Development of computational intelligence models for a smart microalgae-based biorefinery production

Microalgae biorefinery has the potential to produce a clean and sustainable alternative biofuel that may replace fossil fuels. The increasing progress of industry 4.0 will aid the realization of clean and affordable energy which is one of the sustainable development goals (SDGs). One of the technologies that will provide dynamic communication with data and systems is the internet of things (IoT) that can monitor, automate, and predict the output of systems in a biorefinery that leads to high product yield. However, IoT systems will have a hard time computing readily available predetermined mathematical models due to a lack of computational resources. In addition, with the rise of big data, increasing sensitivity of sensors and instruments, mathematical models lack dynamicity in handling big and new datasets. This study enforces computational intelligence models (CI) to predict and optimize microalgae biorefinery systems. This study analyses the microalgae biorefinery process chain such as the cultivation process, drying process, pyrolysis process, and the environmental implications of microalgae biorefinery production. The cultivation process was modeled using Artificial neural network – Artificial Bee Colony algorithm (ANN-ABC) to predict the biomass concentration. The vacuum drying process was modeled using different machine learning techniques. Thermo-gravimetric analysis (TGA) data set was modeled using ANN. The environmental impact of the microalgae biorefinery production was modeled using life cycle assessment (LCA) coupled with Genetic programming (GP) and Genetic algorithm (GA) for prediction and optimization of environmental factors. The significance of the study highlights the potential of CI techniques as an auxiliary to IoT applications used for microalgae biorefinery production for prediction and optimization. Thus, achieving a smart sustainable microalgae biorefinery