Monitoramento do fenômeno de queima no processo de retificação de aços por meio de sinais de vibração e análise de tempo e frequência
Carregando...
Data
2018-01-22
Autores
Alexandre, Felipe Aparecido
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Os processos de usinagem são de extrema importância para a fabricação de peças mecânicas, muitas das quais exigem alta precisão dimensional e geométrica. A retificação, dentro dos pro-cessos de usinagem, é um processo de acabamento amplamente utilizado na indústria sendo uma das últimas fases na cadeia de usinagem. Devido à sua posição na cadeia de usinagem, é essencial que o processo de retificação seja controlado, visando a detecção e até mesmo a pre-venção de falhas, pois a peça neste estágio possui alto valor agregado. O fenômeno de queima é uma das piores falhas que pode ocorrer na retificação de aços. Desta forma, um sistema de monitoramento capaz de identificar a ocorrência deste fenômeno, além de detectar o momento do início do mesmo, é de grande importância. Nesse contexto, esse trabalho propõe um método para detecção do fenômeno de queima e o momento de sua ocorrência no processo de retifica-ção de aços ABNT 1045 e VC131 com rebolo de óxido de alumínio. Para tal, um acelerômetro foi fixado no suporte das peças e um sensor de potência foi instalado no sistema. Ambos sinais foram amostrados por um osciloscópio a uma taxa de 2 MHz. Técnicas de processamento digital de sinais, tais como a transformada rápida de Fourier (Fast Fourier Transform - FFT), trans-formada de Fourier de tempo-curto (Short Time Fourier Transform -STFT), densidade espectral de potência – (Power Density Spectral - PSD) e índices de danos (RMSD e CCDM) foram aplicadas. Para melhor fundamentar o estudo, testes de dureza, inspeção visual e metalografia foram executados. Os resultados mostram que o método é confiável no que tange a identificação do fenômeno de queima, ou seja, se houve ou não queima na superfície da peça. Em relação a identificação do momento que se inicia a queima na superfície da peça, o método apresenta melhores resultados para o aço ABNT 1045. Assim, estudos complementares devem ser efetu-ados em trabalhos futuros para que o método possa ser estendido para outros materiais com maior acurácia.
The machining processes are of extreme importance for the manufacture of mechanical parts, many of which require high dimensional and geometric precision. Grinding, within the machin-ing processes, is a finishing process widely used in the industry and it is generally the last stage of manufacture chain. Thus, it is essential to control the grinding process, aiming at the detec-tion and even the prevention of failures, since the part in this stage has high added value. The grinding burn is one of the worst failures that can occur in the grinding of steels. In this way, a monitoring system capable of identifying the occurrence of the workpiece burn in addition to detect the onset moment is of great importance. This work proposes a method to detect the workpiece burn and the moment at which it has occurred in grinding of steels SAE 1045 and VC131 with aluminum oxide grinding wheel. An accelerometer was attached to the parts holder and a power sensor was installed in the system. Both signals were acquired by an oscilloscope at a rate of 2 MHz. Digital signal processing techniques such as Fast Fourier Transform (FFT), Short Time Fourier Transform (STFT), Power Density Spectral (PSD) and damage indices (RMSD – root mean square deviation and CCDM – correlation coefficient deviation metric) were applied. To better support the study, hardness tests, visual inspection and metallography were performed. The results show that the method is reliable to identify the burning phenome-non, that is, whether there was grinding burn on the workpiece surface. Regarding the identifi-cation of the grinding burn onset, the method presented better results for the ABNT 1045 steel. Thus, complementary studies should be carried out in future works so that the method can be extended to other materials with greater accuracy.
The machining processes are of extreme importance for the manufacture of mechanical parts, many of which require high dimensional and geometric precision. Grinding, within the machin-ing processes, is a finishing process widely used in the industry and it is generally the last stage of manufacture chain. Thus, it is essential to control the grinding process, aiming at the detec-tion and even the prevention of failures, since the part in this stage has high added value. The grinding burn is one of the worst failures that can occur in the grinding of steels. In this way, a monitoring system capable of identifying the occurrence of the workpiece burn in addition to detect the onset moment is of great importance. This work proposes a method to detect the workpiece burn and the moment at which it has occurred in grinding of steels SAE 1045 and VC131 with aluminum oxide grinding wheel. An accelerometer was attached to the parts holder and a power sensor was installed in the system. Both signals were acquired by an oscilloscope at a rate of 2 MHz. Digital signal processing techniques such as Fast Fourier Transform (FFT), Short Time Fourier Transform (STFT), Power Density Spectral (PSD) and damage indices (RMSD – root mean square deviation and CCDM – correlation coefficient deviation metric) were applied. To better support the study, hardness tests, visual inspection and metallography were performed. The results show that the method is reliable to identify the burning phenome-non, that is, whether there was grinding burn on the workpiece surface. Regarding the identifi-cation of the grinding burn onset, the method presented better results for the ABNT 1045 steel. Thus, complementary studies should be carried out in future works so that the method can be extended to other materials with greater accuracy.
Descrição
Palavras-chave
Retificação, Monitoramento, Danos térmicos, Vibração, Dureza vickers, Metalografia, Grinding, Monitoring, Thermal damage, Vibration, Vckers hardness, Metallography