Abstract
Electrical utilities apply condition monitoring on power transformers (PTs) to prevent unplanned outages and detect incipient faults. This monitoring is often done using dissolved gas analysis (DGA) coupled with engineering methods to interpret the data, however the obtained results lack accuracy and reproducibility. In order to improve accuracy, various advanced analytical methods have been proposed in the literature. Nonetheless, these methods are often hard to interpret by the decision-maker and require a substantial amount of failure records to be trained. In the context of the PTs, failure data quality is recurrently questionable, and failure records are scarce when compared to nonfailure records. This work tackles these challenges by proposing a novel unsupervised methodology for diagnosing PT condition. Differently from the supervised approaches in the literature, our method does not require the labeling of DGA records and incorporates a visual representation of the results in a 2D scatter plot to assist in interpretation. A modified clustering technique is used to classify the condition of different PTs using historical DGA data. Finally, well-known engineering methods are applied to interpret each of the obtained clusters. The approach was validated using data from two different real-world data sets provided by a generation company and a distribution system operator. The results highlight the advantages of the proposed approach and outperformed engineering methods (from IEC and IEEE standards) and companies legacy method. The approach was also validated on the public IEC TC10 database, showing the capability to achieve comparable accuracy with supervised learning methods from the literature. As a result of the methodology performance, both companies are currently using it in their daily DGA diagnosis.

Abstract
The power distribution grid is centrally managed concerning the requirements of the end-users, however, with the appearance of smart grids, new technologies arc arising. Therefore, distributed energy resources, mainly, renewables, energy storage systems, electric mobility, and power quality are viewed as encouraging contributions for improving power management. In these circumstances, this paper presents a power electronics perspective for the power distribution grid, considering innovative features, and including a power quality perception. Throughout the paper are presented relevant concepts for a concrete realization of a smart grid, supported by the integration of power electronics devices as the interface of the mentioned technologies. Aiming to support the innovative power electronics systems for interfacing the mentioned technologies in smart grids, a set of developed power electronics equipment was developed and, along with the paper, are shown and described, supporting the most important contributions of this paper.

Abstract
Minimising the total power losses and enhancing the voltage profile is achieved using a proposed multi-objective chaotic salp swarm algorithm with fuzzy logic decision-making. The proposed multi-objective chaotic salp swarm algorithm is utilised to determine the optimal size and location of photovoltaic in radial distribution system to minimise the total power losses, total voltage deviation, and maximise the voltage stability index. In addition, the proposed multi-objective chaotic salp swarm algorithm is used to find suitable scheduling for battery energy storage charge/discharge during 24 h considering the intermittent nature of photovoltaic power generation. The proposed algorithm is tested on standard and practical radial distribution systems (IEEE 33-bus and 94-bus Portuguese systems). The performance of the proposed algorithm is validated by comparing its results with those obtained by other competitive optimisation techniques. The obtained results prove the ability of the proposed algorithm to achieve an efficient setting for photovoltaics and battery energy storages and determine their optimal allocations in order to minimise the power losses and enhance the voltage profile with satisfying all operating constraints.

Abstract
The present work aims to compare several load disaggregation methods. While the supervised alternative was found to be the most competent, the semi-supervised is proved to be close in terms of potential, while the unsupervised alternative seems insufficient. By the same token, the tests with long-lasting data prove beneficial to confirm the long-term performance since no significant loss of performance is noticed with the scalar of the time-horizon. Finally, the patchwork of new parametrization and methodology fine-tuning also proves interesting for improving global performance in several methods.

Abstract
In Europe, clean distributed generation, DG, is perceived as a crucial instrument to build the path towards carbon emission neutrality. DG already reached a large share in the generation mix of several countries and the reduction of technical losses is one of its most mentioned advantages. In this scope, this paper discusses the weaknesses of this postulation using real networks. The adopted methodology involves the power flow simulation of a collection of real networks, using 15 min real measurements of loads and generations for a whole year. The clustering of similar cases allows identifying the situations that cause higher losses. A complementary objective of this research was to define an approach to mitigate this problem in terms of identifying the branches that, if reinforced, most contribute to losses reduction. The results obtained confirm the rationality of the proposed methodology.

Abstract
The financial planning of distribution systems usually includes the prediction of annual mandatory investments, concerning the resources that the DSO is compelled to allocate as a result of new network connections, required by new consumers or new energy producers. This paper presents a methodology to estimate the mandatory investments that the DSO should do in the distribution network. These estimations are based on historical data, load growth expectations and various socioeconomic indices. However, the available database contains very few annual investment examples (one aggregated value per year since 2002) compared to the large number of variables (potential inputs), which is a factor of regression overfitting. Thus, the applicable regression techniques are restrained to simple but efficient models. This paper describes a new methodology to identify the most suitable estimation models. The implemented application automatically builds, selects, and tests estimation models resulting from combinations of input variables. The final forecast is provided by a committee of models. Results obtained so far confirm the feasibility of the adopted methodology.