Search

This paper describes a new method for indirect sensing of the rotor position in switched reluctance motors (SRM's) using pulsewidth modulation voltage control. The detection method uses the change of the derivative of the phase current to detect the position where a rotor pole and stator pole start to overlap, giving one position update per energy conversion. As no a priori knowledge of motor parameters is required (except for the numbers of stator and rotor poles), the method is applicable to most SRM topologies in a wide power and speed range and for several inverter topologies. The method allows modest closed-loop dynamic performance. To start up the motor, a feedforward stepping method is used which assures robust startup (even under load) from standstill to a predefined speed at which closed-loop sensorless operation can be applied. Experimental results demonstrate the robust functionality of the method with just one current sensor in the inverter, even with excitation overlap, and the sensorless operation improves with speed. The method is comparable to the back-EMF position estimation for brushless dc motors in principle, performance, and cost. A detailed operation and implementation of this scheme is shown, together with steady-state and dynamic transient test results. Index Terms - Adjustable-speed drives, sensorless control, switched reluctance motors.

Induction motor (IM) design represents a complex multivariable optimization mathematical problem with constraints. Even recent advanced mathematical tools do not guarantee a global optimum. Based on a nonlinear analytical field-circuit model, this paper introduces a novel practical approach structured in two stages, the optimal lamination and the longitudinal optimization. The first stage, divided into three substages, produces a unique optimal lamination cross section for a given external stator diameter and number of poles, based on the criterion of maximum airgap flux density for given airgap and magnetization MMF. With only the external diameter and current density as independent variables, in the second stage, the chosen optimization objective function is minimized (maximized) in a straightforward iterative manner. Numerous practical examples have produced a global optimum. Index Terms - Induction motors, optimal design.

This paper formulates the objectives associated with high-grade torque control of switched reluctance motors (SRM's). Low torque ripple is a natural outcome of high bandwidth torque control. Smooth torque operation is only uniquely defined together with a secondary objective, such as maximum efficiency, maximum torque-speed capability, maximum power factor, etc., as the SRM does not enjoy the inherent distribution between phases that characterizes ac drives. By introduction of the fluxangle parameter, operation at any point between maximum torque-current ratio and maximum torque-flux ratio is possible. A systematic approach to waveform design is developed and is evaluated using both simulation and experimental testing of an SRM servo-drive system. This approach, which can be regarded as a functional equivalent of ac vector control, starts from machine parameters and gives smooth torque operation with maximized torque-speed range and/or efficiency. Excellent dynamic performance is obtained for speed and position control, together with optimized low torque ripple operation. Index Terms - Reluctance motor drives, torque control.

A new brushless doubly-salient permanent-magnet machine called the flux-reversal machine is described. Design, analysis, construction, and experimental results for a prototype machine built as a single-phase, high-speed generator are presented. Qualitative comparisons are made with the other types of brushless machines in its class, using the flux-MMF diagram technique. An elementary CAD program developed for designing flux-reversal machines and the associated drive electronics is also described. Index Terms - Doubly-salient permanent-magnet machines, permanent-magnet generators.

Acoustic noise in the switched reluctance motor (SRM) is caused primarily by the deformation of the stator lamination stack. Acoustic noise is most severe when the periodic excitation of the SRM phases excites a natural vibration mode of the stack. The natural vibration modes and frequencies of a four-phase, 8/6 SRM are examined. Structural finite element analysis is used to compute the natural modes and frequencies. Impulse tests on the stator stack verify the calculations and show which modes are excited. Heuristic arguments are developed to predict the operating conditions that will excite the natural modes. Measurement of vibration while the machine is under load shows which operating conditions excite the natural modes and verifies the predictions. An approximate formula is derived to predict the frequency of the fundamental vibration mode in terms of lamination dimensions and material properties. The formula is validated by comparison with finite element calculations for several laminations, and hence is shown to be useful in design trade-off studies. Index Terms - Acoustic noise, noise, switched reluctance, and vibration

The flux-MMF diagram has recently been used to predict torque ripple in different machine types. This paper describes a simple and elegant method, based on this technique, for predicting the effect of skew on torque ripple in permanent-magnet machines. Although it is well known that skewing minimizes torque ripple, the paper quantifies the difference in the extent of minimization between the electromagnetic torque ripple and the cogging torque. Index Terms - Skew, cogging, torque ripple.

An improved approach to predicting cogging torque in permanent magnet motors using the flux-MMF diagram technique is presented and validated. A brief review of cogging torque calculation and minimization techniques is included. It is shown that the flux-MMF diagram can be constructed for any one pole of a permanent magnet machine using the demagnetization characteristic of the permanent magnet; then by applying the principle of virtual work, cogging torque can be predicted. It is also shown that the flux-MMF diagram technique is a truly universal technique of cogging torque prediction and gives greater insight into many of the methods used for cogging torque minimization.

This paper uses the flux-MMF diagram to compare and contrast the torque production mechanism in seven common types of electric motor. The flux-mmf diagram is a generalized version of the flux-linkage versus current ([Psi]-i) diagram for switched-reluctance motors. It is illustrated for switched-reluctance, synchronous-reluctance, induction, brushless ac, brushless dc, interior PM and dc commutator motors. The calculated flux-MMF diagrams for motors with the same electromagnetic volume, airgap, slotfill, and total copper loss are shown and are used to compare the low-speed torque and torque ripple performance. The motor designs used were reasonably optimized using a combination of commercially available motor CAD packages and finite-element analysis.

A summary of the design theory associated with dc commutator motors is presented. This is followed by a detailed description of a Computer Aided Design program written specifically for this class of motor. The software engineering content is extremely important: Once the engineering design equations are developed and validated, the productivity of the design engineer depends critically on the efficiency and user-friendliness of the software itself. The program described in the main text permits the integration of the motor design with the design of the electronic controller, together with the simulation of the whole system. When sizing motors for a particular application, speed of execution and the flexibility to evaluate a wide range of design options and parameter variations are essential, while absolute accuracy is only of secondary importance.

The design of a new type of axially-laminated interior permanent magnet motor drive showing an extremely wide constant-power speed range which exceeds 7.5:1 is described. The effect on the field-weakening performance of the number of poles, the pole-piece material, the rotor bolt holes, the airgap size, the rotor insulation ratio, and the magnet type are analyzed using analytical and finite-element techniques. A 7.5-kW synchronous reluctance and a 7.5-kW interior permanent magnet axially-laminated rotor were built for a commercial induction motor stator and experimental low-speed and field-weakening results are presented.

A method is described for calculating the average and instantaneous torque of the synchronous reluctance motor from a knowledge of the trajectory of the phase flux-linkage versus phase current [i-[Psi]] waveform, i.e., the same method as used with the switched reluctance motor. This allows a direct comparison between torque production in the two motors to be made. Analytical and finite-element analysis both show that the [i-[Psi]] loci of the synchronous reluctance motor are ellipsoidal in shape and are not limited to the first and third quadrants as in the switched reluctance motor. The [i-[Psi]] loci of the synchronous reluctance motor are not bounded by the magnetization curves in the same sense as in the switched reluctance motor and rely upon mutual coupling between phases for correct operation.

The paper identifies and analyzes a number of severe fault conditions that can occur in the switched reluctance machine, from the electrical and mechanical points of view. It is shown how the currents, torques, and forces may be estimated, and examples are included showing the possibility of large lateral forces on the rotor. The methods used for analysis include finite-element analysis, magnetic circuit models, and experiments on a small machine specially modified for the measurement of forces and magnetization characteristics when the rotor is off-center. Also described is a computer program (PC-SRD Dynamic) which is used for simulating operation under fault conditions as well as normal conditions. The paper discusses various electrical configurations of windings and controller circuits, along with methods of fault detection and protective relaying. The paper attempts to cover several analytical and experimental aspects as well as methods of detection and protection.