Development, evaluation, and optimization of an automated device for quality detection and separation of cowpea seeds

Abstract

Automation and Artificial intelligence has been used to solve the world’'s most complex problems. The goal of this study is to develop, evaluate and optimize cowpea seeds quality detection and separating device to meet international export standards. The design of the device was divided into metering, automation, and conveyor belt outlet unit. An evaluation was done using samples made up of good and bad (impurity) portions. Response surface methodology was used to evaluate, model and optimize the device performance. The optimized results were validated using regression and prediction interval (PI) analysis test. The separating efficiency, throughput, maximum capacity, and actual utilization obtained; range from 68.966 ‐ –94.118%, 0.5 – –3 kg/hr, 6–36 kg/12 h, 0.083–0.083(8.3%) respectively. These evaluating parameters were significantly affected by the operational factors at P < 0.05. Optimum values obtained are 92%, 2.689 kg/h, 32.781 kg/12 h for impurity separating: efficiency, throughput, and maximum capacity respectively. The prediction interval test shows that the validation experimental mean result lies within calculated prediction intervals. Regression analysis shows a 0.9(90%) coefficient of determination between the model predictions and the validation experimental results. The developed device was recommended to always operate at a metering speed of 20 rpm for optimum performance.

Keywords

• Cowpea
• Development
• Automation
• Optimization
• Artificial Intelligence