Conference Agenda

Abstract—Distribution utilities need high-resolution demand modeling and voltage-condition monitoring. In practice, traditional state estimation requires accurate topology and line parameters that are not always available. This study presents a hybrid statistical-machine learning workflow for the CDA– L273 distribution feeder in Tegucigalpa, Honduras. From 96,486 15-minute records, which resulted in 72,548 validated activepower measurements (Jan 2023-Dec 2025), with only 0.05% outliers. The feeder exhibits stable operation, with a mean demand 7,082 W (peak 11,485 W), voltage stability of 1%, and a consistently power factor 0.96. Time-series analysis reveals strong short-term autocorrelation (r=0.90 at one 15-minute step) and repeatable daily/seasonal patterns. For demand prediction, a multilayer perceptron trained on 23 engineered features (lags, rolling statistics, and cyclical encodings) achieved R^2 = 1.0000, RMSE of 6–8 W, and MAE of 5–6 W (< 0.1% of the mean load). In parallel, an MLP-based model-free mapping from P,Q,PF to phase voltages supports lightweight monitoring without network models, with typical test errors of 0.2–0.6%. Overall, the proposed framework provides a practical, low-cost basis for planning and operational decision-making in data-sparse feeders.

Abstract—Integrating variable renewable energy (VRE) gen-
eration into power systems presents significant challenges due to
its inherent variability and uncertainty. Traditionally, the impact
of this generation is assessed using power flow analysis tools.
However, this method requires modeling the characteristics of the
electrical grid, and its solution can entail excessive computational
effort for real-time applications. In this regard, methods based
on artificial intelligence have emerged as an alternative for
operating electrical power systems. Therefore, this paper proposes
a strategy based on a multilayer perceptron (MLP) to assess the
impact of VRE generation, such as wind and solar, through a
purely data-driven approach. To evaluate the proposed strategy,
numerous operating scenarios were simulated on the IEEE 30-bus
test system using a power flow tool, and the resulting solutions
were used to train the MLP. Once the MLP was trained, the
system voltage profiles could be estimated with a mean squared
error (MSE) of approximately 1.47 ×10^-5 and a coefficient of
determination (R-squared) of 0.987 on the test set. These findings
provide a scalable and adaptable tool for system operators to
effectively manage the increasing penetration of VRE generation.

Low-enthalpy systems for building climate conditioning, such as Canadian wells based on ground–air heat exchange, offer a sustainable alternative for improving indoor thermal comfort by exploiting the thermal inertia of the subsoil. However, their performance strongly depends on the control strategy used to manage air injection and extraction under variable temperature and humidity conditions. This work presents the implementation of an intelligent control system for a low-enthalpy Canadian well installed in an office environment. The proposed solution is based on a small-scale programmable logic controller (PLC) integrated with industrial temperature and humidity sensors, enabling real-time monitoring and adaptive process control. Sensor data are visualized through a web-based interface, while the PLC executes a control algorithm designed to regulate ventilation according to thermal comfort requirements during occupied periods. Baseline measurements of thermal comfort and electrical energy consumption were first obtained. Subsequently, improvements were introduced, including upgraded ventilation equipment, higher-precision sensors, surface insulation measures, and refined control parameters. System performance was evaluated through a comparative analysis of temperature, relative humidity, and electrical energy consumption before and after automation. The results show improved stability of indoor temperature and humidity conditions, at the expense of increased electrical energy consumption. These findings highlight the trade-off between thermal comfort and energy use and demonstrate the relevance of intelligent control strategies to achieve controlled and predictable operation of low-enthalpy systems in office environments.

This research presents the development and evaluation of a smart electrical meter designed to measure, analyze, and remotely monitor key power-quality parameters using low-cost sensors and an ESP32 microcontroller. The system captures voltage and current through SCT-013-100 and ZMPT101B sensors, processes the data locally, and transmits it via Wi-Fi to a cloud platform, where users can visualize measurements in real time through a mobile application. Machine-learning techniques, including Empirical Mode Decomposition (EMD), Kernel PCA, and Support Vector Machines (SVM), were implemented to extract features and classify electrical consumption patterns. Experimental validation showed high accuracy in voltage readings, with errors near ±1.10% for low loads and around 0.5% compared to a professional CW500 meter. Current measurements remained acceptable, despite occasional deviations linked to synchronization differences. The system also accurately registered frequency stability at 60 Hz and enabled harmonic-distortion and active-power analysis. Overall, the smart meter demonstrated reliable performance for real-time monitoring and represents a scalable, low-cost solution for energy-quality assessment

Assistive technologies for visually impaired people (VIP) face critical wireless connectivity challenges in both urban and rural environments. This study presents the technological evolution of systems previously developed by the research group, transitioning from a radio frequency architecture within the ISM band to an architecture based on Bluetooth 5.0 technology. Unlike solutions dependent on network infrastructure, the proposed system relies on a decentralized four-module architecture (MASTER, BUS, STOP, and ETA), optimized for the autonomous navigation of VIP in public transport. The core of the research focuses on an experimental characterization of the propagation channel by monitoring and analyzing the stability of the Received Signal Strength Indicator (RSSI) and latency through Round-Trip Time (RTT). The results demonstrate that the Bluetooth 5.0 implementation, enhanced by the use of high-density integration microcontrollers, not only significantly reduces signal variance but also resulted in a reduction of more than 60% in the area of the implemented electronic boards. The integration of antennas into the PCB constitutes a critical aspect of the system design, as it influences cost reduction and increases robustness against multipath interference. The new system enables more precise localization and superior scalability, contributing to the improvement of inclusive mobility for VIP in limited connectivity environments.

Reliable backup power systems are essential for critical service facilities, particularly in multi-line cremation centers where electrical interruptions may compromise operational safety, environmental compliance, and service continuity. This study proposes an engineering-oriented decision framework based on the Analytic Hierarchy Process (AHP) for the optimal selection of backup power generators operating under 50 Hz electrical systems. The methodology integrates electrical load assessment, expert judgment, lifecycle cost analysis, and environmental considerations within a multi-criteria evaluation model. A real case study conducted at a cremation facility in Lima, Peru, was used to validate the proposed approach. Seven evaluation criteria and four commercial generator alternatives were analyzed. Results indicate that power capacity and voltage/frequency stability represent the most influential factors, accounting for more than 46% of the total decision weight. The selected alternative achieved a global priority score of 0.392 and satisfied recommended safety margins for emergency load coverage and transient performance. Sensitivity analysis and cross-validation using the TOPSIS method confirmed the robustness and consistency of the obtained ranking. The proposed framework provides a replicable and practical tool for improving backup power system design and investment decision-making in critical electrical infrastructures.