A fault tolerant switched reluctance machine 

Context in source publication

Context 1

... definition, fault tolerance of a system is its ability to continue its work even if a failure occurs. The fault tolerant concept emerged for the first time in information technology. It meant an increased level of continuous operation of computer equipment. Later more and more fault tolerant equipments were connected together in order to form a fault-tolerant system [1, 2]. The result was an operational unit having certain fault tolerant level, as a sum of the safety levels of each equipment of the system. A system is reliable when it is capable of operating without material error, fault or failure during a specified period in a specified environment. From another point of view a system is dependable if it is available, reliable, safe, and secure [3]. An electromechanical system is driven by a unit composed of the power converter and of the electrical machine. Both must be fault tolerant. The electrical machine's fault tolerance design has to be in a manner to assure unchanged, as possible, output parameters also in case of fault occurrence. To be able to achieve an optimum solution for a fault tolerant machine, all the advantages and drawbacks have to be taken into account for the new structure. From the inverter's point of view, as the evolution of the power electronics hit an exponential slope, the separation of command and control of each phase will set the required fault tolerance level [4]. Critical electrical machines and drives systems used in diverse fields like aerospace, defense, medical, nuclear power plants, etc. require both special, fault tolerant motor and converter topologies. For example for electric drives used in propulsion applications faults can be critical, since an uncontrolled output torque may have an adverse impact on the vehicle stability, which ultimately can risk the passenger safety. All theses mentioned above have stimulated the researches in the field of fault-tolerant electrical machines and drives [5]. In our days due to the recent technological advances and developments in the area of power electronics and motor control the fault tolerant electrical machine and drive concept reached a level where it begun to be feasible to be used widely in practice There are a lot of elements in an electromechanical system that can be changed to reach for the summits of fault tolerance concept. Hence, modifications in the machine’s topology proved to be the trickiest method for improvement. It was proved that changing a small parameter in a machine’s geometry, raises huge amount of output changes. As a second step the winding can be modified. Diverse way to design the winding, different placement or couplings between the coils were proposed. Using great number of phases became a widely used design method in solving fault tolerant problems. The main idea was to substitute the faulted phase’s contribution by the healthy remaining ones. II. GEOMETRY APROACH FOR FAULT TOLERANCE Electrical machine’s geometries can be changed to obtain fault tolerant designs. Depending on the construction and type of the machine, several improvements can be applied. Usually when a custom machine is designed, there are some basic criteria that the machine has to fulfill (minimum losses, mass, etc.). These criteria must be also taken into account when designing fault tolerant machines. Hence, a possible solution should be the use unsymmetrical stator pole teeth. Its usefulness was proved in [5]. By a small change in the stator structure the winding losses were reduced by near 44%, which is a consistent improvement in the machine's design. Another solution in the case of multi-phase machines should be to increase significantly the number of teeth (see Fig. 1.) [4]. This increased number of teeth will assure low torque ripples also in case of different faults. As much the number of teeth is increased, the torque ripple will be more diminished Other designers proposed models of fault tolerant topologies with a lower number of teeth, a shown in Fig. 2. This of course the simple magnetic circuit structure was completed by a more complex winding: high number of phases and a complex winding plan. This solution also requires a complicated drive unit [7]. Fractional slot winding configurations allow the machine to well operate also in faulty conditions. However, the MMF harmonics in this case are consistent, and this might cause an unbalanced saturation and an unbearable torque ripple. Fault tolerant topologies are also applicable for embedded permanent magnet electrical machines designs. The magnets assure a part of the magnetic flux needed for the force generation, even when a part of the winding is damaged. Changes to be performed on the already existing designs of electrical machines in order to achieve fault tolerance is quite difficult. To optimize permanent magnet fault tolerant synchronous machines, in [8] armature coils are placed in pairs of slots. The walls of the slots are parallel to themselves and to the wall of the other slot of the same slot pair. Between two pair poles, there is a spacer tooth (see Fig. 3). These are narrower than the main teeth to provide greater magnetic flux linkage between them and thus increasing the EMF within the corresponding coil. For a switched reluctance machine, as the rotor is a passive one, this solution cannot be applied to achieve a fault tolerant variant. The solution is to make the rotor active by placing a winding in closed loop around each rotor pole (as shown in Fig. 4). This will assure a reaction for the stator to rotor flux, resulting in a rotor to stator flux. Practically, this emphasizes a magnet; hence, the rotor will generate the force needed for passing over a pole having damaged winding ...