Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 7
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Forecasting Univariate Solar Irradiance using Machine learning models: A case study of two Andean Cities

  • Díaz-Bedoya, Daniel
  • González-Rodríguez, Mario
  • Clairand-Gómez, Jean-Michel
  • Serrano-Guerrero, Johnny Xavier
  • Escrivá-Escrivá, Guillermo|||0000-0002-3202-4571
[EN] The integration of solar energy into power systems is essential for the future sustainability of power systems, particularly for isolated systems, such as microgrids, where establishing a primary transmission network is difficult. Therefore, the development of prediction methods becomes crucial to enable accurate forecasting of solar energy generation, facilitating efficient planning and operation of these systems and ensuring their long-term viability. This study proposes distinct forecasting models for solar irradiance forecasting: an autoregressive (AR) model, a Random Forest model, and a Long Short-Term Memory (LSTM) neural network. The methodology involves preprocessing the historical solar irradiance data and performing feature engineering to extract relevant input features. The architectural design, hyperparameter tuning, and training procedures of each model are discussed in detail. The findings indicate that the LSTM model exhibits enhanced performance compared to the AR model, while maintaining similar predictive accuracy to the Random Forest model in forecasting global solar irradiance. Both models yield a mean absolute percentage error of roughly 25%, with the LSTM exhibiting the lower error rate. Moreover, the LSTM model showcases an advancement over the AR model, resulting in a reduction of approximately 10 W/m2 for both root mean square error and mean absolute error. This finding highlights the effectiveness of LSTM networks in capturing long-term dependencies for accurate solar irradiance forecasting. Furthermore, an analysis of the models' interpretability is conducted, offering valuable insights into the key factors that contribute to the shaping of solar irradiance patterns. These insights hold practical significance for the optimization of renewable energy systems., This work has been funded by Universidad de las Americas, Ecuador, projects SIS.MGR.23.13.01 and IEA.JCG.20.01.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

A new interval prediction methodology for short-term electric load forecasting based on pattern recognition

  • Serrano-Guerrero, Xavier
  • Briceño-León, Marco
  • Clairand, Jean-Michel
  • Escrivá-Escrivá, Guillermo|||0000-0002-3202-4571
[EN] Demand prediction has been playing an increasingly important role for electricity management, and is fundamental to the corresponding decision-making. Due to the high variability of the increasing electrical load, and of the new renewable energy technologies, power systems are facing technical challenges. Thus, short-term forecasting has crucial utility for generating dispatching commands, managing the spot market, and detecting anomalies. The techniques associated with machine learning are those currently preferred by researchers for making predictions. However, there are concerns regarding limiting the uncertainty of the obtained results. In this work, a statistical methodology with a simple implementation is presented for obtaining a prediction interval with a time horizon of seven days (15-min time steps), thereby limiting the uncertainty. The methodology is based on pattern recognition and inferential statistics. The predictions made differ from those from a classical approach which predicts point values by trying to minimize the error. In this study, 96 intervals of absorbed active power are predicted for each day, one for every 15 min, along with a previously defined probability associated with the real values being within each obtained interval. To validate the effectiveness of the predictions, the results are compared with those from techniques with the best recent results, such as artificial neural network (ANN) long short-term memory (LSTM) models. A case study in Ecuador is analyzed, resulting in a prediction interval coverage probability (PICP) of 81.1% and prediction interval normalized average width (PINAW) of 10.13%, with a confidence interval of 80%.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Optimal siting and sizing of electric taxi charging stations considering transportation and power system requirements

  • Clairand, Jean-Michel
  • González-Rodríguez, Mario
  • Kumar, Rajesh
  • Vyas, Shashank
  • Escrivá-Escrivá, Guillermo|||0000-0002-3202-4571
[EN] Electric vehicles (EVs) have become more popular to address transportation-related environmental concerns. However, to integrate a massive fleet of EVs, it is crucial to properly build charging stations by considering the optimal geographical placement and number of charging spots. This task is particularly challenging for users with rigid schedules such as taxi drivers. Hence, an optimal siting and sizing approach for an electric taxi (ET) charging station is proposed in this study, considering both transportation and power system needs. In addition, particular attention to taxi drivers' needs is considered. Fixed installation costs, land costs, and trip costs are the factors evaluated in this proposed approach. A network modeling approach based on a winner-takes-all edge trimming was used to identify interest points of the city in terms of traffic flows. Ecuador's capital, Quito, was considered a case study. A sensitivity analysis was also carried out to address traffic flow uncertainties such as trip expenses and restrictions., This study belongs to the projects SIS.JCG.19.03 and IEA.JCG.20.01 from Universidad de las Americas - Ecuador. The authors would like to express their gratitude to Irvin Cedeno of BYD ~ E-Motors Ecuador for helpful information.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

A charging station planning model considering electric bus aggregators

  • Clairand, Jean-Michel
  • González-Rodríguez, Mario
  • Cerdeño, Irvin
  • Escrivá-Escrivá, Guillermo|||0000-0002-3202-4571
[EN] Many governments are pushing for cleaner transportation. In particular, public transportation allows massive transportation of passengers, but it remains highly polluting, especially in high elevation cities. Thus, the progressive introduction of electric buses (EBS) will allow mitigating these environmental concerns. However, some technological problems must be addressed considering the massive penetration of EBs. The lack of flexibility and the time connection of scheduling for public transit make EBS harder than internal combustion ones. This work studies the impact of charging EBs at the bus station and suggests a new way to take EB aggregators into account to reduce energy costs while fulfilling grid restrictions. In addition, to find a different number of charging spots to be installed, a scheduling analysis is conducted., This works belongs to the projects IEA.JCG.20.01 and SIS.MGR.21.01 from Universidad de las Americas - Ecuador. The authors would like to thank Paulo Guerra-Teran from Universidad de las Americas - Ecuador for the fruitful discussion.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

PSO Tuning of a Second Order Sliding Mode Controller to Adjust Active Standard Power Levels on a Single-Phase Voltage Source Inverter

  • Gonzales-Zurita, Oscar
  • Clairand, Jean-Michel
  • Escrivá-Escrivá, Guillermo|||0000-0002-3202-4571
[EN] Electricity management in non-conventional energy systems requires advanced control algorithms to produce and distribute electricity efficiently. Various countries worldwide have promoted policies to manage electricity from power inverters under typical voltage, current, and power values. This paper proposes a DQ control for active power regulation on a single-phase voltage source inverter (SPVSI) using a Second Order Sliding Mode Control (SMC-2). The SMC-2 tunning is performed by a metaheuristics algorithm like the Particle Swarm Optimization (PSO) to adjust the SMC-2 parameters. PSO appropriately tuns the SMC-2 through MATLAB, where the results showed lower Integral of Absolute Error value (IAE) and Integral of Square Error value (ISE) as performance indexes., This paper belongs to the project IEA.JCG.20.01 from Universidad de las Americas-Ecuador
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

A new sliding mode control strategy to improve active power management in a laboratory scale microgrid

  • Gonzales-Zurita, Oscar
  • Clairand, Jean-Michel
  • Escrivá-Escrivá, Guillermo|||0000-0002-3202-4571
[EN] In recent years, the widespread use of single-phase voltage source inverters (SPVSIs) has led to significant challenges in managing microgrids' electrical energy, particularly regarding efficiency and robustness. Most SPVSIs on the market use proportional¿integral (PI) control for their closed-loop regulation process. Although the PI controller represents an effective method for power management in a microgrid, it exhibits challenges in achieving proper system response and stability for active power generation. This study proposes a robust control solution based on the second-order sliding mode control (SMC-2) algorithm to overcome the mentioned challenges. This algorithm employed a non-conventional sliding surface to improve the microgrid's capacities for energy management. The research was conducted in a PSCAD environment, where a real industrial SPVSI with PI control was emulated and then optimized using the SMC-2. The employed tuning method was the particle swarm optimization (PSO) technique. This method ensures global optimum optimization of the SMC-2 control parameters. Different operating scenarios were performed to demonstrate the superiority of SMC-2 over PI. The SMC-2 achieved in less time than PI, active power references over a range of 0 to 1000 W, and obtained lower performance indices based on reference tracking error. The microgrid under study is located at Universitat Politècnica de València., This paper is part of the project IEA.JCG.20.01, which belongs to Universidad de las Americas-Ecuador, and is part of the Project TED2021-130464B-I00 (INASOLAR) , funded by MCIN/AEI/10.13039/501100011033 and by the European Union "NextGenerationEU"/PRTR directed by Ph.D. Guillermo Escriva-Escriva.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Enhancing transient response in AC microgrids: A multi-objective optimization approach for improved active power management

  • Gonzales-Zurita, Oscar
  • Gonzalez-Rodriguez, Mario
  • Clairand, Jean-Michel
  • Escrivá-Escrivá, Guillermo|||0000-0002-3202-4571
[EN] Many residential and local consumers have embraced single-phase inverters for self-consumption and energy trading. However, their adoption challenges the efficient management of electrical energy within microgrids (MGs), particularly regarding transient responses like rise time and overshoot, an appropriate active power control, or an optimum performance under different operating conditions. Conventional inverter controllers, while easy to program, often face conflicting objectives, where improving one parameter degrades another. This limitation complicates the control of nonlinear systems, risking high-energy transients that can damage components and reduce the lifespan of power semiconductors, leading to costly maintenance. This study proposes a robust strategy focused on primary control using a higher-order sliding mode controller (SMC) with a PI sliding surface tuned by multi-objective optimization (MOO) methods to address these issues. The control of active power is performed under the DQ frame synchronized to the main grid under a PLL method. Our approach aims to improve both the rise time and the overshoot of active power simultaneously. MOO techniques such as Multi-Objective Genetic Algorithm (MOGA), Multi-Objective Differential Evolution (MODE), and Multi-Objective Adaptive Simulated Annealing (MOASA) have shown significant promise in our PSCAD-based research using data from an industrial inverter in an MG laboratory. The results were compared with particle swarm optimization (PSO) techniques. Performance indices like integral of absolute error (IAE), integral of square error (ISE), integral of time and absolute error (ITAE), and integral of time and square error (ITSE) demonstrated that SMC-2+MOO outperforms traditional methods like PSO, offering a superior solution for managing MG efficiency., The authors express gratitude to the Universidad de Las Americas-Ecuador for the financial support granted for this research via project IEA.JCG.20.01.