Electrodynamics Associates (United States)

ABSTRACT This article addresses various technical challenges in developing high speed, lightweight electric generators for applications in direct coupled high speed generators for distributed power systems. These challenges involve the technologies of electromagnetics, thermodynamics, rotor dynamics, and engineering assembly practices. How those challenges are being addressed and the results of activities to date relating to generators from several hundred kilowatts to a few megawatts will be discussed. Current status and the test data relating to these generators will be presented. While the requirements of power and voltage levels as well as the duty cycle vary, the applicable generator technology is mainly impacted by the power level. The electric generator is typically driven by a high speed turbine. The operating speeds of the turbine and the mechanical and thermal interfaces with the generator have a substantial impact on the design details of the generator. The best approach in interfacing the high speed turbine to the generator is to directly couple the two by means of a mechanical coupling and exclude the need for gear, belt or other mechanical transmission systems. The high speed turbine operates most efficiently at their rated speed, which varies with rated power output. Thus while a typical 200 kW turbine may operate at 60000 rpm, a MW class turbine may operate in the range of 15000 rpm to 30000 rpm. It is often details such as the coupling and mounting arrangement and the suitability of coolants and lubricants that affect the generator design and impact the weight and size of the final product. Selection of the appropriate generator technology is highly dependent on the operating speed requirements. Examples of specific power and speed ratings as well as suitability of generator technologies such as permanent magnet (PM), wound rotor synchronous, switched reluctance (SR) and induction will be discussed. Current development programs for 30 kW to 2.5 MW power generators will be presented.

An innovative nonlinear simulation approach for high power density synchronous generator systems is developed and implemented. Due to high power density, the generator operates in nonlinear region of the magnetic circuit. Magnetic Finite Element Analysis (FEA) makes nonlinear simulation possible. Neural network technique provides nonlinear functions for system level simulation. Dynamic voltage equation provides excellent mathematical model for system level simulations. Voltage, current, and flux linkage quantities are applied in Direct-Quadrature (DQ) rotating frame. The simulated system includes main machine, exciter, rectifier bridge, bang-bang control, and PI control circuitry, forming a closed loop system. Each part is modeled and then integrated into the system model.

A theory of optical, infrared, and microwave response of metal- dielectric inhomogeneous films is developed. The generalized Ohm's law is formulated. In this approach electric and magnetic fields outside a film can be related to the currents inside the film. Our computer simulations, show that the local electric and magnetic fields experience giant spatial fluctuations. The fields are localized in small spatially separated peaks: electric and magnetic hot spots. In these hot spots the local fields (both electric and magnetic) exceed the applied field by several orders of magnitude. The high-order field moments that characterize the average enhancement of Raman scattering and nonlinear optical processes are very large and frequency independent in a wide spectral range.

Preliminary design of a meso-scale refrigerator is the focus of this paper. Two design variations for two different applications are presented here. The two intended applications are (1) an integrated heat removal system for electronics or photonic chips or modules, and (2) an actively cooled jacket for personnel. The proposed device is based on a vapor compression refrigeration cycle. The paper starts with a general system description and preliminary design along with a functional decomposition where the overall system function is decomposed into nine major sub-functions. Different alternatives for compression and actuation are considered. Design specifications and some design details for each application are presented next. The specifics of the design requirements were found to suggest the same thermodynamic cycle for both applications, with the same temperature lift of 48

Attempts to use variational grid-generation methods to generate grids on certain surfaces of modest shape failed. There were sufficient points in the grids to well-resolve the surface, so the failures were not easily explained. Similar difficulties were found for variational grid generation on curves; those problems are caused by multiple solutions of the underlying nonlinear algebraic equations.

Additional performance aspects of marching methods for elliptic equations, as embodied in the GEM codes, are presented. These aspects include Neumann boundary conditions in nonorthogonal grids, application to minicomputers and supermicro-computers, vectorization, and comparison to multigrid methods.

This paper is divided into three sections. The first section is a brief present presentation of the powerful capabilities of the symbolic manipulation code MACSYMA developed at MIT. The second section presents results recently achieved by the authors using symbolic manipulation on the problem of three-dimensional bodyfitted coordinates. The third part of the paper surveys previous uses of symbolic manipulation in computational fluid dynamics and related areas, and potential future uses of the code.

Numerical procedures are presented which permit one to generate orthonormal sets of sequences, expand an arbitrary sequence in terms of the set, and reconstruct the arbitrary sequence using only reeursive numerical filtering techniques These sequences approach the uniform samples of an important class of continuous orthonormal functions which include the Laguerre functions, Kautz functions, and others in ~he ge~eral class of orthonormalized exponential functions.