Efficient simulation of Hybrid Renewable Energy Systems

This paper examines the possibilities for significant energy innovation in rural locations in developed countries. It thereby questions the dominant framing of energy experiments and ‘living labs’ as urban. We discuss findings from empirical research with a rural community on Bruny Island, Australia, where a 3-year research project (2016–19) – CONSORT – funded by the Australian Renewable Energy Agency (ARENA), is underway with approximately 35 householders trialing a new residential battery storage and photovoltaic energy system. Bruny Island has a problem of peak demand for electricity during tourist periods, and a back-up diesel generator is currently used to supply electricity during peaks. An alternative solution is being trialled through CONSORT: household-level battery storage, which can be drawn upon by the utility to supply the grid as required. In this paper we explore two energy geography issues: first, how global and national energy challenges are manifesting on Bruny Island through the CONSORT project, and, second, the ways in which the particular sociotechnical context of Bruny Island has influenced the CONSORT project, creating tensions as well as opening up opportunities for energy innovation.

Rural laboratories and experiment at the fringes: A case study of a smart grid on Bruny Island, Australia

Multi-source and multivariate ozone prediction based on fuzzy cognitive maps and evidential reasoning theory

In this paper, effective and systematic steps in the mathematical modeling of high-fidelity battery models for simulating energy storage systems (ESS) will be presented. Two approaches to battery modeling will be discussed in this article: (1) the equivalent electrical circuit approach, and (2) the electrochemical approach. The battery models discussed in this article are developed based on the Modelica Standard Library specification 3.2.1 and commercialized as part of the Battery Component Library in MapleSim ® 2015.

/pdf/developing-mathematical-models-of-batteries-in-modelica-for-4f4wweflwr.pdf

Developing Mathematical Models of Batteries in Modelica for Energy Storage Applications

The spread of electrical storage devices continues to be underpinned by the limited charging currents that can be applied. The limitation arises from the lack of sufficient high power charging stations, either at home or along roads and highways, and from the maximum admissible current that can be applied to the battery before undesirable degradation mechanisms are triggered. Accordingly, most traditional charging protocols limit the charging current as a function of the standing state of charge of the battery. These protocols are designed empirically and restrict the potential benefit of more flexible charging options. However, the alternative to traditional protocols must rely on a more precise knowledge of the operating constraints and on advanced control techniques to compute online the best operating plan. This work presents a model predictive control (MPC) application to minimize the charging time of a lithium-ion battery subject to electrochemical and thermal constraints. The satisfaction of these constraints ensures that the battery degradation is minimized, or at least mitigated. The programming language Modelica and the optimization and simulation framework JModelica.org is used in combination with Python language to assess the computing time and potential use of MPC and the developed cell models in commercial batteries.

/pdf/a-model-predictive-control-application-for-a-constrained-28hq2gpyc2.pdf

A Model Predictive Control Application for a Constrained Fast Charge of Lithium-ion Batteries

Battery energy storage systems (BESS) are one of the most common elements for energy storage. The usage of these systems in renewable energy and traction sectors is gaining wide interest. A BESS has a finite life-cycle, determined by many influencing factors. An electrical storage models offer the possibility to simulate the behavior of these systems and their aging. We developed an equation based model in the open source object-oriented modeling language Modelica. We considered physical phenomena and parameters such as charge/discharge rate, calendar and cycling ageing, battery cell’s internal resistance variations, temperature effects and self-discharge rate. The model has been validated with experimental data from our load management test site. By reducing the order of the model and eliminating discrete modelling functions, it is possible to use the library in combination with an optimal control framework. This enables to test multi-objective optimisation algorithms for BESS coupled with photovoltaic power generation and to estimate the system optimal sizing and control strategies.

A Modelica Library for Simulation of Electrical Energy Storage Coupled with Photovoltaic Systems

The ability to forecast the concentration of air pollutants in an urban region is crucial for decision-makers wishing to reduce the impact of pollution on public health through active measures (e.g. temporary traffic closures). In this study, we present a machine learning approach applied to the forecast of the day-ahead maximum value of the ozone concentration for several geographical locations in southern Switzerland. Due to the low density of measurement stations and to the complex orography of the use case terrain, we adopted feature selection methods instead of explicitly restricting relevant features to a neighbourhood of the prediction sites, as common in spatio-temporal forecasting methods. We then used Shapley values to assess the explainability of the learned models in terms of feature importance and feature interactions in relation to ozone predictions; our analysis suggests that the trained models effectively learned explanatory cross-dependencies among atmospheric variables. Finally, we show how weighting observations helps in increasing the accuracy of the forecasts for specific ranges of ozone's daily peak values.

A data-driven approach to the forecasting of ground-level ozone concentration.

The diffusion of the photovoltaic installations is reaching significant levels in the low voltage electric grid, therefore the need of effective energy management strategies increases. Within the Swiss2Grid (S2G) pilot project we are designing and testing a tool that allows the decentralized management of the distributed generation and consumption by controlling local energy storage systems and shifting the activation of households loads. In order to assess the effects of higher levels of diffusion of photovoltaic and storage system into the low voltage grid, we modeled the pilot residential neighborhood in Modelica, a multi-domain, open-source modeling language. In this paper we describe the components and the selected penetration scenarios. We finally present the results of the simulations, analyzing the effects of the chosen scenarios on neighborhood overall power, voltage instabilities and self-consumption.

Modeling and Simulation of a Residential Neighborhood with Photovoltaic Systems Coupled to Energy Storage Systems

A data-driven approach to forecasting ground-level ozone concentration

The PETRA project aims at developing an computer-based tool for decisions-making in networks for building estate management that encompasses both a rapid assessment and the planning of renovations according to different indicators.
It is addressed to all people or institutions managing a mixed building estate and particularly, to technical services for project management, architects and engineers, owner‟s associations and expert offices. In order to establish a correct maintenance management programme, these « clients » are in need of a method for analyzing buildings that allows to rapidly and simply
«scan » the building estate and estimate the typology and the investment costs of rehabilitation interventions.
The aim of the project is to bring innovation by offering extra speed and universality for better answering the real estate market‟s new requirements in the area of maintenance and renovation. Indeed, the market needs fast and precise methods that help the expert to reconcile
the needs of heritage protection and new energy requirements. Within the next 10 to 30 years, the biggest building estate conversion ever will take place in Switzerland. This implies the refurbishment of buildings from the years 1950-1970 that are now at the end of their lifetime and that have a 2-3 energy consumption reduction factor. Given this reality, an owner cannot only trust the manager‟s intuition ; this intuition must be reinforced by appropriate analytical tools such as the new method that will be developed with this project. PETRA develops a new evaluation methodology that allows to analyze the buildings‟ conservation state and to
calculate the renovation costs of each building in a simple and quick way. In order to carry this out, a process of data aggregation and element grouping, based on existing methods previously developed by the work group (EPIQR+1, INVESTIMMO2), will be set up.

/pdf/computer-based-tool-petra-for-decision-making-in-networks-3ikjc4s2xa.pdf

Davide Strepparava

Papers

A Modelica Library for Simulation of Electrical Energy Storage Coupled with Photovoltaic Systems

A data-driven approach to the forecasting of ground-level ozone concentration.

Modeling and Simulation of a Residential Neighborhood with Photovoltaic Systems Coupled to Energy Storage Systems

A data-driven approach to forecasting ground-level ozone concentration

Computer-based tool « PETRA » for decision-making in networks about the maintenance and renovation of a mixed building estate