SIMULACIÓN DEL CONTROL DE UN SISTEMA DE PRIMER ORDEN EMPLEANDO UN ALGORITMO NEURO INSPIRADO (SIMULATION OF THE CONTROL OF A FIRST ORDER SYSTEM EMPLOYING NEURO-INSPIRED ALGORITHM)
Resumen
Actualmente el desarrollo de sistemas autónomos se ha convertido en un tópico bastante importante, su relevancia continúa creciendo en diversos ámbitos de la sociedad. Sin embargo, la autonomía de los nuevos sistemas se ve limitada por la capacidad de las baterías. En este artículo se propone un control neuronal que utiliza una red neuronal de tipo función base radial. Utilizando el software MATLAB – Simulink, el control neuro inspirado es implementado sobre un sistema dinámico de primer orden con presencia de perturbaciones. Se realiza un análisis comparativo del control propuesto con un controlador PI clásico para los casos de una perturbación constante, perturbación variante en el tiempo y perturbación dependiente del estado. En general, el algoritmo neuro inspirado muestra un comportamiento rápido, logra aproximar la posición deseada en tiempo reducido y mantiene un error mínimo. Esto permite reducir tiempos de simulación, lo que se traduce en menor costo computacional y energético.
Palabras Clave: Control Neuronal, Función Base Radial, PI, Redes Neuronales.
Abstract
Currently, the development of autonomous systems has become a highly significant topic, and its relevance continues to grow in various fields of society. However, the autonomy of these new systems is constrained by battery capacity. This article proposes a neuronal control approach that utilizes a radial basis function neural network. Using MATLAB – Simulink software, the neuro inspired control is implemented on a first-order dynamic system with the presence of disturbances. A comparative analysis is conducted between the proposed control and a classic PI controller in scenarios involving constant disturbances, time-variant disturbances, and state-dependent disturbances. In general, the neuro-inspired algorithm exhibits rapid response, achieves close approximation to the desired position in a short amount of time, and maintains minimal error. This leads to reduced simulation times, translating to lower computational and energy costs.
Keywords: Neural Networks, Neuronal Control, PI, Radial Basis Function.
Texto completo:
459-473 PDFReferencias
Doug, A. Neural networks: history and applications. Nova Science Publishers, Incorporated, 2020.
Galeone, P. Hands-on neural networks with TensorFlow 2.0: understand TensorFlow, from static graph to eager execution, and design neural networks. Packt Publishing Ltd, 2019.
García, C. Redes neuronales de funciones base radiales. ULL. Facultad de Ciencias. 2017.
Gómez, J. et al. Control PI neuro-adaptable en tiempo real de la humedad en el suelo usando un modelo híbrido. Revista Iberoamericana de Automática e Informática industrial. vol. 20, no 1, p. 93-103. 2023.
Jing, Y. et al. Inverted pendulum RBF neural network PID controller design. International Symposium on Computer, Consumer and Control. IEEE, 2014.
Panhale, A. et al. Robust motion control using novel first order sliding modes. 20th International Conference on Control, Automation and Systems (ICCAS). IEEE, 2020.
Pérez, C. et al. Control por modos deslizantes de un sistema de intercambio de calor: validación experimental. Enfoque UTE 9.4: 110-119, 2018.
Regalón, O. et al. Aplicación de algoritmos de control clásico, adaptable y robusto a sistemas dinámicos de parámetros variables. Ingeniería Energética 33.3: 184-195, 2012.
Revanesh, M. et al. Artificial neural networks-based improved Levenberg–Marquardt neural network for energy efficiency and anomaly detection in WSN. Wireless Networks: 1-16, 2023.
Shengnan, L. et al. Energy efficiency and coding of neural network. Frontiers in Neuroscience 16, 2023.
Shuai, L. et al. Neural Networks for Robot Arm Cooperation with a Full Distributed Control Topology. Neural Networks for Cooperative Control of Multiple Robot Arms: 49-74, 2018.
Tian, H. et al. Research on Adaptive Sliding Mode Robust Control Algorithm of Manipulator Based on RBF Neural Network. Chinese Automation Congress (CAC). IEEE, 2020.
Utkin, V. et al. Sliding mode control in electro-mechanical systems. CRC press, 2017.
Wang, H. & Meng, F. Control Method of Nanomaterial Numerical Control Electronic Processing Based on RBF Neural Network. Advances in Materials Science and Engineering, 2022.
Wang, R. et al. Fuzzy neural network PID control based on RBF neural network for variable configuration spacecraft. IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC). IEEE, 2018.
Zhang, X. et al. Research on fixed time sliding mode control strategy based on RBF neural network. Journal of Measurement Science & Instrumentation 14.2, 2023.
URL de la licencia: https://creativecommons.org/licenses/by/3.0/deed.es
Pistas Educativas está bajo la Licencia Creative Commons Atribución 3.0 No portada. 
TECNOLÓGICO NACIONAL DE MÉXICO / INSTITUTO TECNOLÓGICO DE CELAYA
Antonio García Cubas Pte #600 esq. Av. Tecnológico, Celaya, Gto. México
Tel. 461 61 17575 Ext 5450 y 5146
pistaseducativas@itcelaya.edu.mx