Electrical Engineering

We report on a passively mode-locked InP/InGaAsP multiple quantum well semiconductor ring laser that operates at a 20 GHz repetition rate and around 1575 nm wavelength. The device has been realized using the active–passive integration technology in a standardized photonic integration platform. We demonstrate experimentally for the first time to our knowledge that the relative positioning of the amplifier and absorber in a monolithically integrated ring laser can be used to control the balance of power between counterpropagating fields in the mode-locked state. The directional power balance is verified to be in agreement with a model previously reported.

Flux-switching permanent magnet (FSPM) machines are double salient machines with a high energy density suitable for e-mobility. For a fast design process, machine specialists need easy-to-use motor models. For the FSPM model, analytical methods cost high efforts to create and to improve them. Numerical methods such as the finite element method (FEM) have been extensively studied in the literature with little emphasis given to their alternatives.

With the emergence of energy related issues in the automotive sector, there is a tendency to find new efficient solutions to replace existing electrical machinery. One promising candidate is the flux switching permanent magnet machine (FSPMM). Due to its challenging structure and nonlinear characteristic, in the investigation of the machine, generally finite element method (FEM), and rarely the magnetic equivalent circuit (MEC), are implemented. The following paper introduces an alternative analytical modeling technique by means of a hybrid model, which combines the advantages of the MEC and the Fourier analysis.

Weight limitations in electric/hybrid cars demand the highest possible power-to-weight ratio from the traction motor, as in double salient permanent magnet (PM) machines. Their high flux densities in the air gap result in nonlinear analytical models, which need to be time optimized. The incorporated reluctance networks are sensitive to the correctness of air gap permeances. Conventionally, in these networks, air gap permeances are calculated by approximating flux paths; however, it is time inefficient. For an improved simulation time, spatial discretization techniques are presented to calculate air gap permeances in double salient PM machines. The spatial techniques discussed here cover the tooth contour method and Schwarz-Christoffel (SC) mapping as semidiscrete methods, which are used to discretize only the air gap region. Their results are verified by the spatial discrete method and finite element method, which discretizes the whole machine geometry. For consistency in this paper, all methods are explained on a three-phase 12/10 flux-switching PM motor. Obtained air gap permeances show a very good agreement with only 0.8% difference. Further on, machine characteristics such as phase flux linkage and cogging torque are also compared to show the impact of the modeling techniques. The total machine simulation time is improved by 20% using the SC method. Although methods are explained particularly for double salient PM machines, formulas are generalizable for other machine types as well.

With the emergence of energy related issues in the automotive sector, there is a tendency to find new efficient solutions to replace existing electrical machinery. One promising candidate is the flux switching permanent magnet machine (FSPMM). Due to its challenging structure and nonlinear characteristic, in the investigation of the machine, generally finite element method (FEM), and rarely the magnetic equivalent circuit (MEC), are implemented. The following paper introduces an alternative analytical modeling technique by means of a hybrid model, which combines the advantages of the MEC and the Fourier analysis.

For applications demanding a high torque density and high speed capability, the flux switching permanent magnet machine is an excellent candidate. However, the double salient structure and nonlinear behavior increases the challenge to model the magnetic field distribution and torque output. To date, only the magnetic equivalent circuit (MEC) is employed to model the magnetic field in an analytical manner. However, the MEC method suffers from a coarse discretization and the need for a relative complex adjustment when rotor movement or a parametric sweep is considered. Therefore this paper discusses an alternative technique based on the harmonic or Fourier model which solves these difficulties.

The conservation of energy and the development of energy-efficient products at minimum costs is a critical challenge facing the world today. A recently introduced class of machine, the flux switch ing permanent magnet machine (FSPM) [1], could play a pivotal role in answering some of these issues for the automotive and energy sector. To date, in these applications permanent magnet syn chronous machines (PMSM), due to their high power density, or switched reluctance machines (SRM), due to their simple rotor structure capable of reaching higher speeds, are evermore applied. In this respect, the FSPM embodies the combined advantages of both machine types. The aim of this paper is to illustrate a new way to investigate the FSPM. To date, these machines are primarily designed using finite elements (FE) and only a very limited analytical approaches are available [2]. Therefore, this paper proposes a novel analytical hybrid modeling technique, which combines the advantages of both the magnetic equivalent circuit (MEC) and Fourier analysis. The obtained results are verified with 2D FE. Although the FSPM allows for unconventional stator rotor pole combinations, for a convenient comparison to published articles, a 12 10 pole structure ( ) has been considered to verify the analytical hybrid model. 2) HYBRID MODEL The coupling of energy sources, magnetic and electrical, that use an identical magnetic circuit results in a highly saturated FSPM, hence requires an iterative non-linear magnetic analysis method. One way to overcome this is to model the FSPM using a MEC at the cost of a coarse dis cretization compared to the FE analysis [2]. An alternative or supplementary technique, the Fouri er analysis, obtains a fine discretization, however it can only deal with linear magnetic modeling [3]. The proposed analytical hybrid model combines the advantages of both methods in order to predict the output torque. In this new analytical hybrid model the presence of non-linear soft-magnetic material is included by virtually changing a global machine parameter, i.e. the airgap length. This parameter variation continues until an error-criteria is satisfied, which compares the reference nonlinear MEC flux den sity values in the airgap center with the linear Fourier model. Once an acceptable flux density dis tribution is achieved, the result of the Fourier analysis with adapted airgap length is used in the Maxwell Stress Tensor method to obtain the output torque. Various error criteria have been researched, where the first two, i.e. A and B, consider a point-wise comparison of the flux density distribution. These initial criteria make use of a mean error function, which compares the solutions in the airgap center from the two analytical methods, the nonlinear MEC and linear Fourier analysis. As such, criterion A differs from B in that it only compares the points where a flux density above 2T is reached, because these locations show the largest deviation from the real machine behavior in the presence of saturation. Criterion B considers all points in the airgap using the same error equation as A. The last criterion C differs from the previous two in the sense that instead of a point-wise, a mean comparison is considered, where each step in the model makes a numerical integration over the airgap area for the two methods, which are compared later in the decision step. 3) RESULTS AND CONCLUSIONS Depending on the choice of error criterion, a different level of accuracy has been achieved in flux density and mean torque calculation, ). Criterion A has a very short simulation time (e.g. 307 s) with an error in the rated torque output of 330o. The second criterion B has the highest accu racy with the lowest error of 6°o but a larger simulation time (e.g. 487 s). The last criterion C gives a balanced result regarding both aspects with an error of 1 8% and a simulation time between the other criteria. The hybrid analytical model combines the advantages of the nonlinear analysis capability of the MEC model with the fine discretization of the linear Fourier analysis in order to predict the rated torque output of a FSPM machine with a highly saturated nature. The various error criteria give the designer the freedom to select between calculation time (e.g. for optimization) and model accura cy.

This paper evaluates what are the conditions, in terms of increased overhead time and number of optical network units (i.e., ONUs), in which cyclic sleep based techniques are effective in Long Reach PONs. OCIS codes: (060.4250) Networks; (060.4252) Networks, broadcast

This paper evaluates what are the conditions, in terms of increased overhead time and number of optical network units (i.e., ONUs), in which cyclic sleep based techniques are effective in Long Reach PONs. OCIS codes: (060.4250) Networks; (060.4252) Networks, broadcast