A new general analytical model based on elastic foundation beam theory is proposed for adhesively bonded double cantilever beam (DCB) specimens. The new model accounts for any stress state (from plane stress to plane strain) in the adhesive and therefore, the model is able to predict the mechanical response of a DCB specimens whatever the adhesive type (from rigid to flexible), the stress state, and the width/thickness relationship of the specimen are. The model is validated with DCB tests for flexible Silkron-H100 and stiff Araldite-2021 adhesives. A comparison with other models from the literature demonstrates that more accurate predictions in terms of stress distribution in the middle plane of the adhesive layer, adherend deflections and load predictions are obtained. The model is especially suitable when flexible adhesives and/or thick adhesives are used. In addition, a new method to accurately obtain Poisson's ratio for flexible adhesives is proposedThe authors gratefully acknowledge the financial support of the Spanish government through DGICYT under Contract No. TRA2015- 71491-

Different methods have been developed to estimate the energy efficiency of induction motors. The accuracy of these methods vary with the load factor, the unbalanced voltage (UV) and harmonics. The feasibility of these methods for efficiency estimation in real-time were theoretically and experimentally assessed during the operation under different operational conditions (i.e. balanced sinusoidal voltage (BSV), harmonics, UV and harmonics with UV). Results show that for load factors over 80%, the air-gap method is applicable under any condition, while the slip method is only applicable under BSV or balanced harmonic voltage. Moreover, for load factors over 40%, the nameplate method is applicable under BSV. Other methods result in errors over 8% and optimization methods are not applicable for real-time monitoring. Electric systems generally operates with some degree of UV and harmonics, while induction motors mostly operate with load factors below 60%, limiting the use of these methods for real-time measurement.

No abstract

No abstract

No abstract