Experimentally validated finite element modeling of linear alternators for predicting electrical power and electromagnetic force due to resistive loads

Abstract

High energy density conversion systems (HEDCS) have been of growing interest within the scientific community. The applications of these HEDCS in the aerospace industry are seemingly unending as they can be small and lightweight compared to current energy conversion systems. NASA’s use of unmanned aerial vehicles (UAV) in everyday scientific study and operations creates a need for an energy efficient method of operating these aircraft. While these UAVs have already been proven useful on Earth, NASA’s discovery of materials on Mars that can perform a combustion cycle, namely O₂ and CH₄, creates an outlet for implementing HEDCS on future Mars UAVs as well. Working in tandem with Wichita State University, Kansas University, Pulse Aerospace, Aerojet Rocketdyne, and NASA Glenn Research Center with NASA funding, the objective of this project was to develop a design methodology for cost-efficient HEDCS. Previous efforts of the research team led to the decision of designing a miniature internal combustion engine (MICE) with a free piston architecture coupled to a linear alternator. This architecture was chosen because of the inherent high efficiency they demonstrate compared to common internal combustion engines resulting in a system with less friction loss and a more compact design. After the physical design of a MICE was decided, a mathematical model for predicting optimal engine performance was needed. A portion of the research team developed a model for engine kinematics and combustion dynamics. The research in this thesis focuses on attaining and validating model parameters that will be used to couple the aforementioned engine model to an electromechanical alternator model. As the magnets (mover) of a linear alternator pass through the coils (stator) an electrical current is generated in the coil along with an electromagnetic force that opposes the magnet motion. This force is similar to a damping force and is directly proportional to velocity of the mover. To create a comprehensive electromagnetic model for the linear alternator that can be integrated with the engine dynamics models, the time series values for power generation and this electromagnetic force must be found. This work details experimental methods used to validate finite element analysis (FEA) predictions of the relationships between power, resistive load, force, and mover velocity in a simple permanent magnet linear alternator. An experiment that could measure these values of interest whose motion could also be modeled in the electromagnetic simulation software, EMWorks, was developed. Two experiments were run, a drop test and a shaker table experiment. The drop test experiment consisted of a permanent magnet being dropped through a coil while the magnet motion and the coil voltage across a resistive load was collected. This experiment proved to be a useful venture as it helped with learning the basics of the simulation software and provided a power output comparison but could not provide a reliable acceleration measurement from the collected velocity data to calculate the opposing force. The shaker table experiment was used to imitate a linear alternator by oscillating the magnet in a coil while measuring the same values as the drop test in addition to measuring force directly with a statically and dynamically calibrated load cell. Through comparisons of this experimental and simulated data, the finite element model was validated for use in predicting power generation and the electromagnetic force acting against the mover of a linear alternator providing the means necessary to create a comprehensive MICE alternator model.