Powering frequently utilised DC loads like LEDs, laptops, and adjustable DC motor drives is where the DC microgrid truly shines. The DC microgrid, on the other hand, is constrained by substantial voltage differences between each converter and an unequal distribution of current among the converters, which leads to a lot of circulating current. The suggested distributed droop controller effectively lowers the constraints and provides an improved response in both transient and steady states. The distribution of currents from the equality line is also quantified on the inequality curve, which resembles the Lorenz curve in economics. In order to store extra power and then give it back to the bus, energy storage devices are also incorporated into DC buses. In this case, specific controller regulates the charging and discharging modes. While transitioning from charging to discharging and vice versa, the controller is likewise put to the test. The DC system is designed to accomplish the targeted purpose as a consequence.
This paper presents the characterization of an unbalanced three-phase system based on the analysis of the Voltage Unbalance Factor (VUF) defined by the IEC standard concerning its effect on the performance of the induction motor. It is shown that for a given VUF, there are infinite values of the positive and negative sequence voltage that would impact the performance of a three-phase induction motor differently. Finally, an alternative for characterizing the unbalanced voltage system considering its effect on the induction motor performance is proposed.
En este documento se presenta una metodología de diseño de una red de telecomunicaciones con estándar IEC61850-7-420, para soportar los procesos requeridos en Redes Inteligentes y Generación Distribuida. Mediante el uso de poderosas herramientas de simulación de redes de telecomunicaciones como OPNET y MATLAB, se desarrollan y modelan los principales elementos de una "smartgrid". Esta prueba de concepto simula el comportamiento de un sistema con despacho óptimo y una respuesta a la demanda en tiempo real de donde se extraen estadísticas que permiten evaluar la posible implementación del sistema de telecomunicaciones.
Integrating smart grids with the renewable energy present in Colombia expands the country's energy matrix and strengthens it through distributed generation in non-interconnected areas. Therefore, integrating distributed generation sources is of great importance to meet the growing demand, security of electricity supply, and system efficiency. The integration of distributed generation sources needs several management algorithms, control schemes, and an adequate environment to verify the proper functioning. This paper shows the implementation in an emulated microgrid environment based on Solar, Wind, Water, Biomass, and Diesel generation. Each generation source and control was implemented on a Raspberry PI 3 and linked with a server to manage the microgrid. Implementation schemes, control algorithms, and microgrid testbed are displayed throughout the paper.
This paper aims to investigate the use of machine learning techniques for PV power forecasting. In this study, a bibliometric review was conducted to establish a baseline, which involved the use of a SARIMA model as a statistical model, XGBoost model for machine learning, and LSTM for deep learning. These models were trained and evaluated using a sliding window technique and a validation set. The study’s development utilized the generation data collected from May 2018 till February 2023. The dataset underwent preprocessing to fill in the gaps in the data and to analyze trends, seasonality, and stationarity. The study compared three models and found that all of them could predict PV power generation. However, the XGBoost model outperformed the other two in training and validation, resulting in an R-squared value of 0.99, an nMAE of 0.35%, and an nRMSE of 0.55%, indicating that its predictions were the closest to actual values. The study’s limitations and potential areas for future research were discussed at the end.
