Dynamic Modelling of Automated Weight-Based Fish Sorting Machine

Abstract

This paper presents the dynamic modelling and performance analysis of an automated weight-based fish sorting machine developed for small-scale aquaculture and fish processing operations. The machine integrates a motorised conveyor belt subsystem, a load cell weighing station modelled as a second-order dynamic system, and pneumatically actuated gate mechanisms for three-category weight-based classification of fish (small: <300 g; medium: 300–600 g; large: >600 g). Mathematical models were developed for the conveyor belt drive dynamics using Newton's second law, the load cell response using a damped harmonic oscillator formulation, and the pneumatic actuator using a first-order transfer function. Simulation results demonstrate a load cell settling time of approximately 0.65 s with 8.4% overshoot at a natural frequency of 12 rad/s and a damping ratio of 0.55. The optimal belt speed of 0.30 m/s was identified, yielding a maximum throughput of 60 fish per minute with 92.3% sorting accuracy. The design and dynamic behaviour of each subsystem are presented alongside a complete bill of materials and performance characterisation tables, providing a validated framework for the realisation of the machine.