Locating multi energy systems for a neighborhood in Geneva using K-means clustering

To determine the optimum location for non dispatchable renewable energy systems, this study comes up with an integrated tool to place energy systems considering distributed energy demand and renewable energy potential. The Citysim urban energy planning software is used to compute the hourly heating demand of 371 buildings in Jonction, a neighborhood in Geneva. Hourly solar irradiation on the roof tops of each building is computed using the Citysim model. The electricity demand profile for each building is generated using the hourly profile for Geneva using the databases of Swissgrid and Swiss building database. Subsequently, k-means algorithm is used to cluster the buildings based on spatial distribution. An energy economic model is used to evaluate the losses in the thermal and electrical distribution networks and initial investment. Sensitivity of cluster size is evaluated using the energy economic model to obtain an optimum number of clusters and the locations for the energy systems.

Design optimization of electrical hubs using hybrid evolutionary algorithm

Dasaraden Mauree, Vahid Moussavi Nik, Amarasinghage Tharindu Dasun Perera, Jean-Louis Scartezzini

Integration of non-dispatchable renewable energy sources such as wind and solar into the grid is challenging due to the stochastic nature of energy sources. Hence, electrical hubs (EH) and virtual power plants that combine non-dispatchable energy sources, energy storage and dispatchable energy sources such as internal combustion generators and micro gas turbines are getting popular. However, designing such energy systems considering the electricity demand of a neighborhood, curtailments for grid interactions and real time pricing (RTP) of the main utility grid (MUG) is a difficult exercise. Seasonal and hourly variation of electricity demand, potential for each nondispatchable energy source and RTP of MUG needs to be considered when designing the energy system. Representation of dispatch strategy plays a major role in this process where simultaneous optimization of system design and dispatch strategy is required. This study presents a bi-level dispatch strategy based on reinforced learning for simultaneous optimization of system design and operation strategy of an EH. Artificial Neural Network (ANN) was combined with a finite state controller to obtain the operating state of the system. Pareto optimization is conducted considering, lifecycle cost and system autonomy to obtain optimum system design using evolutionary algorithm.

Amer Soc Mechanical Engineers2016

SCCER JA-RED - Analysis of initial case and identification of potential use for renewables and waste heat at the building and district level - Deliverable 1.1.1

Lionel Benjamin Bloch, Luc Girardin, Luise Middelhauve, Mathias Raphaël Niffeler

This deliverable summarizes the current situation of all relevant aspects for the future planning and optimization of the local energy system at the demonstrator site of this Joint Activity (Rolle and Mont-sur-Rolle VD). The two communities are analyzed in terms of settlement structure, buildings, renewable energy potential and waste heat. The analysis of the settlement structure yields areas that will require special attention during planning and implementation of future changes to the local energy system – for environmental, social or technical reasons. Furthermore, it lays out existing structures, like road networks, electrical and gas-grids. The analysis of buildings shows that fossil fuels are the predominantly used source of heating energy in Rolle and Mont-sur-Rolle. Hence there is a large potential for the reduction of greenhouse gas emission. A fast simulation model to calculate time series for heating and electricity demands had been developed. This models allowed us to generate time series for each of the 1650 houses in less than 1 hour. Exemplary results for individual buildings are presented in this deliverable. These models will be an essential component in achieving the next milestones of this Joint Activity and will enable practitioners to use time-resolved demand curves in their planning processes. In the last part of this deliverable, the potential for renewable energy generation is analyzed. Estimations for thermal energy that can be provided by locally available wood, Lake Geneva and geothermal probes, as well as, estimations of the wind energy and PV potential lead us to the conclusion that the current energy demand could be entirely covered using renewable sources. Hence this deliverable prepared the basis for the next steps of developing the energy concept for Rolle and Mont-sur-Rolle.

2019

Redefining energy system flexibility for distributed energy system design

Vahid Moussavi Nik, Amarasinghage Tharindu Dasun Perera, Jean-Louis Scartezzini

A novel method is introduced in this study to consider flexibility taking into account both system design and operation strategy by using fuzzy logic. A stochastic optimization algorithm is introduced to optimize the system design and operation strategy of the energy system while considering the flexibility. GPU (Graphics Processing Unit)-accelerated computing is introduced to speed up the computation process when computing the expected values of the objective functions considering a pool up to 5832 scenarios. Subsequently, a Pareto optimization is conducted considering Net Present Value (NPV), Grid Integration (GI) level (which represents the autonomy level of the energy system) and system flexibility. The case study assesses an energy system design problem for the city of Lund in Sweden. According to the obtained NPV and GI Pareto front, a renewable energy penetration level covering more than 45% of the annual demand of the energy hub (an integrated energy system consisting of wind turbines, solar PV panels, internal combustion generator and a battery bank) can be achieved. However, the flexibility of the system notably decreases when the renewable energy penetration level exceeds above 30%. Furthermore, the results show that poor system flexibility notably increases the risk of higher-loss of load probability and operation cost. It is also shown that the utility grid acts as a virtual storage when integrating renewable energy sources. In this context, a grid dependency level of 25-30% (of the annual energy demand) is sufficient while reaching a renewable energy penetration level of 30% and maintaining the system flexibility.

2019

SCCER JA-RED - Analysis of initial case and identification of potential use for renewables and waste heat at the building and district level - Deliverable 1.1.1

Lionel Benjamin Bloch, Luc Girardin, Luise Middelhauve, Mathias Raphaël Niffeler

2019