Sistema imunológico artificial aplicado no reconhecimento de falhas de uma placa de material compósito
Carregando...
Data
2023-02-24
Autores
Almeida, Estevão Fuzaro de
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A sociedade moderna depende fortemente de sistemas estruturais e mecânicos e, nesse contexto, o uso de materiais compósitos tem se expandido significativamente nas últimas décadas. Eles têm sido impulsionados pelas indústrias aeroespacial, automotiva, agrícola, de consumo e de energia renovável devido às suas propriedades superiores em relação aos materiais metálicos, como alta resistência e rigidez combinadas com baixa densidade e excelente resistência à corrosão. Apesar das propriedades enumeradas, os compósitos geralmente apresentam estruturas internas complexas e anisotrópicas que acabam por levar a vários tipos de danos, dificultando a identificação e previsão de potenciais falhas estruturais. Além disso, vários desses sistemas estão se aproximando do fim de sua disponibilidade de serviço e, considerando que a sua substituição é economicamente inviável, deseja-se continuar utilizando-os com segurança mesmo quando sua operação for estendida. Em todas essas situações apresentadas, o surgimento de danos deve ser detectado com antecedência, a fim de evitar falhas que possam ter impacto econômico significativo e colocar em risco a segurança da vida humana. Neste contexto, o Monitoramento da Integridade Estrutural (SHM) refere-se ao processo de detecção de danos às estruturas de engenharia. Um dos métodos propostos para SHM são os chamados Sistemas Imunológicos Artificiais (SIA), que simulam o sistema imunológico humano, e cujo princípio básico de funcionamento se baseia no processo de diferenciação das células endógenas para as células exógenas, tomando ações que visam a destruição dos possíveis invasores. O SIA é uma área de pesquisa que combina os campos da imunologia, ciência da computação e engenharia para resolver problemas computacionais complexos. Este trabalho visa, portanto, a implementação de uma abordagem SHM completa para detecção e localização de danos em uma placa de material compósito submetida a mudanças de temperatura e crescimento gradual de dano com base em dados de voltagem extraídos de transdutores PZT. Através de seis análises diferentes, realizando-se combinações entre tipos diferentes de leitura e processamento de sinais, alcançou-se uma metodologia otimizada para esta aplicação envolvendo tanto a análise dos sinais no domínio do tempo quanto no domínio da frequência via método de Welch que apresenta um F1-score igual a 1, 100% de acurácia e 100% de probabilidade de detecção de danos e que inclusive é capaz de localizar o caminho no qual o dano está inserido. Alcançou-se, assim, uma aplicação supervisionada que pode ser aplicada no monitoramento contínuo de estruturas de material compósito sofrendo variação de temperatura devido a sua velocidade de processamento média de 2 segundos e que possui alta capacidade de memória.
Modern society is strongly reliant on structural and mechanical systems, and the usage of composite materials in this context has grown dramatically in recent decades. Because of their better qualities over metallic materials, including as high strength and stiffness paired with low density and great corrosion resistance, they have been driven by the aerospace, automotive, agricultural, civil, and renewable energy industries. Regardless of the mentioned qualities, composites often have complex and anisotropic internal structures that eventually lead to various types of damage, making possible structural failures difficult to diagnose and anticipate. Furthermore, several of these systems are nearing the end of their service availability, and because replacement is monetarily prohibitive, one would like to continue using them securely even if their operation is extended. Damage must be discovered early in all of these instances in order to avoid failures that can have a large economic impact and endanger human life safety. In this context, Structural Health Monitoring (SHM) refers to the process of identifying damage to engineering structures. One of the proposed solutions for SHM are so-called Artificial Immune Systems (AIS), which replicate the human immune system and whose primary operating principle is based on the differentiation process between endogenous and exogenous cells, executing activities directed at the death of prospective invaders. AIS is a field of study that integrates immunology, computer science, and engineering to address complicated computational problems. As a result, the goal of this work is to implement a complete SHM approach for damage identification and localization in a composite material plate subjected to temperature variations and progressive damage growth based on voltage data extracted from PZT transducers. An optimized methodology for this application involving both signal analysis in the time domain and in the frequency domain via Welch's method was achieved through six different analyses, performing combinations of different types of readout and signal processing, and that presents an F1-score equal to 1, 100% accuracy, and 100% probability of damage detection, and that is even capable of locating the path in which the damage is inserted. As a result of its average processing speed of 2 seconds and substantial memory capacity, a supervised application that can be applied to the continuous monitoring of composite structures suffering temperature changes was created.
Modern society is strongly reliant on structural and mechanical systems, and the usage of composite materials in this context has grown dramatically in recent decades. Because of their better qualities over metallic materials, including as high strength and stiffness paired with low density and great corrosion resistance, they have been driven by the aerospace, automotive, agricultural, civil, and renewable energy industries. Regardless of the mentioned qualities, composites often have complex and anisotropic internal structures that eventually lead to various types of damage, making possible structural failures difficult to diagnose and anticipate. Furthermore, several of these systems are nearing the end of their service availability, and because replacement is monetarily prohibitive, one would like to continue using them securely even if their operation is extended. Damage must be discovered early in all of these instances in order to avoid failures that can have a large economic impact and endanger human life safety. In this context, Structural Health Monitoring (SHM) refers to the process of identifying damage to engineering structures. One of the proposed solutions for SHM are so-called Artificial Immune Systems (AIS), which replicate the human immune system and whose primary operating principle is based on the differentiation process between endogenous and exogenous cells, executing activities directed at the death of prospective invaders. AIS is a field of study that integrates immunology, computer science, and engineering to address complicated computational problems. As a result, the goal of this work is to implement a complete SHM approach for damage identification and localization in a composite material plate subjected to temperature variations and progressive damage growth based on voltage data extracted from PZT transducers. An optimized methodology for this application involving both signal analysis in the time domain and in the frequency domain via Welch's method was achieved through six different analyses, performing combinations of different types of readout and signal processing, and that presents an F1-score equal to 1, 100% accuracy, and 100% probability of damage detection, and that is even capable of locating the path in which the damage is inserted. As a result of its average processing speed of 2 seconds and substantial memory capacity, a supervised application that can be applied to the continuous monitoring of composite structures suffering temperature changes was created.
Descrição
Palavras-chave
Monitoramento da integridade estrutural, Sistema imunológico artificial, Algoritmo de seleção negativa, Materiais compósitos, Structural health monitoring, Artificial immune system, Negative selection algorithm, Composite materials