This paper presents a sensorless method for indirectly estimating the moisture content of shredded organic food waste during mechanical pressing, based on real-time analysis of the press motor electrical current. The proposed approach exploits the established relationship between motor torque and stator current: as moisture is extracted from the material, mechanical resistance increases, producing a measurable change in motor current. Three motor topologies are evaluated through detailed simulation: a three-phase squirrel-cage induction motor (IM), a permanent magnet synchronous motor (PMSM) with field-oriented control (FOC), and a hybrid stepper motor with microstepping drive. For each motor, mathematical models are developed and integrated with a physics-based mechanical load model representing the press dynamics. Two estimation strategies are compared: direct polynomial inversion of the empirical I(W) relationship, and an Extended Kalman Filter (EKF) combining a process model of moisture extraction with the motor current measurement. Simulation results demonstrate that both the IM and PMSM with EKF estimation achieve a moisture estimation root-mean-square error of approximately 3.18%, compared to 23 to 28 percent with polynomial inversion alone. The stepper motor is shown to be unsuitable for current-based moisture sensing due to its constant-current driver characteristic. Energy consumption analysis indicates that all three motor types are viable for compact household applications. The method has been incorporated as a core technical feature of a patented household organic waste processing device (USPTO Provisional Patent Application No. 63/995,028) and represents a novel application of motor current signature analysis to the food waste processing domain. Full simulation code is provided for free reproduction of the results.
Yurii O. Khomuilo (Mon,) studied this question.