Short-term local forecasting by artificial intelligence techniques and assess related social effects from heterogeneous data

• Autores: Bing Gong
• Directores de la Tesis: Joaquín Bienvenido Ordieres Meré (dir. tes.)
• Lectura: En la Universidad Politécnica de Madrid ( España ) en 2017
• Idioma: español
• Tribunal Calificador de la Tesis: Antonio Bello García (presid.), Ángel Uruburu Colsa (secret.), Maurizio Bevilacqua (voc.), Ana González Marcos (voc.), Manuel Castejón Limas (voc.)
• Programa de doctorado: Programa de Doctorado en Economía y Gestión de la Innovación por la Universidad Autónoma de Madrid; la Universidad Complutense de Madrid y la Universidad Politécnica de Madrid
• Materias:
  • Ciencias tecnológicas
    • Tecnología de los ordenadores
      • Ordenadores analógicos
    • Ingeniería y tecnología del medio ambiente
      • Control de la contaminación atmosférica
• Texto completo no disponible (Saber más ...)
• Resumen

  • This work aims to use the sophisticated artificial intelligence and statistic techniques to forecast pollution and assess its social impact. To achieve the target of the research, this study is divided into several research sub-objectives as follows:

    First research sub-objective: propose a framework for relocating and reconfiguring the existing pollution monitoring networks by using feature selection, artificial intelligence techniques, and information theory.

    Second research sub-objective: comprehensively assess the influences factors of socio-economic activities on pollutions at the district level in a macro environment, in order to explore the pollution formation mechanism.

    Third research sub-objective: introduce an innovative method of pollution forecasting, including regional factors, and apply it to a case study to predict ozone threshold exceedances.

    Fourth research sub-objective: provide a way to develop a multi-level energy usage awareness system, which can provide energy usage, CO2 emission, and the cost of energy usage at operation, product, and order level.

    Fifth research sub-objective: identify the patterns of dynamic changes of energy usage, CO2 emission, and GDP by using cluster algorithms and find the corresponding driven factors for the clusters differences.