A hybrid PSO optimized SVM-based model for predicting a successful growth cycle of the Spirulina platensis from raceway experiments data
作者：
Highlights：
• A hybrid PSO–SVM-based model is built as a predictive model of the S. platensis growth cycle.
• Chlorophyll a is a relevant parameter used to estimate the biomass production.
• The remaining physical–chemical variables in this process are studied in depth.
• The obtained regression accuracy of the hybrid PSO–SVM–RBF-based model is about 99%.
• The results show that PSO–SVM-based models can assist in the diagnosis of the S. platensis presence.
摘要
•A hybrid PSO–SVM-based model is built as a predictive model of the S. platensis growth cycle.•Chlorophyll a is a relevant parameter used to estimate the biomass production.•The remaining physical–chemical variables in this process are studied in depth.•The obtained regression accuracy of the hybrid PSO–SVM–RBF-based model is about 99%.•The results show that PSO–SVM-based models can assist in the diagnosis of the S. platensis presence.
论文关键词：Support vector machines (SVMs),Particle Swarm Optimization (PSO),Spirulina platensis,Chlorophyll a monitoring,Hyperparameter selection
论文评审过程：Received 15 October 2014, Revised 22 December 2014, Available online 16 January 2015, Version of Record 15 August 2015.
论文官网地址：https://doi.org/10.1016/j.cam.2015.01.009
