El presente artículo describe los modelos de optimización recientemente aplicados al diseño y operación de los sistemas de potencia hacia la conformación de las redes inteligentes e identifica las tendencias, barreras y posibles brechas en esta área. Se describen modelos para optimizar el diseño y la operación de los sistemas de potencia considerando las energías renovables, la generación distribuida, las micro redes, la gestión de la demanda y los sistemas de almacenamiento de energía. Se concluyó que es necesario validar muchos de los modelos que se han formulado recientemente para la optimización de la operación mediante pruebas con datos reales y a gran escala. Además, la gestión de la demanda y las micro redes son aspectos en los cuales se requieren desarrollar modelos para el flujo óptimo de potencia. Finalmente, es necesario predecir con mayor precisión las variables estocásticas para que estos modelos se adapten al comportamiento real del sistema.
The growing integration of renewable energy sources and the electrification of multiple sectors have heightened the need for optimized planning and operation of modern electrical distribution systems. A critical challenge for distribution network operators is enhancing the resilience and reliability of their grids by identifying effective solutions. One promising approach to achieving this is through the deployment of battery energy storage systems, which can rapidly inject power to mitigate the impacts of network disturbances or outages. This study investigates the use of advanced genetic algorithms as a practical methodology for the optimal siting of batteries in modern distribution networks. By incorporating historical data on demand and network failures, the algorithm generates statistical models that inform the optimization process. The model integrates both the technical and economic aspects of battery systems to identify locations that minimize reliability indices such as SAIDI and SAIFI, while also reducing investment costs. Tested on a real distribution system comprising 1837 nodes, the proposed approach demonstrates the ability of genetic optimization to deliver efficient solutions compared with traditional methods, providing a high likelihood of identifying strategic battery locations that respond to variable demand, system failures, and technical constraints.
Dissolved gas chromatography (DGA) analysis and Megger DC insulation tests are performed to diagnose the condition of a power transformer. In this context, the objective of this study is to assess whether there is a relationship between both types of tests. To this effect, a database with DGA and Megger DC test protocols is analyzed, using the theory of variable correlation as well as current DGA techniques. The results allow stating that there is indeed a relationship between both tests under specific conditions. Therefore, the DC isolation status of a transformer can be estimated via DGA tests.
In this research, a modification of the particle swarm optimization method was proposed, which is based on an intervention technique for the design of antennas. The proposed method was tested through its application on typical benchmark functions. Moreover, it was applied to the optimal design of a narrow band antenna and two ultra-wideband antennas through shape and hybrid optimization. The objective function was based on minimizing the S11 magnitude in the desired frequency range to improve impedance matching. The modified method had a better performance than the original particle swarm optimization in the typical benchmark functions, and the best results were obtained for the antenna optimization process. Therefore, this method is a good alternative to be applied in these processes because it allows obtaining a better quality of solution and reducing the number of evaluations of the objective function.
This paper presents a formulation of the optimal power flow (OPF) problem for transmission systems with the presence of wind generation. The OPF was modeled as a robust design optimization problem focused on ensuring that the probability that the power balance constraints are observed within a tolerance and that the boundaries of the design variables is respected. A simulation tool was developed implementing the model and it was tested with two power systems, two different wind probability distributions and two solution methods. Test results validated the proposed formulation giving optimal solutions that satisfy the power balance probabilistic constraint with the selected tolerance. The formulation is a simple model to calculate an optimal operation point for a power system with wind farms. This is an initial work in the formulation of an optimal operation model for large scale hybrid power system.
This article is focused on the estimation of reliability indices (SAIDI, SAIFI and ENS) in distribution systems with microgrids containing PV generation and energy storage implementing a Latin hypercube sampling (LHS) method integrated with a Monte Carlo (MC) simulation to combine the failure model of the distribution system and the failure model of the microgrid. The methodology takes into consideration the behavior of the PV generation and the energy storage inside the microgrid and its contributions to improving the reliability indices. The results indicate that the LHS method reduces the amount of replications needed for the convergence of the MC simulation, particularly when a low tolerance in the average value of the ENS is needed.