This study proposes a novel approach to monitoring the grinding wheel during the dressing operation by using acoustic emission (AE) signals and the statistical Counts method. AE signals were acquired during the dressing passes and processed in MATLAB^®. The Counts matrices were segmented according to the grinding wheel rotation, and the metric termed z-ratio, which combines mean and standard deviation statistics, was calculated for each subwindow. The vectors were then filtered, normalized, and represented in polar coordinates. The results demonstrate the method’s ability to track the evolution of dressing and detect grinding wheel eccentricity, offering a promising tool for the indirect monitoring of the surface conditions of the grinding tool during the dressing operation of conventional grinding wheels.

A dressagem é fundamental para garantir a eficiência dos processos de retificação com rebolos convencionais. Contudo, seu monitoramento depende majoritariamente da experiência do operador, comprometendo a precisão e reprodutibilidade dos resultados. Este trabalho propõe uma abordagem de monitoramento indireto baseada em sinais de emissão acústica (EA) para identificar alterações no estado do rebolo. Os ensaios foram conduzidos em uma retificadora plana sob condições controladas. Um sensor de EA foi fixado ao suporte do dressador de ponta única e os sinais foram digitalizados via sistema de aquisição de alta velocidade. Osdados foram processados mediante as estatísticas RMS (Root Mean Square) e razão-z, sendo posteriormente representados em coordenadas polares e complementados pelo cálculo do RMSE (Root Mean Square Error), utilizado como métrica de desvio em relação à condição ideal de circularidade. Os valores de RMS quantificaram a energia acústica, enquanto a razão-z evidenciou a uniformidade dos sinais. A representação polar e o RMSE sintetizaram a distribuição angular da energia acústica, fornecendo o índice objetivo da uniformidade do rebolo. Nos primeiros passes, os valores de RMS apresentaram maior variabilidade, refletindo a topografia irregular da superfície de corte do rebolo. Com o avanço da operação de dressagem, observou-se maior uniformidade, com curvas polares próximas do círculo unitário e valores decrescentes de RMSE, indicando melhora da condição do rebolo. A metodologia provê uma basequantitativa para a otimização da dressagem, permitindo a interrupção da operação no ponto de máxima uniformidade geométrica da ferramenta, minimizando os custos do processo. Como perspectivas futuras, sugere-se integrar a abordagem a sistemas inteligentes de monitoramento em tempo real.

The study proposes a new methodology based on time-frequency analysis for the indirect monitoring of the dressing operation of conventional grinding wheels. Through a low-cost piezoelectric diaphragm (PZT), acoustic signals are captured during the process. The analysis is based on the coefficient of variation in the Short-Time Fourier Transform (STFT). The results indicate that the signal instability is high in the first passes but progressively decreases, reaching stability between passes 10 and 15. This suggests that the surface of the grinding wheel is regularized and ready for grinding. The methodology can serve as an objective indicator to assist the operator in interrupting the dressing process at the optimal moment, thereby optimizing grinding quality and reducing operational costs.

This study proposes a method to indicate the optimal moment to interrupt grinding wheel dressing using the counts statistics extracted from acoustic emission signals. Experiments were conducted on a surface grinding machine equipped with an aluminum oxide grinding wheel and a single-point diamond dresser. A Multilayer Perceptron neural network, trained via the Scaled Conjugate Gradient algorithm, classified the grinding wheel condition into three states: undressed, intermediate, and dressed. In three external tests not included in the training set, the model achieved an accuracy of 88.13%.

Carbon fiber-reinforced polymers (CFRPs) are increasingly taking the place of metals in lightweight structures and as a result require efficient structural health monitoring (SHM). To achieve this, four PZT sensors (12–27 mm) were bonded to both CFRP and aluminum plates to examine the effects of material and diameter. Time and frequency domain analysis showed that the influence of sensor diameter was non-monotonic, which was statistically confirmed (p < 0.001 for CFRP; p = 0.003 for aluminum). The 20 mm sensor proved to be the most sensitive, showing a peak amplitude of 2.96 V in CFRP, while the 12 mm sensor had a higher signal-to-noise ratio (~34 dB) in both materials. The CFRP concentrated energy below 50 kHz and its decay time was 42–49% lower than that of aluminum, which shows selective attenuation of high frequencies. Even though all the sensors detected the event, the 20 mm sensor is recommended because of its sensitivity and the 12 mm sensor because of its noise robustness.