Automatic detection of thermal damage in grinding process by artificial neural network

Esse trabalho tem por objetivo o desenvolvimento de um sistema inteligente para detecção da queima no processo de retificação tangencial plana através da utilização de uma rede neural perceptron multi camadas, treinada para generalizar o processo e, conseqüentemente, obter o limiar de queima. em geral, a ocorrência da queima no processo de retificação pode ser detectada pelos parâmetros DPO e FKS. Porém esses parâmetros não são eficientes nas condições de usinagem usadas nesse trabalho. Os sinais de emissão acústica e potência elétrica do motor de acionamento do rebolo são variáveis de entrada e a variável de saída é a ocorrência da queima. No trabalho experimental, foram empregados um tipo de aço (ABNT 1045 temperado) e um tipo de rebolo denominado TARGA, modelo ART 3TG80.3 NVHB.

Improving minimum quantity lubrication in CBN grinding using compressed air wheel cleaning

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Analysis of the Influence of Sparkout Time on Grinding Using Several Lubrication/Cooling Methods

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Evaluation of the thermal damages in inlet engine valves varying the grinding wheels and the cutting fluids

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Application of the Minimum Quantity Lubrication (MQL) Technique in the Plunge Cylindrical Grinding Operation

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Performance of aerodynamic baffles in cylindrical grinding analyzed on the basis of air layer pressure and speed

Over the years, grinding has been considered one of the most important manufacturing processes. Grinding is a high precision process, and the loss of a single workpiece in this stage of the production is unacceptable, fir the value added to the material is very high due to many processes it has already undergone prior to grinding. This study aims to contribute toward the development of an experimental methodology whereby the pressure and speed of the air layer produced by the high rotation of the grinding wheel is evaluated with and without baffles, i.e., in an optimized grinding operation and in a traditional one. Tests were also carried out with steel samples to check the difference in grinding wheel wear with and without the use of baffles.

Influence of the grinding wheel's dive speed on the damage of rectified hardened steels

The world tendency is the increase of the productivity and the production of pieces more and more sophisticated, with high degree of geometric and dimensional tolerances, with good surface finish and low cost. Rectification is responsible for the final finish in the machining process of a material. However, damages generated in this production phase affect all the resources used in the previous processes. Great part of the problems happennig in the rectification process is due to the enormous temperature generated in this activity because of the machining conditions. The dive speed, which is directly related to the productivity, is considered responsible for the damages that occur during rectification, limiting its values to those that do not cause such damages. In this work, through the variation of the dive speed in the process of cylindrical grinding of type ABNT D6 steel, rationalizing the application of two cutting fluids and using a CBN (cubic boron nitrate) abrasive wheel with vitrified blond, the influence of the dive speed on the surface damages of hardened steels was evaluated. The results allowed to say that the dive speed, associated to an efficient cooling and lubrication, didn't provoke thermal damages (including heated zones, cracks and tension stresses) to the material. Residual stresses and the roughness of rectified materials presented a correlation with the machining conditions. The work concluded that it is possible to increase the productivity without provoking damages in the rectified components.

Evaluation of grinding fluids in the grinding of a martensitic valve steel with CBN and alumina abrasives

In this paper the performances of different cutting fluids and grinding wheel types were analysed in the grinding of SAE HVN-3 workpieces. The resulting residual stress, wheel wear and roughness were evaluated. The influence of the cutting fluid jet velocity v(j) was also analysed. As a conclusion, the lubrication ability seems to be the governing factor in the cutting fluid performance. The use of CBN wheels can significantly reduce the thermal damage in grinding, leading to compressive residual stresses. The CBN wheel and the cutting oil give an optimum combination for performing this grinding operation.

The minimum quantity of lubricant (MQL) technique in the surface grinding of steel

This research aimed to analyze the viability of the minimum quantity of lubricant (MQL) technique towards different methods of lubri-refrigeration in surface grinding of steel, considering process quality, wheel life and the viability of using cutting fluids The proposal methods were the conventional (abundant fluid flow), the minimum quantity lubrication (MQL) and the optimized method with Webster nozzle (rounded) This analysis was carried out in equal machining conditions, through the assessment of variables such as grinding force, surface roughness, G ratio (volume of removed material/volume of wheel wear), and microhardness The results showed the possibility of improvement of the grinding process Besides, there is the opportunity for production of high quality workpieces with lower costs The MQL technique showed efficiency in machining with lower depths of cut The optimized method with Webster nozzle applies the fluid in a rational way, without considerable waste Hence, the results show that industry can rationalize and optimize the application of cutting fluids, avoiding inappropriate disposal, inadequate use and consequently environment pollution

The behaviour of resin-bond diamond wheels in the grinding of advanced ceramics

Results are reported of the behaviour of the plane tangential grinding process using diamond grinding wheels. Grinding performance is analysed in terms of the wear behaviour of the wheel in the grinding of ceramic. Discussion of the results concentrates on the wear mechanism of the diamond wheel and the process of material removal.

Production grinding burn detection using acoustic emission and electric power signals

This work uses a monitoring system based on a PC platform, where the acoustic emission and electric power signals generated during the grinding process are used to investigate superficial burning occurrence in a surface grinding operation using two types of steel, three grinding conditions and an Al203 vitrified grinding wheel. Acoustic emission signals on the workpiece and grinding power were measured during a surface plunge operation until the grinding burn happened. From the results the standard deviation of the acoustic emission signal and the maximum electric power were calculated for each grinding pass. The proposed DPO parameter is the product between the power level and acoustic emission standard deviation. The results show that both signals can be used for burning detection, and the parameter DPO is the best indicator for the burning studied in this work. This can be explained by the high dispersion of the acoustic emission RMS level associated to the high power consumption when the grinding wheel lose its sharpness.

The improvement in grinding of inlet engine valves by the adoption of the most effective cuting fluid type

In this paper is presented an experimental research in which the grinding of seating surfaces of inlet engine valves was improved by the adoption of the most effective cutting fluid type, matching the new requirements of cutting fluid application. Four different types of cutting fluids (straight oil and three different types of soluble oils) were analyzed. As qualitative and quantitative evaluation parameters of the performance of the cutting fluids, the roughness, the grinding wheel wear, the cutting force and the workpiece residual stress were determined. As a conclusion, the straight oil was the cutting fluid that presented the best results in all of the parameters analyzed. Copyright © 2000 Society of Automotive Engineers, Inc.

Neural network detection of grinding burn from acoustic emission

An artificial neural network (ANN) approach is proposed for the detection of workpiece `burn', the undesirable change in metallurgical properties of the material produced by overly aggressive or otherwise inappropriate grinding. The grinding acoustic emission (AE) signals for 52100 bearing steel were collected and digested to extract feature vectors that appear to be suitable for ANN processing. Two feature vectors are represented: one concerning band power, kurtosis and skew; and the other autoregressive (AR) coefficients. The result (burn or no-burn) of the signals was identified on the basis of hardness and profile tests after grinding. The trained neural network works remarkably well for burn detection. Other signal-processing approaches are also discussed, and among them the constant false-alarm rate (CFAR) power law and the mean-value deviance (MVD) prove useful.