1 November 2020 in Solar Energy
CRCarlos D. Rodríguez-Gallegos
Carlos D. Rodríguez-Gallegos
27 December 2017 in IEEE Transactions on Industry Applications
CRCarlos D. Rodriguez-Gallegos
Carlos D. Rodriguez-Gallegos
+1
This paper proposes a new method to determine the sizing and siting of diesel generators (DGs), photovoltaic (PV) solar panels, and batteries for off-grid systems. In this work, the objective is to reduce the total system cost while fulfilling the load demand and maintaining the power quality of the system, among other constraints. This is achieved by employing a two-stage particle swarm optimization algorithm; the first for siting and sizing of the system and the second for its scheduling. Three case studies, namely only DGs, DGs + PV, and DGs + PV + batteries systems, based on the energy requirement of an Indonesian island are presented. In addition, real solar irradiance and temperature data are utilized together with a measured load profile. The results reveal that a system that relies only on DGs is not able to keep the system voltage within the tolerable range (from 0.95 to 1.05 p.u.), while yielding a lifetime cost of 104 million USD. The DGs + PV system requires a lower lifetime cost, 67 million USD. Nevertheless, the voltage of the system reaches values lower than 0.95 p.u. for certain nodes during night-time as the PV are not able to provide power. Finally, the DGs + PV + batteries system yields the lowest cost, 64 million USD, and is able to keep the values of the voltage within the desired range, as batteries can provide power when PV are not operating.
1 January 2018 in Applied Energy
CRCarlos D. Rodríguez-Gallegos
Carlos D. Rodríguez-Gallegos
19 May 2020 in IEEE Transactions on Smart Grid
One of the most groundbreaking structures in deep learning is ResNet, which utilizes skip connections to make neural networks (NNs) significantly deeper. While there are many successful applications of skip connections in computer vision (CV) and natural language processing (NLP), the applications of skip connections in electric load forecasting are still quite limited. Moreover, as compared to the deep NNs used in CV and NLP, most NNs used in load forecasting are relatively shallow NNs, whose performance is partially restricted by the relatively shallow structures. To improve probabilistic forecasting by deepening NN, we investigate the applicability of the skip connections in probabilistic load forecasting NN by proposing new structures and studying the philosophy behind the forecasting improvements brought by skip connections. Case studies show that by deepening the NNs, the proposed structures can generate more accurate probabilistic load forecasts than state-of-the-art methods, which implies that the ideas of skip connections have significant value in probabilistic load forecasting. It is also validated in observations and mathematical proofs that the proposed structures improve the probabilistic forecasting by alleviating the gradient vanishing and exploding problems in deep NNs.
21 July 2021 in IEEE Transactions on Sustainable Energy
Although the integration of solar photovoltaic (PV) systems is gaining widespread acceptance, the intermittency and instability of PV power generation lead to several operational challenges. PV power generation can be impacted by multiple environmental factors, such as the soiling of solar PV panels. There are some conventional methods proposed to deterministically estimate the solar power loss caused by soiling. However, the error of deterministic estimation cannot be eliminated due to the inherent volatility of solar power. Therefore, this paper proposes a probabilistic quantification method, namely SolarQRNN, to estimate the power loss by leveraging images captured by surveillance cameras. Specifically, the proposed model employs a novel quantile loss function and deep learning structures (backbone networks based on residual convolution units), which combines quantile regression and computer vision models for the first time. The proposed method has been extensively tested on a solar panel soiling image dataset. Test results indicate that SolarQRNN outperforms benchmark classification models with at least 51% improvements in evaluating metrics.
24 April 2018 in IEEE Transactions on Power Systems
This work proposes an analytical approach for reactive power dispatch and energy arbitrage in grid-connected distribution systems with distributed energy resources. The comprehensive approach minimizes the operational cost of the system while incorporating the price of electricity, cost of reactive power, reactive power charge, line losses, and energy arbitrage. Mathematical programming and metaheuristic algorithms have been employed as benchmarks, and the efficacy of the analytical approach was validated on 69 and 119-bus radial distribution systems with real solar irradiance and electricity price data. The proposed approach removes the dependence on initial solutions and parameter tuning, shows superior results compared to the benchmark algorithms, and is up to 100 times faster compared to existing algorithms commonly used. It is therefore suitable for both day-ahead and real time power dispatch applications.