article
Biochar provides an important renewable energy source produced through biomass thermochemical processes, but final yields vary according to specific operating conditions. Conventional computational techniques, such as the least square support vector machine, often face challenges including high time complexity and a tendency to become trapped in local optima. To overcome these limitations, an adaptive neuro-fuzzy inference system was developed and trained using a particle swarm optimisation algorithm to forecast biochar production. The predictive model incorporates five key input parameters: heating rate, pyrolysis temperature, moisture content, holding time, and sample mass. It forecasts both biochar mass and overall biochar yield. Evaluation against standard benchmarks demonstrates superior performance, achieving a root mean square error of 0.2673, a coefficient of determination of 0.9842, and an average absolute percent relative error of 3.4529.
Biochar is a valuable form of renewable energy, yet predicting how much yield a thermochemical process will generate remains difficult. By applying particle swarm optimisation to a neuro-fuzzy system, producers and researchers can model output accurately across varying operational variables like temperature and moisture, helping refine biomass conversion processes without excessive trial and error.
This computational model could assist biomass processing operators and bioenergy producers seeking to optimise thermochemical conversion settings before running physical operations. By relying on common operational inputs, the method could be embedded into industrial process monitoring or design software. However, the work represents early-stage predictive modelling evaluated on computational error metrics, with no real-world pilot or deployment pathway detailed in the text.
AI-generated from the published abstract. Always read the original work before citing.
This paper proposed an intelligent approach to predict the biochar yield. The biochar is an important renewable energy that produced from biomass thermochemical processes with yields that depend on different operating conditions. There are some approaches that are used to predict the production of biochar such as least square support vector machine. However, this approach suffers from some drawbacks such as get stuck in local point and high time complexity. In order to avoid these drawbacks, the adaptive neuro-fuzzy inference system approach is used and this approach is trained with a particle swarm optimization algorithm to improve the prediction performance of the biochar. Heating rate, pyrolysis temperature, Moisture content, holding time and sample mass were used as the input parameters and the outputs are biochar mass and biochar yield. The results show that the proposed approach is better than other approaches based on three measures the root mean square error, the coefficient of determination and average absolute percent relative error (0.2673, 0.9842 and 3.4529 respectively).
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1109/powerafrica.2017.7991209
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
