• Energy Research

Abstract Energy Storage Systems (ESSs) are progressively becoming an essential requisite for the upcoming Smart Distribution Systems thanks to the flexibility they introduce in the network operation. A rapid improvement in ESS technology efficiency has been seen, but not yet sufficient to drastically reduce the high investments associated. Thus, optimal planning and management of these devices are crucial to identify specific configurations that can justify ESSs installation. This consideration has motivated a strong interest of the researchers in this field that, however, have separately solved the optimal ESS location and the optimal ESS schedule. In the paper, a novel multi-objective approach is presented, based on the Non-dominated Sorted Genetic Algorithm – II integrated with a real codification that allows joining in a single optimization all the main features of an optimal ESS implementation project: siting, sizing and scheduling. The methodology has been tested on a real-size rural distribution network.

The paper presents a risk-based distribution network planning procedure to perform a comparison (in terms of costs and associated residual risks) among conventional planning solutions and the exploitation of flexibility purchased from distributed energy resources through bilateral contracts or local markets. The procedure has been integrated within software developed by the Authors in the past decades for distribution network expansion planning. The software already includes many of the main distinctive characteristics for a modern planning tool, such as abandoning the traditional worst-case approach, resorting to non-network planning options, and implementing the stochastic network assessment to consider generation and demand uncertainties. Since many flexibility resources are connected to the low voltage system, both medium voltage and low voltage networks have to be jointly analysed to account for their mutual interactions. The planning process has been applied to distribution networks representative of the Italian distribution system. The low voltage system has been represented by replicating few real networks provided by the leading Italian Distribution System Operator. Consumption and generation patterns have been modelled from real anonymised measurements.

The growing interest on environmental issues and the increasing cost of fossil fuels have determined the conditions for high shares of Distributed Generation in the distribution system. Nowadays distribution networks are approaching a critical point whereby the connection of DG will require an active approach. The paper presents a Distribution Management System to manage an active distribution network economically and safely.

This paper presents a new approach to the problem of system decoupling in a power-system state estimation problem. The complexity of power systems is growing, thus challenging the way measurements for state estimation are traditionally managed. Following a previous experience in defining a decentralized solution for state estimation, the authors here propose a decentralized dynamic state estimation method for a large-scale power system in combination with a procedure to automatically identify how and which state information to exchange for reconstructing the states starting from partial knowledge. In particular, the problem of selecting the variables that each observer has to estimate is partially solved within the framework of a stochastic approach, i.e., the so-called collocation method. An optimization algorithm based on dynamic programming is also developed to determine the optimal set of strongly coupled variables necessary for a sufficiently accurate estimation. The developed method is evaluated by applying to an IEEE test bus system.

Energy management systems for the operation of distributed energy resources, distribution storage devices, and responsive loads will be embedded in distribution management systems (DMSs) as advanced functions that rely on accurate input data and fast communication signals. For a proper DMS design, the impact of the state estimation uncertainties and of the communication system delays should be evaluated. The paper presents an integrated software package for the cyber-physical simulation of DMS taking into account the accuracy of state estimation and measurements. The effectiveness of a wireless Wi-Max communication system has been tested on the rural Italian representative network produced by the research project ATLANTIDE. Results proved the need of cosimulation packages in modern distribution systems.

The concept of electrical-mobility, in opposition to the present oil-mobility, is attracting the attention of politicians and of civil society worldwide. Electrical mobility means the usage of battery powered Electric Vehicle (EV) and Plug-in Hybrid Electric Vehicle (PHEV) as the main future technology to combat greenhouse gas emissions. The burden of electric mobility will be mainly on the distribution system that, particularly during the peak hours, will be exposed to critical operation conditions by a high number of high density simultaneous loads. Vehicle-to-Grid technology by adding control capabilities to charge and discharge of cars' batteries can exalt the benefits from their whole energy storage capacity. Distributors can then be helped in the active management of the network by the services offered (e.g., VAR/volt regulation, frequency regulation, spinning reserve, integration of renewable generation). Vehicle-to-Grid is perfectly part of the emerging Smart Grid technology and is based on intelligent stations fully integrated within the distribution management system. For a full exploitation of Vehicle-to-Grid potentialities, the role of the aggregator is essential to create value to customers by offering services to the distribution system operator. In the paper, a Particle Swarm Optimization is used to define the aggregator's optimal control strategy to optimize the recharge/discharge patterns of a fleet of EVs taking into account financial contracts, driver's behavior, energy prices, etc.

The foreseen increasing diffusion of Distributed Generation (DG) in the distribution networks is giving the impetus to the study of active distribution networks. The recent EU directive 54/2003 may help to accomplish these objectives by defining DG as a planning alternative for distribution expansion problems. In the paper, considering the importance of interconnection in active networks, meshed schemes have been deeply studied both in steady state and transient condition in order to point out the effect on the integration of DG. In particular, different types of meshed distribution networks have been examined and compared to the radial scheme. Steady state simulations have been performed in order to assess the impact on the voltage profile, on the short circuit level, on the power losses, on the continuity of service, and on the exploitation of lines and transformers. Finally, the greater attention has been paid to analyse the dynamic behaviour of meshed distribution networks. In order to complete the study a mix of different DG sources — directly connected generators or with static converters — have been considered.

A heuristic optimization algorithm based on the Dynamic Programming theory is proposed to find the optimal placement of measurement devices, i.e. to determine their number and position. The optimization procedure explicitly considers network reconfigurations (caused by random faults or by the active management of the network), so that the final measurement system allows the distribution state estimation to provide an accurate estimate of the system status in all the possible practical conditions. The branch currents are taken as state variables for improving the quality of the solution of the state estimator that exploits field measurements and load pseudo-measurements. The uncertainties introduced by the measurement chain are simulated with a Monte Carlo algorithm. Variations of both load demand and network parameters are also modeled in the Monte Carlo algorithm. The provided examples show the effectiveness of the optimization process.

Voltage dips are the most common power quality disturbances. They can often cause serious damages to the end- users, especially to the industrial ones or those involved in services that today are almost entirely based on digital processes. Thus, their mitigation is assuming a key role for both customers and distributors. The first ones look for eliminating any obstacle to the correct operation of their processes, and are often willing to stipulate premium contracts that guarantee a high level of quality to their power supply. The second ones have to solve power quality problems in their networks with restrict available budget, first of all, to achieve the Regulatory targets on quality but also to satisfy the most exigent customers. In previous works, the authors proposed a distribution network planning methodology that adds to the traditional technical constraints also the target of the power quality. In terms of voltage dips, the power quality target to be achieved may be limiting the voltage dips frequency in each node of a given network within a prefixed threshold. In this paper, an improved algorithm able to find the most economical solutions to comply with this constraint has been developed. This improved algorithm considers by the same standards different corrective actions, with the aim at finding the best compromise between investment and benefits. Accordingly with this view, the goal is minimizing the cost that the distributors have to sustain for the protective measures, the cost paid by the customers for the damages caused by the not mitigated disturbances and for the premium quality contracts. The effectiveness of the proposed approach is proved by examples in a test network and in a small one derived from real cases.