Development of a microscale traffic emission simulation system. Application to madrid for the definition of air quality improvement measures / desarrollo de un sistema de simulación de emisiones del tráfico rodado a microescala. Aplicación a madrid para la definición de medidas de mejora de la calidad del aire

• Autores: Christina Quaassdorff
• Directores de la Tesis: Rafael Borge Garcia (dir. tes.)
• Lectura: En la Universidad Politécnica de Madrid ( España ) en 2018
• Idioma: español
• Tribunal Calificador de la Tesis: Adolfo Narros Sierra (presid.), Juan Manuel de Andrés Almeida (secret.), José Luis Santiago (voc.), Tomás Ruiz Sánchez (voc.), Sotiris Vardoulakis (voc.)
• Programa de doctorado: Programa de Doctorado en Ingeniería Ambiental, Química y de los Materiales por la Universidad Politécnica de Madrid
• Materias:
  • Matemáticas
    • Ciencia de los ordenadores
      • Simulación
  • Ciencias de la tierra y del espacio
    • Meteorología
      • Contaminación atmosférica
  • Ciencias tecnológicas
    • Ingeniería y tecnología del medio ambiente
    • Planificación urbana
      • Transporte
• Enlaces
  • Tesis en acceso abierto en: Archivo Digital UPM
• Resumen

  • Road transport is often the main source of air pollution in urban areas, significantly affecting air quality in cities such as Madrid (Spain). Due to the adverse effects of atmospheric pollution, the European Union has established legislation to improve air quality and reduce air pollution. Although Madrid is expected to satisfy air quality limit values at a mesoscale level, pollutant concentrations are influenced by very local phenomena, and exceedances or non-compliance situations at some specific points are assumed for much larger areas. In this sense, macroscale models have been successfully applied to Madrid. Nevertheless, they do not adequately reflect the driving effects, simplifying acceleration-deceleration processes that significantly influence emissions at a finer scale and do not allow their calculation in short periods of time. It is essential to develop emission reduction measures that focus on hot-spots and highly polluted micro-environments. On these specific points, finer-scale tools are needed due to the complexity of the processes that determine emissions from mobile sources and also to estimate the contribution of traffic to atmospheric emissions at local level.

    For those cases, a microscale emission estimation methodology has been developed and the performance of a suitable combination of traffic and emissions micro-simulation models that accurately define the emissions in a specific area has been tested. These models do not only describe the behaviour of the single entities which define the traffic such as individual vehicles and their interactions, but they also take into account traffic signs and lane changes for each vehicle. This results in stop & go cycles that produce braking-acceleration patterns with a huge influence on emissions. So that, in order to obtain detailed traffic emission results a modelling system based on a microscale traffic flow simulation model (VISSIM) fed with real-measured data collected by measurement campaigns that serves inputs for a microscale emission model (VERSIT+micro through the ENVIVER interface) was implemented.

    Several microscale traffic simulations representative for different types of urban hot-spots (roundabout, signalized intersections, combination of urban roads) and for different daily and weekly traffic conditions were carried out. Speed management influences in a complex way vehicle emissions so that differentiating between congested and non-congested scenarios is useful to suggest control strategies. Traffic results were used as input for the microscale emission simulation model to obtain detailed estimations of NOX and PM10 emissions related to different vehicle types with five meters resolution.

    This modelling system provides detailed traffic emissions for complex urban hot-spots locations. Nevertheless, as an evaluation test, NOX results for passenger cars were compared with those obtained from the well-known mesoscale emission model COPERT 4. Also, modal emission models PΔP and PHEM-light have been successfully applied, analysed and compared to VERSIT+micro for two case studies. Results points out the relevance of congestion and a detailed vehicle composition definition for accurate emission estimations. In the case of very detailed modal models implementation also engine power is a very important parameter. In turn, these parameters relay on an accurate estimation of the speed-acceleration pattern, road gradient and vehicle loading. So that, larger number of vehicle classes included in a particular model implies a better chance to provide representative emissions estimates.

    These detailed emission results from VERSIT+micro has been successfully coupled to stationary simulations from highly detailed air quality models based on CFD-RANS. Nevertheless, non-stationary simulations with CFD in urban areas needs second-by-second resolved emissions that can be provided by modal emission models.

    However, real-world emission measurements are essential towards the validation of emission factors and total emissions computed by these modelling techniques. In this sense, a first approach to the validation of the high resolution emission estimation modelling system VISSIM-VERSIT+micro/ENVIVER on a real hot-spot using mobile laboratory data has been performed.