Anàlisi mitjançant simulació intervalar de llaços de control amb incertesa estructurada

• Autores: Josep Cugueró Escofet
• Directores de la Tesis: Teresa Escobet (dir. tes.), Jordi Saludes i Closa (codir. tes.)
• Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2011
• Idioma: catalán
• Tribunal Calificador de la Tesis: Vicenç Puig (presid.), Ramón Estruch Riba (secret.), Jorge Bondía Company (voc.), Joaquim Armengol Llobet (voc.), Jesús Picó (voc.)
• Materias:
  • Matemáticas
    • Investigación operativa
      • Sistemas de control
  • Ciencias de la tierra y del espacio
    • Ciencias de la atmósfera
      • Simulación numérica
  • Ciencias tecnológicas
    • Ingeniería y tecnología aeronáuticas
      • Estabilidad y control
    • Tecnología de la instrumentación
      • Ingeniería de control
• Texto completo no disponible (Saber más ...)
• Resumen

  • The stability analysis of control loops is usually based on the analysis of the model equations. Different model types yield different analytical frameworks, e.g. linear system theory, hybrid system theory, among others. This thesis takes the model simulation process as the framework for analyzing stability and asymptotical stability tendency of trajectories. Pointwise simulation of one trajectory is extended to set based simulation, which takes into account all the possible trajectories. Different works using set based simulation are discussed, with special emphasis on the ones having a similar analysis aim. Without significative loss of generality this thesis is focused on interval sets.

    Problems derived from the extension from pointwise to interval based models are described and discussed. These discussions are one of the contributions of this thesis, especially with regard to the relation between set properties and stationarity properties of models.

    Furthermore, the main contribution of this work is to present a set of interval based algorithms which allow stability and asymptotical stability tendency verification of trajectories for discrete and continuous time models of dynamic systems. These algorithms have the property of being less complex as the degree of stability being verified is greater.