HUES GSoC Projects 2017

From HUES Platform Wiki


Logo cropped2.png

Background - the HUES Platform

The Holistic Urban Energy Simulation (HUES) Platform is an open source ecology of computational resources to support distributed energy system (DES) design and control. By bringing together diverse computational resources reflecting cutting-edge DES research, we aim to accelerate research and facilitate effective deployment of DES.

The HUES platform is managed by the Urban Energy Systems Laboratory at Empa. Empa is the Swiss Federal Laboratories for Materials Science and Technology, an interdisciplinary Swiss research institute for applied materials sciences and technology.

Application instructions: Please submit your application via the GSoC portal. We ask that you please submit a brief CV/resume and a brief (e.g. 1 page) proposal describing: (1) Which of the projects you would like to be involved in - see our "Ideas List" page? (2) How you would go about completing the project? (3) What aspects of your background/experience prepare you for this?

Questions: For questions or further information about a specific project, please contact the relevant project mentor.

Projects for GSoC 2017

1. GIS Interface for energy hub and network definitions

Mentors: Julien Marquant, Georgios Mavromatidis, Boran Morvaj

Project description: Spatial aspects are very crucial to the effective realisation of energy hubs as their placement and the possible connections between hubs and/or buildings must be considered. This project aims to facilitate the description of candidate energy hub configurations in urban environments. More specifically, the GUI-based user-friendly assignment of energy technologies for different locations in a city and the possible network configurations that interconnect energy entities.

The project’s task list encompasses the following:

  • GIS data manipulation for defining elements like roads, buildings, custom nodes etc. and imposing of constraints associated with them
  • Development of a GUI for the user-friendly definition of such constraints.
  • GUI-based assignment of energy technologies to each energy hub pulling from a set of candidate technologies.
  • GUI-based specification of network type (e.g. electrical, thermal etc.) and components.
  • Connection of the energy system elements defined via the GUI with a Python-based energy hub modelling environment (see GSoC project: Modular energy hub modelling framework).
  • Visualisation of the simulation results in the GUI.

The developed tool will be openly available to researchers and practitioners on the HUES website.

Required knowledge: Good Python programming skills, experience with programming and GUI development in GIS environments (QGIS)

Difficulty level: Easy/Moderate

2. Visualization Dashboard for Empa-NEST

Mentors: Philipp Heer

Project description: Empa-NEST (Next Evolution of Sustainable Technologies) is a research building which represents a vertical city district. The goal is to study and optimize the energy flows between different parts of the district. This includes living homes, working places, and leisure activities and many more. The heart of the building is an energy hub, which provides stores and distributes energy in form of thermal, chemical and electrical energy. As a result of the complexity of the energy hub many sensors like temperature, pressure, energy flow and actors like heat pump, fuel-cell and batteries are installed and produce data, big data. The data is logged into a SQL Database in the NEST-Cloud, which is accessible for researchers and technicians from all over the world. To bring the big data to life it is essential to have an attractive visualization based on Google Charts.

The aim of this project is to develop a visualization dashboard based on Google Charts which has a connection to the Microsoft SQL Database and can be integrated into a Website.

Required knowledge: HTML, PHP, Java Script, Google Charts, Microsoft SQL Database

Difficulty level: Moderate

3. Visualization module for the Ehub Modeling Tool

Mentors: Andrew Bollinger

Project description: The Energy Hub (Ehub) Modeling Tool is a set of scripts for writing and executing a district energy system optimization. The purpose of the tool is to aid the design of decentralized energy systems at the building and district level. The current version of the Ehub Modeling Tool is programmed in Matlab, but is being transported to and further extended in Python.

This project focuses on developing a Python-based visualization module for the Ehub Modeling Tool. The results from a run of the model include the optimal dispatch schedule of different technologies, the technologies to be installed and the optimal capacities of these technologies, and other data. This project will involve developing a set of scripts to import and visualize the results data from a set of experiments. The challenge is to do this in a flexible way that can also be extended and tailored by users.

Required knowledge: Python, Data visualization, General knowledge of energy systems concepts

Difficulty level: Easy/Moderate

4. Script development and data analysis in the DES data portal

Mentors: Ashreeta Prasanna

Project description: This project involves the further development of an existing open source database. Development of scripts to collect and analyse additional data in the platform, including statistics on energy demand, supply, production technologies and prices. This task will entail:

  • The development of scripts to automatically extract relevant data from selected websites which publish data on energy production, costs, etc, and write this data to a common database.
  • Further development of a tool for querying, visualizing and downloading data from the database.
  • Development of a tool to run basic statistic functions on data from the database and present it in specified formats.

The developed portal will be used by researchers and practitioners for the development of models and analyses of distributed energy systems. Some knowledge or interest in energy, energy production would be appreciated.

Required knowledge: Knowledge of python, basic statistics, interest in energy, renewable energy, data analysis.

Difficulty level: Moderate/Easy

5. Developments on Fast Fluid Dynamics (FFD) model

Mentors: Christoph Waibel

Project description: This project aims to contribute to further development of an open-source Fast Fluid Dynamics library in C#.Net, which was developed at the last year's GSoC program ( [1] ). Anticipated additions to the current library are: adding energy (Boussinesque approximation) and pollution (Fick's law) transportation, adding and benchmarking numerical solvers of partial differential equations (PDE) for finite difference, and parallelizing the code (e.g. the PDE solvers) on graphics processing unit (GPU). An alternative possibility on contributing to the FFD library is the implementation of a meta-model to approximate the spatially resolved turbulent viscosity of the fluid domain - currently only a single constant for the kinematic viscosity is used throughout the domain. This includes the identification of relevant features, selection and evaluation of different meta-models, training and implementing the model into the FFD code, as well as verification and validation with data from CFD. The application purpose of the FFD model is for urban wind studies for providing a fast model for an informed design process of architects and urban planners. The library has since been implemented in a 3D-CAD tool for this purpose ( [2] ).

Required knowledge: Numerical solution of Partial Differential Equations, Fluid Dynamics, coding on GPU. Alternatively experience with meta-modelling / function approximation.

Difficulty level: Moderate

6. Analysis aid for energy research

Mentors: Philipp Heer

Project description: Empa-NEST (Next Evolution of Sustainable Technologies) is a research building which represents a vertical city district. The goal is to study and optimize the energy flows between different parts of the district. This includes living homes, working places, and leisure activities and many more. The heart of the building is an energy hub, which provides stores and distributes energy in form of thermal, chemical and electrical energy. As a result of the complexity of the energy hub many sensors like temperature, pressure, energy flow and actors like heat pump, fuel-cell and batteries are installed and produce data, big data. The data is logged into a SQL Database in the NEST-Cloud, which is accessible for researchers and technicians from all over the world.

The aim of this project is to develop a tool which is able to find causal correlations of events to ease analysis of vast amounts of data form the energy hub. The tool will be used by various researchers and it needs to cope with a changing environment as the present set of sensors and actors changes over time.

Required knowledge: Microsoft SQL Database

Difficulty level: Moderate

7. Modular energy hub modeling framework

Mentors: Julien Marquant

Project description: A modular and flexible platform is needed to facilitate the implementation of large scale models for the complex study of the interactions of multiple energy systems at different scales. A framework has been developed in an object-oriented fashion to easily construct and configure systems of modular entities (energy demands, sources, convertors, storages and network links). These systems are hierarchically nestable, facilitating the construction of large systems at high level of details. The platform has been developed in Python, and is based on an open source energy modelling framework Oemof. The goal of the project is to continue the platform development to increase its functionalities. The following tasks will be performed:

  • Linear programming: improvements of the hierarchical structure of the platform, allowing constraints and variable domain definition management.
  • Approximation methods: aid model solvability.
  • Programming: advanced nesting functionalities.
  • Mapping of the structure with existing database.

The developed framework will be integrated in a decision making tool for the optimisation of the interactions of multi-energy sources and carriers. This tool will be openly available to researchers and practitioners on the HUES website.

Required knowledge: Knowledge of python, object-oriented, program architecture, good programming skills required, interest in energy/energy modelling and linear programming.

Difficulty level: Moderate

8. Development of a Mediawiki library for multi-model initiatives

Mentors: Andrew Bollinger

Project description: Increasingly, researchers and practitioners in the energy domain need to consider insights from multiple models and datasets. This leads to the proliferation of multi-model projects - projects involving the development, management and integration of potentially numerous models, algorithms, datasets and other resources. Currently lacking is a web-based infrastructure for enabling the development and management of these resources in an effective and coherent manner.

The purpose of this project is to develop a MediaWiki library to facilitate the development and management of multi-model initiatives. The setup of the HUES Model Repository provides a starting point for this, but requires further extension and refinement, and consolidation in the form of a MediaWiki library. For instance, the current setup should: (1) be extended with an improved interface for exploring models and datasets in a multi-model project, (2) be adapted to integrate and automatically visualize (potentially large) datasets. The setup should be easily extensible and applicable to multi-model initiatives in different industries/domains.

Required knowledge: MediaWiki

Difficulty level: Moderate

9. Propose your own project

We are open to original project proposals generally in line with the development needs of the HUES Platform.