Modelling covering energy and climate modelling integrating behavioural changes
date: 18/12/2019
- At the University College Cork researchers are developing a new global energy model using TIMES framework with upgraded spatial and temporal resolution and inclusion of high-performance computing. They are generating physical demand for transport using machine learning algorithms that use custom deep learning architectures. The transport sector data set is currently generated for OECD countries with yearly resolution. The key modelling challenge addressed by the project is to project demands using Shared Socioeconomic Pathways (SSP) scenario drivers. A prototype Neural Network architecture[1] was developed to project transport demands. This is the first application of deep learning architectures in transport service demand. The generated energy service demands is expected to feed into the CHIMERA (China Ireland Modelling Energy Research Assessments) project to build a global energy model using TIMES (presentation online).
- Researchers at Imperial College London have published an article in Applied Energy on the integration of energy systems and gas infrastructure optimisation models in Brazil. They have developed a novel soft-link integration of the simulation-based integrated assessment model MUSE (ModUlar energy systems Simulation Environment) and a new Gas INfrastructure Optimisation model (GINO), which contains details on the position and availability of gas resources in five Brazilian states. The model was applied to assess whether the existing gas infrastructure in the five southern states of Brazil could meet the expected increase in gas demand by 2050. Results suggest that additional investments in the regional gas infrastructure would be required. Depending on the outcome of the ongoing gas imports renegotiations between Brazil and Bolivia (i.e. either maintaining constant, reducing by half, or halting altogether the imports of Bolivian gas), natural gas demand could be covered by a share of alternative supply options, such as an increase in pre-salt production, LNG imports, and imports from a new Argentinian pipeline. The research aids policy makers by modelling cost-effective natural gas infrastructure pathways. Energy planning decisions based on up-to-date models and data can contribute to better planning for climate change mitigation actions.
[1] A set of algorithms, modelled loosely after the human brain, that are designed to recognize patterns.