Browsing by Author "Nuñez Retamal, Felipe Eduardo"

Now showing 1 - 9 of 9
  • Thumbnail Image
    Item
    A multi-cast algorithm for robust average consensus over internet of things environments
    (2019) Oróstica, Boris; Nuñez Retamal, Felipe Eduardo
  • Thumbnail Image
    Item
    An Overview of Microgrids Challenges in the Mining Industry
    (2020) Gómez, Juan S.; Rodriguez, Jose; Garcia, Cristian; Tarisciotti, Luca; Flores-Bahamonde, Freddy; Pereda Torres, Javier Eduardo; Nuñez Retamal, Felipe Eduardo; Cipriano, Aldo; Salas, Juan Carlos
    The transition from fossil fuels to renewable energies as power sources in the heavy industries is one of the main climate change mitigation strategies. The carbon footprint in mining is related to its inherent extraction process, its high demand of electric power and water, and the use of diesel. However, considering its particular power requirements, the integration of microgrids throughout the whole control hierarchy of mining industry is an emergent topic. This paper provides an overview of the opportunities and challenges derived from the synergy between microgrids and the mining industry. Bidirectional and optimal power flow, as well as the integration of power quality have been identified as microgrid features that could potentially enhance mining processes. Recommendations pertaining to the technological transition and the improvement of energy issues in mining environments are also highlighted in this work.
  • Thumbnail Image
    Item
    Blockchain-Driven on-Demand Control Loops Over Iot Environments
    (IEEE, 2019) Langarica Chavira, Saúl Alberto; Nuñez Retamal, Felipe Eduardo
    The feedback control loop is the atomic unit in a control system. Typically, feedback control loops are rigid objects that involve a dynamical system, or plant, which has a set of its output states measured by dedicated sensors, which in turn feed a processing unit, known as controller, that calculates actions to be applied as inputs to the plant, via elements known as actuators, in order to drive the outputs to a desired value or trajectory. The appearance of the Internet of Things (IoT) paradigm, where a large number of sensors and processing units interact over a communication network, offers an underlying infrastructure to operate and configure control loops using a different logic: an on-demand strategy. This work introduces the concept of on-demand control loop, and proposes the use of blockchain technology as the enabling infrastructure for generating on-demand control loops over large-scale IoT environments. General design guidelines are given and a simple implementation example over the Ethereum blockchain is presented, which shows the feasibility of the proposed technique.
  • Thumbnail Image
    Item
    Consensus-Based Distributed Control of a Multilevel Battery Energy Storage System
    (IEEE, 2020) Neira Castillo, Sebastian Felipe; Poblete Durruty, Pablo Martín; Pereda Torres, Javier Eduardo; Nuñez Retamal, Felipe Eduardo
    Battery Energy Storage Systems (BESS) based on modular multilevel cascaded topologies allow splitting the battery array into the converter sub-modules, improving efficiency and reliability of the solution. Modular converters can perform active balance of the battery packs and regulation of the output power through the use of multi-objective controllers commanding the power of each sub-module accordingly. However, despite the hardware modularity of multilevel BESS, the controller is usually based on centralized designs, which present issues in terms of computational load and scalability. This paper proposes a two-layer distributed control scheme, based on a high-level consensus algorithm to perform the State of Charge (SoC) balance and low-level individual multi-variable controllers to regulate the operation of the converter sub-modules. The controller was implemented in a cascaded topology, where each sub-module determines and regulates its output power considering the information of its neighbours. The proposed solution allows to fully exploit the modularity of the converter, distributing the control units in the power modules to improve the overall flexibility and scalability of the system. Simulations results validate the operation of the proposed system, showing how the converter sub-modules distribute its power outputs to equalize the SoC levels. Furthermore, the distribution of the multi-variable controllers reduces the computational burden, as each unit just regulates the operation of the respective sub-module.
  • No Thumbnail Available
    Item
    Distributed Current Control of Cascaded Multilevel Inverters
    (IEEE, 2019) Poblete Durruty, Pablo Martín; Pereda Torres, Javier Eduardo; Nuñez Retamal, Felipe Eduardo; Aguilera, Ricardo P.
    Multilevel inverters are widely used in the medium voltage energy market. Among them, the Cascaded H-Bridge (CHB) and Modular Multilevel Inverter (MMC) stand out for their modular hardware, which allows the easy replacement of their modules in the case of failures. Normally, these converters are operated with centralized control schemes, which require a large processing capacity, multiple digital outputs and wires for the switching signals of the transistors. The use of a centralized controller reduces the modularity of the system and generates difficulties as the number of inverters connected in series (cells) increases. In this work, a distributed control strategy for CHB converters is proposed, allowing each cell to have an independent local controller, which increases the modularity of the multilevel converter without deteriorating the performance of the system. Extensive simulations are carried out, showing the advantages and disadvantages of the proposed distributed control strategy offers when compared to a PI current controller on dq axes and Shifted PWM. The proposed control scheme is also tested under fault conditions and is finally applied in a Field Oriented Control (FOC) of a permanent magnet synchronous machine (PMSM), in order to evaluate its performance in electrical drives applications.
  • Thumbnail Image
    Item
    Expert fault detection and diagnosis for the refrigeration process of a hydraulic power plant
    (IEEE, 2008) Berrios Alvarez, Rodrigo Andres; Nuñez Retamal, Felipe Eduardo; Cipriano, Aldo; Paredes, Rodrigo
    In this paper we discuss the components and design of each part of an expert system for fault detection and diagnosis in the refrigeration process of a hydraulic power station. Given a minimum group of measurements, it is possible to extract behavior patterns for the whole system. The knowledge base is obtained by simulating the dynamic behavior of the process. The expert system was designed for a 580 MW power plant called Central Pehuenche, located in Chile. The system shows a 96% of detection effectiveness under different operation conditions.
  • Thumbnail Image
    Item
    Forecasting copper electrorefining cathode rejection by means of recurrent neural networks with attention mechanism
    (2021) Correa Hucke, Pedro Pablo; Cipriano, Aldo; Nuñez Retamal, Felipe Eduardo; Salas, Juan Carlos; Löbel Díaz, Hans-Albert
    Electrolytic refining is the last step of pyrometallurgical copper production. Here, smelted copper is converted into high-quality cathodes through electrolysis. Cathodes that do not meet the physical quality standards are rejected and further reprocessed or sold at a minimum profit. Prediction of cathodic rejection is therefore of utmost importance to accurately forecast the electrorefining cycle economic production. Several attempts have been made to estimate this process outcomes, mostly based on physical models of the underlying electrochemical reactions. However, they do not stand the complexity of real operations. Data-driven methods, such as deep learning, allow modeling complex non-linear processes by learning representations directly from the data.We study the use of several recurrent neural network models to estimate the cathodic rejection of a cathodic cycle, using a series of operational measurements throughout the process. We provide an ARMAX model as a benchmark. Basic recurrent neural network models are analyzed first: a vanilla RNN and an LSTM model provide an initial approach. These are further composed into an Encoder-Decoder model, that uses an attention mechanism to selectively weight the input steps that provide most information upon inference. This model obtains 5.45% relative error, improving by 81.4% the proposed benchmark. Finally, we study the attention mechanism’s output to distinguish the most relevant electrorefining process steps. We identify the initial state as a critical state in predicting cathodic rejection. This information can be used as an input for decision support systems or control strategies to reduce cathodic rejection and improve electrolytic refining’s profitability.
  • No Thumbnail Available
    Item
    Neuroevolutive Control of Industrial Processes Through Mapping Elites
    (IEEE, 2021) Langarica Chavira, Saúl Alberto; Nuñez Retamal, Felipe Eduardo
    Classical model-based control techniques used in process control applications present a tradeoff between performance and computational load, especially when using complex nonlinear methods. Learning-based techniques that allow the controller to learn policies from data represent an appealing alternative with potential to reduce the computational burden of real-time optimization. This article presents an efficient learning-based neural controller, optimized using evolutionary algorithms, designed especially for maintaining diversity of individuals. The search of solutions is conducted in the parameter space of a population of deep neural networks, which are efficiently encoded with a novel compression algorithm. Evaluation against strong baselines demonstrates that the proposed controller achieves better performance in most of the chosen evaluation metrics. Results suggest that learning-based controllers are a promising option for next-generation process control in the context of Industry 4.0.
  • No Thumbnail Available
    Item
    Simulating Railway and Metropolitan Rail Networks: From Planning to On-line Control
    (IEEE, 2010) Nuñez Retamal, Felipe Eduardo; Reyes Leiva, Francisco; Grube Krebs, Pablo; Cipriano, Aldo
    Train-based systems are the principal means of public transportation in many of the world's cities, and continue to grow in the face of rising demand. Expanding infrastructure is costly, however, and at a certain point becomes unsustainable. When this occurs the only feasible solution is to improve the management system. This is done by using either offline or online intelligent transportation systems which requires prior analysis and testing. These previous activities are not easy to carry out in the transportation system itself because of high costs and possible drawbacks. The usual solution in these cases involves conducting simulations. Simulating a train system is a complex problem for which several software applications have been designed, using different models, programming approaches, and simplifications. Therefore, selecting the best simulator for testing a particular intelligent system is a hard task that needs atention. In this work, the requirements that a simulator must fulfill in order to be suitable for testing a particular system are stated. For each class of application, examples of available simulators are given and their main characteristics are then analyzed. Finally, as a practical example, the problem of evaluating skip-stop policies in a multi-line Metro system is studied using a novel event-driven simulator.