Chlorella Vulgaris Surface-Mount Photobioreactor with Vision-Based Growth Signature Prediction Optimized by Electromagnetism-Like Mechanism
Ronnie Concepcion II1, Michael Jon Alain Saavedra2, Jonnel Alejandrino3, Maria Gemel Palconit4
1Ronnie Concepcion II*, Electronics and Communications Engineering Department, De La Salle University, Manila, Philippines.
2Michael Jon Alain Saavedra, Electronics Engineering Department, University of Perpetual Help System DALTA, Las Piñas City, Philippines.
3Jonnel Alejandrino, Electronics and Communications Engineering Department, De La Salle University, Manila, Philippines.
4Maria Gemel Palconit, Electronics and Communications Engineering Department, De La Salle University, Manila, Philippines.
Manuscript received on October 06, 2020. | Revised Manuscript received on October 25, 2020. | Manuscript published on November 30, 2020. | PP: 378-387 | Volume-9 Issue-4, November 2020. | Retrieval Number: 100.1/ijrte.D5007119420 | DOI: 10.35940/ijrte.D5007.119420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Industrial waste disrupts the natural production of microalgae cultures. Cultivation of microalgae in a controlled environment highly results to biomass with lower contamination necessary as high-valued economic product. In response to the emerging challenges of sustainable energy production, the integration of computational intelligence and biosystems engineering is considered as an open research area. In this study, Chlorella vulgaris microalgae were cultivated in BG-11 growth medium on three customized surface-mount light bioreactors that are equipped with digital camera for growth monitoring in terms of accumulated biomass surface area and color reflectance intensity via IoT. Feature-based machine learning models predicted microalgae growth area in terms of water temperature, pH level and turbidity, and light intensity. Microalgae cultures were exposed to combinations of white artificial light source of 2000 ± 1000 lux and water temperature of 27 ± 5°C using Peltier plate to discriminate biomass growth within a 30-day cultivation period. A total of nine environmental conditions were employed to clearly discriminate the impacts of environmental stressors to microalgae growth. Combined neighborhood component analysis and ReliefF was used to select high impact color features of C, Ye, M, H, and S with biomass area. Electromagnetism-like mechanism optimized-RBNN bested RNN and generalized processing regression with R2 of 0.985 and RMSE of 6.262. There is also considerable growth in biomass surface area for certain combinations of light intensity and water temperature (2125 ± 625 lux and 28.75 ± 3.25°C), and turbidity and water pH concentrations (3.85 ± 0.15 NTU and 8.025 ± 0.775). However, the photobioreactor with 27°C and 2000 lux exposure is considered having the exact optimum controlled environment condition in cultivating Chlorella vulgaris based on the generated growth in biomass surface area of 38.314%. This developed intelligent system is scalable for seamless microalgae production of any strands for renewable energy resource.
Keywords: Bioreactor, computational intelligence, computer vision, microalgae.