The current project is within the framework of Scientific Initiation in a project related to artificial neural networks for damage detection in structure. The project is lead by Prof. Herbert M. Gomes. The concept is based on measuring the dynamic characteristic of structures and identifying patterns associated with the damage present (fissure) in simple structures. The damage detection in structure is complex, and there are many methodologies to address the problem. Among these methods, the changes introduced in dynamic features are the ones with the most prominent results. Some use parametric models, while others are based on the damaged structure response. Independent of the type of research, both lack reference data from experimental and theoretical solutions. Numerical methods have been demonstrated to be an effective technique to access the problem, detecting cracks in robotic arms and real-time detection in raisers and industrials pipes. This project addresses an analytical proposal for the natural frequency modification in simple beam structures as the position of the cracks change. The study is focused on Euler-Bernoulli beams only.