Off-grid houses can be considered an important concept to increase the access to electricity throughout the world. Although there are quite some initiatives in place to improve the access to electricity, the implementation rate of practical solutions is far below the UN Sustainable Development Goal 7: Energy (SDG 7) + for 2030. This situation is most apparent in Sub-Saharan Africa, where the current trends of electricity access calculated by the World Bank indicate that this region will not be able to achieve the SDG 7 target. Another worldwide trend which may help to increase electricity access is that currently, a lot of renewable energy generation is realized locally in houses (especially Solar Photovoltaics (PV)). This paper reviews the recent developments to increase the access to electricity in the world and the implementation of off-grid houses in different scenarios. The focus here is on the different efforts to create off-grid houses considering their challenges on a macro and micro level. Moreover, potential research directions for technologies in off-grid houses are presented in more detail. For this, a case description of a possible off-grid house in the Netherlands is presented together with some initial simulations results for this case using solar PV, the Sea-Salt battery, and a Glycerol Fuel Cell. The simulations use the DEMkit software and the analysis is performed using measured house load data for a period in winter and in summer.

This work uses experimental data to estimate the size of storage needed for an isolated off-grid household in two different regions (Austin, Texas, US, and Nunspeet, NL). In our study, an off-grid house is considered to be supplied with 100% renewable energy during the summer period, in which cooling demand is neglected, and a solar photovoltaic (PV) system and batteries are the main electrical energy providers. Based on results achieved with the DEMkit simulation package we can conclude that, in both cases, using a solar PV system and a Sea-Salt battery would have been sufficient to provide the necessary electricity without showing a blackout during the summer of 2016. The Austin household needs a solar PV system of 38 kWp and storage of 452 kWh; in the case of Nunspeet, a solar PV system of 11.5 kWp and storage of 90 kWh is sufficient. Furthermore, using the DEMkit model, it is possible to determine an optimal value for the size of storage to half of the initial battery capacity (226 kWh for Austin and 45 kWh for Nunspeet) and still be able to provide enough power to cover the load demand of the households during the summer. In a second part, data of the solar PV system and load from Austin for one specific week was used to create data of a ‘typical’ but downscaled day. This day was used to determine the fluctuation of electricity for a real Sea-Salt battery for the considered off-grid scenario in Austin. The downscaling of the data was needed in order to have load values that fit to the size of the real battery. The tests show that the Sea-Salt battery under real electricity fluctuations is possibly adequate for off-grid scenarios.

In this work, we investigate scheduling problems for electrical energy storage systems and formulate an algorithm that finds an optimal solution with minimal charging cycles in the case of a single device. For the considered problems, the storage system is used to reduce the peaks of the production and consumption within (part of) the electricity distribution grid, while minimizing device wear. The presented mathematical model of the storage systems captures the general characteristic of electrical energy storage devices while omitting the details of the specific technology used to store the energy. In this way, the model can be applied to a wide range of settings. Within the model, the wear of the storage devices is modeled by either: (1) the total energy throughput; or (2) the number of switches between charging and discharging, the so-called charging cycles. For the first case, where the energy throughput determines the device wear, a linear programming formulation is given. For the case where charging cycles are considered, an NP-hardness proof is given for instances with multiple storage devices. Furthermore, several observations about the structure of the problem are given when considering a single device. Using these observations, we develop a polynomial time algorithm of low complexity that determines an optimal solution. Furthermore, the solutions produced by this algorithm also minimize the throughput, next to the charging cycles, of the device. Due to the low complexity, the algorithm can be applied in various decentralized smart grid applications within future electricity distribution grids.

The Ecovat is a seasonal thermal storage solution consisting of a large underground water tank divided into a number of virtual segments that can be individually charged and discharged. The goal of the Ecovat is to supply heat demand to a neighborhood throughout the entire year. In this work, we extend an integer linear programming model to describe the charging and discharging of such an Ecovat buffer by adding a long-term (yearly) planning step to the model. We compare the results from the model using this extension to previously obtained results and show significant improvements when looking at the combination of costs and the energy content of the buffer at the end of the optimization. Furthermore, we show that the model is very robust against prediction errors. For this, we compare two different cases: one case in which we assume perfect predictions are available and one case in which we assume no predictions are available. The largest observed difference in costs between these two cases is less than 2%.

An increase in the number of volatile renewables in the electricity grid enhances the imbalance of supply and demand. One promising candidate to solve this problem is to improve the energy storage. The Ecovat system is a new seasonal thermal energy storage system currently under development. In this paper, an integer linear programming model is developed to describe the behaviour and potential of this system. Furthermore, it is compared with a previously developed model, which is simplifying the behaviour of the Ecovat system much more, but is much less computationally expensive. It is shown that the new approach performs significantly better for several cases. For controlling a real Ecovat system in the future we may incorporate a number of improvements identified by our comparison analysis into the previously developed approach, which may help increase the quality of the obtained results without increasing the computational effort too much.