PhD research

Past (already completed) PhD projects

Dr. Stefano Mandelli: Strategies for access to energy in developing countries: methods and models for off-grid power systems design

This doctoral thesis is part of the research topic related to “energy for sustainable development in developing countries.” In general terms this theme refers to the analysis of specific problems of the energy sectors of developing countries (DCs) and the resulting consequences on the sustainable development of these countries. Looking at the energy sector of DCs, a “problem of energy” is typically mentioned. This basically consists in the strong energy dependence on traditional biomass (firewood and charcoal) and low rates of electrification and low per capita consumption of electricity. From these two aspects derive a series of negative consequences that hinder the development process at local, but also at country level. In particular, the population of rural areas of DCs is one that most suffer the consequences of the problem of access to energy. In fact, these areas are often not reached by the electricity grid and the only energy source is traditional biomass. In particular the absence of electricity supply severely limits the ability to improve the capability to meet basic services both at home and community level. In many rural areas, however, strong technical and economic feasibility constraints limit the provision of electricity supply via the traditional paradigm of the centralized system. In these cases, off-grid systems based on renewable energy sources (RESs) are the only viable solution. Nevertheless, the process leading to the identification of the best technical off-grid solution within a specific application context is not trivial, and there are many research topics open in this regard. In this context, this thesis mainly focuses on the problem of access to electricity in DCs and on the analysis of off-grid systems for electrification of rural areas. The thesis is organized into two parts which deal, through different chapters, with two specific topics respectively. The first part offers an in-depth analysis and capitalization of the issue of access to energy in DCs with particular attention to the problem of rural electrification.

Dr. Moshi Godfrey Gladson: Optimal planning of hybrid microgrid

Microgrids stand as the best approach for integration of Distributed Energy Resources (DERs), Energy Storage Systems (ESSs) and conventional generators in order to electrify rural or remote areas in least developed countries (LDCs). Statistics show that 97% of population without access to electricity live in LDCs in sub-Saharan Africa and developing Asia. Studies have also shown that by 2030 only 30% of these un-electrified areas can be reached by grid extensions. In most cases grid extensions are infeasible due to large distances from the main grids, difficult terrains, and low population with highly dispersed settlements patterns. However, most of un-electrified regions, such as sub-Saharan Africa, are rich of fossil fuels, natural gas, and important renewable resources such as solar, wind, hydro, geothermal. The remaining challenge is on the planning of rural off-grid electrification systems. Proper planning is of paramount importance in order to realise all technical, economical, and environmental benefits of various types of DERs and ESSs applied for off-grid electrification. In particular, this thesis addresses the problem of optimal planning of hybrid microgrids with solar Photovoltaic (PV) arrays, Wind Turbines (WTs), Storage Battery Banks (SBBs), and conventional Diesel Generators (DGs). Microgrid planning requires joint optimization of operation and selection of capacities, quantities, and combination of components of different types and technologies. Integration of renewable energy based generation technologies, storage systems, and conventional generators present technical and economic challenges that must be considered in the planning of hybrid microgrids. Economically, conventional generators such as DGs have lower investment costs but higher operation costs; whereas renewable energy based generators such as solar PVs and WTs have higher investment costs but lower operation costs. Technically, generation from PVs and WTs, are subject to uncertainties and variations of weather conditions, and thus they are not fully dispatchable. These uncertainties and high variations in microgrid demand make Battery Energy Storage System (BESS) necessary, particularly for stand-alone microgrids. One of the main challenges in microgrid planning is to ensure that the components to be installed will offer minimum life cycle operational cost while fulfilling all required technical constraints. Consequently, hourly dispatching of DGs, BESS and other sources, which determine the overall life cycle operational cost, must be considered in the planning of hybrid microgrids. This thesis applies mathematical programming and optimization approach in planning of hybrid microgrid considering long term operational constraints. The aim is to obtain the optimum capacities, combination, and number of components to install in a microgrid in order to ensure reliable and continuous supply of its demand at minimum cost. A novel Mixed Integer Linear Programming (MILP) deterministic model for microgrid planning is proposed. The overall microgrid long term operation is integrated in this planning model. A technique called Clustered Unit Commitment (CUC) is applied in order to reduce the number of discrete variables required to model DGs operation. In additional, in order to make the model computationally tractable, K-medoids clustering algorithm is applied to select typical representative days with profiles of renewable resources and demand data. Piecewise Linear Approximation (PWLA) of components nonlinear characteristics is carried out to enable the use of CPLEX and GUROBI solvers in General Algebraic Modeling System (GAMS) The deterministic planning model is extended to include uncertainties in renewable resources and electric demand in microgrid planning, based on Two Stage Stochastic Integer Programming (2SSIP) and Robust Optimization (RO) frameworks. Applicability of the proposed models are demonstrated by using microgrid planning case studies.

Dr. Claudio Brivio: Battery energy storage systems: modelling, applications and design criteria

Nowadays, the specific costs of battery energy storage systems (BESSs) are decreasing exponentially and at the same time their installations are increasing exponentially. BESS are in fact becoming pivotal in the development of several heterogeneous industrial sectors like energy, automotive, electronics, telecom etc. However, BESS performances (energy density, power density, efficiency, lifetime) cannot be assumed expandable from one application to another and from one technology to another. Therefore, methods and models have to be developed to end up with a proper design criteria for the selected application. The General objective of the thesis is to contribute in expanding the knowledge about BESSs by focusing on appropriate methodologies capable of linking the technological studies with the economic analyses required in real life applications. The dissertation is centred on electrochemical batteries, considering power electronics well-established with respect to both industrial applications and mathematical modelling. Specific objectives of the thesis are: the development of a reference framework related to technologies, performances and modelling of BESS; the proposal of innovative BESS models representing dynamic and aging phenomena; the development of proper methodologies to analyse the techno-economic performances of BESS when deployed in stationary applications. The work is theoretical, numerical and experimental. A theoretical framework serves to identify and formulate the correct BESS models. The experimental activities are fundamental in developing and tuning the models. The numerical analyses, based on field data, are needed to test and validate the models on real applications. These themes are specifically developed for lithium-ion battery technology and stationary applications. With respec tto the developing countries issues, Chapter 9 deals with the design of off-grid power systems for rural electrification. A novel sizing methodology is proposed which is composed of separated blocks addressing the different sizing phases: data elaboration, load and source profiles formulation, modelling of the main components (e.g. BESS) and their simulation, heuristic optimization method to formulate the robust design from a technoeconomic perspective. The procedure has been applied to design a PV+BESS microgrid system in supplying power to a rural village of Tanzania. Also in this case, simulations show that different BESS models can bring to different sizing results, especially if very simplified empirical models (based on literature/manufacturers data) are adopted. However, given the less stressful application, proper empirical models (i.e. based on laboratory test) can bring to a similar conclusion with respect to more complex electrical models. This fact provides a great opportunity in energy planning analyses like the one proposed because it can reduce the simulation time by more than ten times. The analyses are based on real data gathered within the framework of the Energy4growing project. The methodology is proposed in the form of a computational tool in MATLAB® named Poli.NRG (POLItecnico di Milano –Network Robust design).

Dr. Mina Mirbagheri: Renewable Energies Integration on Distribution Grid

Currently, renewable energy development is emphasized for sustainable development goals accomplishment and the better realization of sustainable development globally. Several African countries such as Tanzania, like other developing countries, is striving to adopt different ways of ensuring affordable and accessible energy supply to its socioeconomic and political sectors to achieve renewable energy development. To secure affordable and accessible energy in the country, renewable energy is termed as an alternative energy source because of it is environmentally friendly. If renewable energy is produced and utilized in a modern and sustainable manner, it will help to eliminate energy problems in these countries. However, upward trends of installing dispersed generators cause some power quality challenges for distribution system operators, such as harmonics, voltage regulation issues, interface protection problems and power quality in general.

Therefore, the focus of the thesis is on appropriate management of distributed generation. However, as available electrical network data in emerging country is limited, it is required to estimate the unknown grid parameters using available measurements data such as active power, reactive power and voltage magnitudes. Then, the evaluation of maximum active power injection to the grid by distributed generation units (named Hosting Capacity), its applications and its voltage control are studied.

To do so, a novel procedure to estimate the single bus hosting capacity even in case of uncertainties in grid parameters or lack of data is presented, named Bricks approach. This approach could be used when the possibility of data collection is very low or complex. Then, multi-generator hosting capacity is evaluated using the combination of Bricks approach and a suited Monte Carlo procedure. In addition, electric vehicle integration as the other discussed application in this thesis is looked into by Monte Carlo simulation for different charging processes.

At the end, to increase the hosting capacity a novel procedure based on optimal power flow is proposed. The goal is to avoid over and under voltages violations and preserve the grid efficiency. Actually, the procedure is proposed in order to optimally set-up the standard voltage control setting, i.e. to be directly integrated into the already in place voltage regulators. All of the mentioned simulations in this thesis work are coded in MATLAB and then validated for a real-life case studies in Tanzania.

Dr. Matteo Moncecchi: Renewable Energy Communities

When we speak about Energy Communities (ECs) we generally refer to groups of citizens who organize themselves to actively contribute to the energy transition, producing energy and meeting their energy needs through the exploitation of renewable sources. Beyond this, the EU has recently provided formal definitions for the ECs and all the Member States are required to introduce them into their national legislation, ensuring an enabling framework to promote and facilitate their development. These recent innovations have stimulated new interests in methods and models to properly deal with the ECs peculiarities. ECs are not isolated microgrid; the energy produced is shared among the members of the community using the public infrastructure or also exchanged with other actors on the market. Therefore, self-sufficiency is an important aspect but it is not a technical requirement and it is not mandatory to achieve a complete autarky. On the contrary, the optimal planning and operation of an EC are driven by economical evaluations that take into account both energy exchanges among community members and with the external energy system. This thesis provides some elements to investigate benefits and risks correlated to ECs, evaluating the issue from three different perspectives: the one of the EC as a whole, the one of the EC members (i.e. citizens, municipalities and SMEs that participate to the EC), and the one of the system in which they are hosted (i.e the public distribution network). The first part of the thesis concerns the definition of a reference framework, that is composed by a legislative framework and a research one. In the legislative framework, the models of ECs defined in the European Directives are presented and analysed. These are the Renewable Energy Community (REC) and the Citizen Energy Community (CEC). Moreover, a detailed description of the Italian scenario is proposed starting from the historical energy cooperatives to the current process of transposition of the European Directives, and the characteristics of the experimental phase currently ongoing in Italy is analysed. The second element of the reference framework focuses on the ECs state of the art in the scientific literature. This clearly shows that the interest in ECs has grown very fast in recent years, but most of the studies are not yet aligned with the new EU definitions. Some classifications of community-based initiatives are reported and a review of the most interesting research projects currently focusing on ECs is provided. The second part of the thesis deals with methods and models for the analysis of ECs. A model capable to evaluate energy and economical exchanges within a REC is proposed. The peculiarity of the model is the ability to consider separately the self-consumed energy and shared one, to properly evaluate their economic values based on the different tariff structures. The goal of the model is to find the optimal DERs portfolio in terms of installed generators and storage capacity, optimizing the net present value of the EC investment. The proposed model and methodology constitute a tool that supports the EC planning. Then, the issue of benefit distribution among the EC members is addressed. Game theory algorithms are identified as a suitable approach for this purpose. Therefore, some elements from the cooperative and non-cooperative game theory are presented and examples of application in energy sharing situations are considered. The proposed REC model is formalized as a cooperative game, and a two steps distribution rule, based on the Shapley value among clusters of users followed by a proportional allocation, is proposed. The methodology is applied to a real-life case study of EC with more than one hundred members based on the Italian scenario. Finally, changing perspective, the point of view of the distribution system operator is also considered, and the impact that ECs could have on the MV distribution network is tackled. The relationship between distributed generation and EC is discussed and a review of the hosting capacity concept is provided. Then, a methodology based on Monte Carlo simulation is proposed to evaluate the capacity of a network to host new ECs. Two study cases with different characteristics have been built and the procedure is tested on different real-life MV networks.



The Thesis work considers the electricity markets in the presence of Battery Energy Storage Systems (BESS), Renewable Energy Sources (RES) and Distributed Energy Resources (DERs). It aims to provide a panoramic view of the possible regulatory evolutions of electricity markets, with particular focus on the Ancillary Services Markets (ASMs).
The work starts from analyzing in detail the ongoing evolution of ASMs and the possible trade-offs that it arises, by both the perspectives of the System Operator (SO) and the Balancing Services Provider (BSP). To quantitatively assess the performance of BESS (standalone or integrated with other DERs) in the provision of ancillary services, a numerical model of BESS is developed. This is aimed to overcome a set of lacks found in the battery models present in literature, that makes them unsuitable to be used for testing ASM participation. The importance of control strategies for improving the reliability on BESS on the markets is highlighted. A multiple-service control strategy is developed for providing energy management, RES integration, flexibility to the power system with a high degree of reliability and improving economics. The Multiservice strategy exploits dynamic service stacking and guarantees revenue stacking. The main outcome of the work is the estimation of optimal ASM arrangements for the best interaction of BESS, RES, and markets. A sensitivity analysis on a set of ASM parameters, recognized as potential regulatory barriers, is carried out. The optimal values of the ASM parameters are returned. Finally, the impact on the grid of fast frequency regulation by BESS is estimated with a novel methodology.
The work is structured to offer valid technical modeling, market insights and advice to policymakers.


Dr. Silvia Corigliano: Geospatial based methodology for rural electrification planning

The research work presented in this manuscript has the general goal of addressing the problem of energy access in rural areas of the Global South, proposing effective solutions that could foster the electrification process considering both on and off-grid technologies. More specifically, the present thesis aims to give a significant contribution to the literature models for rural electrification planning by creating a new open source and open access modeling framework, usable by different stakeholders, from university researchers, to Non Governmental Organizations and private companies. A novel procedure, named Gisele (GIS for electrification) has been developed, coded and validated over the real-life case studies. The proposed approach is a comprehensive model, subdivided into four blocks, able to estimate the energy needs of an area, size the optimal generation portfolio for off-grid systems, identify the optimal electrification solution and design the distribution network. With respect to other literature tools its main strengths reside in the ability of including a detailed sizing of hybrid microgrid and routing of the distribution grid, the reliance on optimization algorithms for the optimal electrification solution identification and the integration of a multi-objective optimization including environmental and social dimensions.
The procedure has been tested for designing the rural electrification plan in a area in Mozambique and Lesotho. In both cases, that differ in terms of load distribution and geographical characteristics it performed well being able to plan an electrification strategy considering both grid extension and off-grid systems.


Dr. Marina Petrelli:

Marina Petrelli is carrying out a PhD in the area of rural electrification, with a particular focus on microgrids. She is working on a tool for microgrid optimization in the Global South. The aim is to develop an instrument able to account for the complexity of a rural electrification project by including multi-year evaluations (degradation of assets, demand growth) and multi-objective decision making, for a comprehensive assessment of the technical, economic, social and environmental sustainability of the project.

Ongoing PhD projects

Dr. Benson Mbuya

Benson Mbuya is another PhD candidate cooperating with Energy4Growing team. He is an Arusha Technical College staff currently pursuing PhD studies at NM-AIST in Sustainable Energy Science and Engineering and specializing in renewable energy. He is currently working on Distributed energy management system, and load forecast studies of hybrid microgrid in emerging countries. Therefore, Energy4Growing team has Arusha Technical College (ATC) and The Nelson Mandela African Institution of Science and Technology (NM-AIST) as collaborating partners.

Dr. Aleksandar Dimovski

Aleksandar Dimovski is developing his PhD research in the area of rural electrification and development of emerging countries, mostly focusing on the planning of the national network and optimizing its development. The ultimate goal of his project is to develop procedures and tools capable of planning the expansion of the national grid considering different generation portfolios and the feasibility of operating the grid in such a topology, while adopting publicly available open-source data. The outputs of the procedures give indications on substation placement, generation portfolios and accurate routing of the electrical network. For such a complex task, various known algorithms, including optimization and clustering techniques are adopted and being improved. The target is not only to focus on the steady state planning, but also on the dynamic feasibility of the proposed solutions.
Moreover, a second direction of the research is to focus on the planning of microgrids for remote area, not only considering a techno-economical steady state optimization, but also taking into account the dynamic phenomena. This research path will be adopted not only for emerging countries, but remote areas such as islands in the developed world as well.

Dr. Darlain Edeme

Darlain Edeme is carrying out a PhD in the area energy planning and rural electrification, named “Multi-scale and multi-year GIS-based rural electrification planning in Developing Countries”. He is in particular working on a GIS-based tools for energy needs assessment for energy access planning. The aim is to develop an instrument capable of identifying communities in large areas, characterizing them on the basis of socio-economic parameters, and on the basis of these predicting current and future electric energy needs with a scenario-based approach.

Dr. Matteo Spiller

Matteo Spiller is carrying out a research project focused on Battery Energy Storage Systems devoted to develop steady state and transient models capable to properly evaluate which services a battery could effectivelly provide in a market environment. Within the research area, topics related to the stability of an electric power system strongly based on inverter generators (and batteries) are going to be investigated.