Conference Agenda

Inaccuracies inherent to the dynamic weighing of mineral concentrates during container filling operations can result in substantial economic losses. This study addresses this problem by developing and implementing an intelligent compensator based on a Long Short-Term Memory (LSTM) recurrent neural network. The proposed system processes real-time sensor data—namely, load cell voltage, conveyor belt speed, and inclination angle—as a multivariate time series to predict and correct weighing errors on-the-fly. Following a systematic hyperparameter optimization, the optimal architecture (20 LSTM units, learning rate of 0.001) reduced the Mean Absolute Percentage Error (MAPE) from 8.5% to 3.01% on the validation dataset, representing a 64% improvement in accuracy. Subsequent validation on an independent test set confirmed the model's robustness and generalizability, yielding a coefficient of determination (R²) of 0.97. A feature importance analysis revealed that the load cell signal is the primary contributor, accounting for 55% of the predictive power, thereby aligning the model's behavior with the underlying physical principles of the process. This research offers two primary contributions: (1) a methodological framework for integrating LSTM networks into industrial weighing systems, and (2) a prototype validated under real-world operating conditions, thereby providing a scalable solution for process optimization in the mining sector.

This work presents an analysis of the behavior of an induction motor under field-oriented control in response to variations in rotor resistance. Different field-oriented control schemes are simulated under varying rotor resistance. The responses of electromagnetic torque, fluxes, currents, and speed to load variations are analyzed. The responses of different control schemes, including decoupling, PID with a synchronous controller, and closed-loop control with PI, are compared under different resistance values. Finally, the results of the responses are analyzed.

The objective of this study was to examine the operational condition of Feeder 5006 at the Juliaca Electrical Substation through the application of infrared thermography as a predictive maintenance tool, with the aim of identifying hot spots in its components and reducing the risk of unplanned failures. The main problem addressed was the lack of a systematic thermal diagnosis that would allow the timely detection of thermal anomalies associated with overloads, defective connections, and equipment deterioration, which affect the reliability and continuity of service. The methodology employed was based on field thermographic inspection, under previously established technical parameters and applying the NETA standard for the classification of thermal faults. Fifty-one medium-voltage structures of the feeder were evaluated, recording ambient, maximum, and reference temperatures, from which the thermal delta and the priority level of each detected anomaly were determined. The results revealed the presence of thermal faults of varying severity, with severe and moderate conditions predominating, as well as the identification of critical points with temperature differences exceeding forty degrees Celsius, which require immediate intervention. Finally, it was demonstrated that infrared thermography is an effective tool for predictive maintenance, as it enables the early identification of anomalous conditions that cannot be detected through conventional visual inspections, contributing to timely technical decision-making and the prioritization of corrective actions. Likewise, it strengthens the operational reliability of the feeder and optimizes institutional maintenance resources.

The growth of the electric automotive sector is undeniable, which is why numerous applications for the control and interface of electric vehicles have been researched and developed, as well as improvements to their performance. This article presents an electric motor control system based on frequency-to-voltage conversion, which offers, among other features, very precise power regulation. The premise of all of the above is based on the use of pulse width modulation (PWM) of a square wave signal, which will be responsible for modifying its duty cycle at the user's request by means of a potentiometer, thus simulating the accelerator pedal of an electric vehicle, varying the operating frequency of the rest of the circuit responsible for controlling the power demand required by the electric motor at any given moment. The final design has been implemented on a board to avoid transfer losses between its components as much as possible and to ensure that everything results in a single solid block that is as compact as possible, so that it can be used in teaching, both for analysis and synthesis. The end result is a high level of control over the motor through PWM signal regulation, achieving maximum values of 12 V and 1.8 A, and a maximum power of 21.6 W.

This paper describes the development of a power conversion system based on a cascaded single-phase multilevel inverter with a common variable DC source for electric traction applications. The proposed architecture uses high-frequency transformers to ensure galvanic isolation and enable power distribution from a single DC source, optimizing energy integration and DC bus variability. A phase-array modulation (PDPWM) technique is implemented to improve output signal quality and significantly reduce harmonic content. The result is a DC bus design with a variability between 45 V and 145 V, controlled by the pulse width of a PWM signal and a 13-level multilevel voltage signal with a total harmonic distortion (THD) of 9.38%.