The design of aerodynamic airfoils are optimized for certain conditions. For instance, the shape of the wings of fixed-wing aircrafts are designed and optimized for a certain flight condition (in terms of altitude, speed, aircraft weight, etc.). However, these flight conditions vary significantly during the flight. Currently, aircraft are provided with control surfaces such as flaps, slat and ailerons, normally governed by powerful but heavy hydraulic mechanisms. These moving parts allow the aircraft to fly under many different flight conditions, although usually with non-optimal performance. Moreover, these mechanisms introduce hinges and surface discontinuities between parts which cause undesirable effects such as turbulences and noise or a decrease of the lift-to-drag ratio. This issue motivates the research and development of the so-called ‘morphing aircraft’ or ‘morphing wings’. Ideally, a morphing aircraft is able to modify quickly the shape of its wings in-flight, thus reaching optimum aerodynamic performance under any flight condition. This idea is applicable to any other aerospace applications such as rotorcraft or wind turbines. Morphing applied to aerodynamic airfoils brings along interesting benefits: reduction of mechanical fatigue which has a special importance in wind turbines and rotorcrafts (by minimizing vibrations on the structure), reduction of the wing root bending moment, reduction of fuel consumption of flying machines and increase of the performance of wind turbines by increasing the lift-to-drag ratio of the wings or blades, and the reduction of generated noise. This dissertation gives an introduction on the concept of morphing applied to aerodynamic airfoils and describes the benefits and challenges brought by morphing structures. Subsequently, the potential of smart materials to develop novel actuation systems is introduced. The description of the motivation for this work leads to the purpose of this research, which aims at developing a prototype of a general-purpose morphing flat surface based on embedded shape memory alloy wires with increased working frequency for morphing airfoils. In order to achieve this purpose, the research is divided into two research objectives. The first research objective is to develop a profound understanding of the behaviour of shape memory alloys (SMAs) and, subsequently, to devise a method to increase their actuation frequency. The second research objective is to develop novel control algorithms and actuator technology as well as an integration technology for the SMA wires with increased actuation frequency. Following the introduction, this thesis describes the potential of smart materials in general, and SMAs in particular, to develop novel smart actuators for morphing wings. The main differences between conventional and smart actuators are explained together with their advantages and introduced challenges in terms of design and control. The selection of SMAs is justified as the best candidates to achieve the research objectives. A detailed description, working principles, features, capabilities, limitations and applications of SMA based actuators is given as well. In order to fulfill the first research objective, a series of logical steps were followed. Once the requirements of the SMA based actuator were stated, a commercial SMA wire was chosen and characterized (i.e., the phase diagram of the SMA wire was obtained) which shows the phase composition of the SMA when it is subjected to different levels of stress and temperature. The phase diagram was obtained from data collected from isothermal, isobaric and differential scanning calorimeter tests. During an isothermal test, the temperature is kept constant and the SMA wire is subjected to increasing and decreasing levels of stress in order to find those stresses at which the SMA’s phase transforms. Similarly, an isobaric test keeps the stress on the SMA constant over time and the temperature is increased and decreased to find those at which the SMA’s phase transforms. Subsequently, the functional fatigue of the SMA wire was studied by training the SMA wire. This training process consist in applying repetitive heating and cooling cycles in an isobaric configuration. High repeatability was found on the results. After an average of 7289 training cycles, the wire was able to recover only 77% of its original recoverable length for the tested conditions of applied current, heating time, cooling time and applied stress. Subsequently, an SMA model for SMA wires is implemented in a finite element analysis software. The equations that describe the model and their physical meaning are explained. One of the advantages of the chosen model is the ease of obtaining the parameters required by the model, which can be obtained from a few experiments. In addition, and also as a part of the model’s required parameters, the characteristics of the forced cooling system are studied here, and the heat transfer coefficient of such airflow for different airflow rates is measured. Finally, the model is satisfactorily validated. One of the limitations of SMA based actuators is their poor actuation frequency (usually lower than 0.1Hz). This is due to the fact that they are thermally activated, normally by Joule heating, which is a quick process. However, they must be cooled before the next actuation cycle, which normally happens by natural convection. This cooling process can be accelerated by means of active cooling systems. In this research, an active cooling airflow at room temperature is used. The effects of applying different heating and cooling rates on the time that it takes for the wire to contract and elongate (respectively) were experimentally measured using an isobaric configuration. In addition, the same experiments were simulated by the model for SMA wires implemented previously. There is an overall quantitative disagreement between the results yielded by the simulations and by the experiments. However, the results are useful qualitatively. It is found that the contracting and the cooling times are significantly decreased as the applied power and airflow is increased. More importantly, it is found that when the wire works at low working frequencies, the heating rate is the limiting factor whereas the cooling rate is the limiting factor when it works at high working frequencies. In addition, these experiments show that increasing the level of applied stress results in slightly higher working frequencies of the SMA wire. The knowledge acquired in the previous experiments and simulations leads to the development of a method to improve the attainable actuation frequency of the SMA wires. This method is based on the idea that, in many SMA based applications, the SMA wires do not work throughout their full recoverable strain but they work only within a portionofit. Taking advantage of the nonlinearity of the strain-temperature relationship for SMAs, the method proposed here is able to increase the SMA’s actuation frequency by three and a half times just by making the SMA wire work within the most suitable range of strains, without varying the heating, cooling and stress conditions. The development of this method fulfills the first research objective. Later on in this thesis, the design, manufacturing and assembling processes of the SMA based actuator as well as their challenges are detailed. First, a beam-like module of the SMA based actuator was conceived, designed and manufactured. Subsequently, the design was expanded in the spanwise direction, thus obtaining a modular SMA based actuator that forms a morphing plate. The morphing plate was tested by heating the SMA wires embedded in the actuator. This test revealed an unexpected behaviour of the plate. The expected behaviour was that it would bend upwards when the wires on the top side were heated. However, instead of bending uniformly along the spanwise direction, the center of the plate bended upwards and the sides downwards. A similar response was observed when the wires on the bottom side were heated. This was found to be caused by inhomogeneous thermal expansion (in the spanwise direction) through the thickness of the plate. Due to the unexpected response observed on the morphing plate, a single beamlike module SMA based actuator was controlled under two different control strategies, fuzzy logic control (FLC) and proportional-integrative-derivative (PID) control. In both cases, the SMA based actuator was made to track sinusoidal and step signals in order to measure its performance. The overall performance of the system under FLC is better than that under PID. Under FLC, it reaches actuation frequencies above 0.6Hz for working at amplitudes of up to 6mm, and frequencies above 1Hz for amplitudes of up to 3mm while tracking the reference signal accurately (maintaining relative error below 10%). Under PID control, it reaches actuation frequencies above 0.5Hz when working at amplitudes up to 3mm, and frequencies above 0.7Hz for amplitudes up to 2mm. The actuator is able to track step signals (that is, to reach and maintain a constant deflection over time) under both types of controller, although under FLC it is significantly more stable. These developed design, manufacture, assembly and control methods fulfill the second research objective. This thesis presents novel methods aiming to increase the accuracy and actuation frequency of SMA based actuators. In particular, this work is focused on the development of a morphing surface intended to be integrated in morphing airfoils. However, the methods and ideas developed in this research are applicable to other SMA based applications, especially those which require fast, cyclic and accurate actuation.