Eduardo Perez-Anaya

The importance of power quality has increased these days due to the growth in the use of renewable energies and nonlinear loads. Although the use of renewable energies provides power generation sources that help in reducing greenhouse gas emissions, they might have a detrimental effect on the power quality due to their intermittency and dependence on weather conditions. Due to the importance of keeping an optimal power quality, in this work, a novel methodology is developed whose main contribution relies on the use of entropy features and envelope analysis for the detection and quantification of power quality disturbances. The proposed method is implemented within a machine learning framework, where linear discriminant analysis (LDA) is employed to optimize entropy-based features. Subsequently, a neural network classifier performs an automatic classification and quantifies the magnitude of affectation associated with grid disturbances. The training is performed using synthetic signals, and validation is conducted with real signals from a photovoltaic park and from an IEEE working group. The results obtained are compared with those provided by other methodologies proving the accuracy and the viability of the proposed approach.

Energy generation through renewable processes has represented a suitable option for power supply; nevertheless, wind generators and photovoltaic systems can suddenly operate under undesired conditions, leading to power quality problems. In this regard, the development of condition-monitoring strategies applied to the detection of power quality disturbances becomes mandatory to ensure proper working conditions of electrical machinery. Therefore, in this work we propose a diagnosis methodology for detecting power quality disturbances by means of the continuous wavelet transform (CWT) and convolutional neural network (CNN). The novelty of this work lies in the image processing that allows us to precisely highlight the discriminant patterns through spectrograms into 2D images; the images are cropped and reduced to a standard size of 128x128 pixels to retain the most relevant information. Subsequently, the identification of six power quality disturbances is automatically performed by a convolutional neural network. The effectiveness of the proposed method is validated under a set of synthetic data as well as a real data set; the obtained results make the proposal suitable for being incorporated into the monitoring of power quality disturbances in renewable energy systems.

The use of renewable energies is increasing around the world in order to deal with the environmental and economic problems related with conventional generation. In this sense, photovoltaic generation is one of the most promising technologies because of the high availability of sunlight, the easiness of maintenance, and the reduction in the costs of installation and production. However, photovoltaic panels are elements that must be located outside in order to receive the sun radiation and transform it into electricity. Therefore, they are exposed to the weather conditions and many environmental factors that can negatively affect the output delivered by the system. One of the most common issues related to the outside location is the dust accumulation in the surface of the panels. The dust particles obstruct the passage of the sunlight, reducing the efficiency of the generation process and making the system prone to experimental long-term faults. Thus, it is necessary to develop techniques that allow us to assess the level of dust accumulation in the panel surface in order to schedule a proper maintenance and avoid losses associated with the reduction of the delivered power and unexpected faults. In this work, we propose a methodology that uses a machine learning approach to estimate different levels of dust accumulation in photovoltaic panels. The developed method takes the voltage, current, temperature, and sun radiance as inputs to perform a statistical feature extraction that describes the behavior of the photovoltaic system under different dust conditions. In order to retain only the relevant information, a genetic algorithm works along with the principal component analysis technique to perform an optimal feature selection. Next, the linear discrimination analysis is carried out using the optimized dataset to reduce the problem dimensionality, and a multi-layer perceptron neural network is implemented as a classifier for discriminating among three different conditions: clean surface, slight dust accumulation, and severe dust accumulation. The proposed methodology is implemented using real signals from a photovoltaic installation, proving to be effective not only to determine if a dust accumulation condition is present but also when maintenance actions must be performed. Moreover, the results demonstrate that the accuracy of the proposed method is always above 94%.