The growing demand for new technologies in several fields : recreation, medicine, transport, telecommunications and the space industry, etc. is increasingly related to the field of the management, transformation and use of electric power. In this domain, the development of no-ferromagnetic integrated electromechanical actuators finds wide application and thus implies an usefulness over a large diversity of geometries. The absence of ferromagnetic materials does not allow for quick determination of an equivalent electromagnetic circuit for this type of actuator. The interaction of the magnetic induction fields produces phenomenas like the skin effect and the proximity effect which degrade its performance. In such circumstances, there is a compromise to be made between the analytical model's complexity and the precision of the results. Often, a modeling based on numerical approaches, which uses finite element methods, is the best adapted approach. Nevertheless, this type of analysis requires much experience and in particular, the parameterization of the actuator specifications and the exploitation of such a model in the optimized design is not very practical and requires considerable computing time. The use of discretized analytical models constitutes a good compromise between the degree of complexity and the model's correspondence to reality. Accordingly, in this thesis, we propose a suite of discretized analytical models allowing for the design and the optimization of this type of actuator. The development of a general rule to determine the optimal discretization step allows an arbitration between computing time, degree of complexity, and the precision of the results. The identification and the validation of the analytical models describing the skin and proximity effects have made it possible to improve the computation of self-inductance and winding resistance. The development of a design and optimization methodology for this type of actuator permits each application to be optimized in a systematic and efficient way, by proposing one or more possible solutions with their advantages and their disadvantages. Two distinct applications were conceived and optimized adopting this design methodology : the Montrac® system and the Iglus® system. In the Montrac® project, it was a question of modifying the power system of a modular assembly line. This system is composed of a fixed track with motorized shuttles for use in a clean room environment. The replacement of the electrical contacts by a contactless energy transmission system was dimensioned and optimized in order to guarantee an operation compatible with the requirements of the above mentioned environment. In the Iglus® project, a subcutaneous biomedical actuator for glycemic level detection, with the aim of measuring the glucose level of diabetes patients was optimized. The measurements and results from these two applications, made it possible to confirm the validity of the models expounded in this thesis.