization

A key contribution of the work lies in its use of AI as a decision-support tool in process engineering. Machine learning models were used not only for prediction but also for identifying the relative importance of different process variables, offering deeper insight into the biodiesel production system.

“AI gives us the ability to see the system differently,” Amenaghawon notes. “It shifts our focus from just conducting experiments to understanding the underlying relationships within the process.”

By revealing nonlinear interactions between variables, the ML framework provided a more comprehensive understanding of how feedstock blends and operating conditions influence performance. This approach significantly enhances the efficiency of process design and optimization.

Toward More Sustainable and Scalable Energy Systems

Beyond performance improvements, the integration of AI into biodiesel production contributes to sustainability by reducing resource use and experimental waste. The ability to predict optimal conditions minimizes unnecessary trials, saving time, energy, and materials. 

Additionally, the blended feedstock approach enables the use of diverse and potentially lower-cost oil sources, improving resilience and adaptability in biodiesel production systems.

“Sustainability is not just about the feedstock,” Amenaghawon explains. “It is also about how efficiently we use resources. AI helps us achieve both.”

This dual strategy of feedstock blending and AI-driven optimization provides a scalable framework that can be applied across different production environments, particularly in regions with variable resource availability.

Bridging Research and Real-World Application

Amenaghawon’s work reflects a strong commitment to translational research, where advanced computational tools are integrated with practical engineering solutions. By combining experimental validation with AI-driven insights, the study offers a robust framework that can be adopted in industrial biodiesel production.

“The strength of this work lies in the integration,” he emphasizes. “We are bringing together data, models, and experiments to create solutions that are both innovative and implementable.”

This approach aligns with the growing trend of digital transformation in chemical engineering, where AI is increasingly used to enhance process efficiency and decision-making.

Expanding the Frontiers of AI in Renewable Energy Research

Beyond biodiesel, Amenaghawon’s broader research portfolio continues to explore the application of artificial intelligence in sustainable energy systems, process optimization, and bioresources valorization. His work highlights the transformative role of AI in accelerating innovation and addressing global energy challenges. 

Through mentorship and collaboration, he is also equipping students and young researchers with the skills needed to operate at the intersection of engineering and data science.

“The future of energy research will be driven by the integration of domain knowledge and intelligent systems,” he says. “We must prepare the next generation to think in both dimensions.”

Amenaghawon’s latest study stands as a compelling example of how artificial intelligence can be harnessed to improve not just the efficiency of biodiesel production, but the entire approach to sustainable energy development.