Virtual Sensors for On-line Wheel Wear and Part Roughness Measurement in the Grinding Process

Abstract

Grinding is an advanced machining process for the manufacturing of valuable complex and accurate parts for high added value sectors such as aerospace, wind generation, etc. Due to the extremely severe conditions inside grinding machines, critical process variables such as part surface finish or grinding wheel wear cannot be easily and cheaply measured on-line. In this paper a virtual sensor for on-line monitoring of those variables is presented. The sensor is based on the modelling ability of Artificial Neural Networks (ANNs) for stochastic and non-linear processes such as grinding; the selected architecture is the Layer-Recurrent neural network. The sensor makes use of the relation between the variables to be measured and power consumption in the wheel spindle, which can be easily measured. A sensor calibration methodology is presented, and the levels of error that can be expected are discussed. Validation of the new sensor is carried out by comparing the sensor's results with actual measurements carried out in an industrial grinding machine. Results show excellent estimation performance for both wheel wear and surface roughness. In the case of wheel wear, the absolute error is within the range of microns (average value 32 μm). In the case of surface finish, the absolute error is well below Ra 1 μm (average value 0.32 μm). The present approach can be easily generalized to other grinding operations.

A new approach for dynamic modelling of energy consumption in the grinding process using recurrent neural networks

“…Besides, most of the published works provide a solution for a given wheel-workpiece combination. Although there are few cases where solutions for more than one wheel are provided, in few works the solution can be generalized to new wheels that have not been used during the training process of the ANN [11]. This is the main reason why industrial application of ANN is not widely extended in grinding yet.…”

“…Thus, for analysing the influence of the time characteristics and the number of points, three different net configurations (HU-D) have been selected (5HU-10D, 12HU-10D and 15HU-10D) due to the complexity of the task and experience gained from previous works [11]. These ANN configurations are trained ten times with time steps set to 5 mm 3 /mm (time series 1, see Table 2), 10 mm 3 /mm (time series 2, see Table 2) and 15 mm 3 /mm (time series 3, see Table 2).…”

Estimation of the surface roughness of K9 optical glass in precision grinding using the brittle material removal fraction based on an improved image processing algorithm

“…Agustina et al evaluated the surface roughness obtained during robot-assisted polishing processes by the analysis of acoustic emission signals in the frequency domain [10] . Arriandiaga et al developed a virtual sensor for online monitoring of wheel wear and part roughness in the grinding process [11] . Santos et al investigated the relevance for the detection of surface and subsurface cracks in ceramic tiles by using ultrasonic techniques [12] .…”

A Fast and in-Situ Measuring Method Using Laser Triangulation Sensors for the Parameters of the Connecting Rod

“…Ito et al [8] used a laser displacement sensor (LDS) to measure the surface form of ceramics parts. Due to the limitation of its accuracy, however, this kind of method is commonly used for form measurement but not dimension inspection, such as in [9,10,11,12]. The main reason is understandable as the coordinates offered by the NC machines are less accurate, and thus the final measuring accuracy is not comparable to that of the CMM yet.…”