Predicting drill wear using an artificial neural network (original) (raw)
Related papers
Drill wear monitoring using back propagation neural network
Journal of Materials Processing Technology, 2006
Present work deals with prediction of flank wear of drill bit using back propagation neural network (BPNN). Drilling operations have been performed in mild steel work-piece by high-speed steel (HSS) drill bits over a wide range of cutting conditions. Important process parameters have been used as input for BPNN and drill wear has been used as output of the network. Inclusion of chip thickness as an input in addition to conventional parameters leads to better training of the network. Performance of the neural network has been found to be satisfactory while validated with experimental result.
Using of Artificial Neural Networks to Predict Drill Wear in machining processes
In machining operations a hard tool is engaged with work piece along with process. Tool is harder than work piece. However, tool wear occurrence in machining processes is inevitable. Tool wear will results in scraped parts and also it makes tool to weaken and then a tool failer will happen in the end. Therefore, an operator is needed to follow the process and change the tool when it is going to break. But this is a serious problem against automation. To create an automation system, we need to develop a monitoring system to predict tool wear rate by on-line and substitute it with an operator. In this paper by using of a wear model and experimental data and also motor current block diagram ,tool wear rate in drilling process will be predicted .To investigate the results, neural network method is used .The results compared with the real data show that the neural network results have a close fitness with the real data.
Prediction of Drill Flank Wear Using Radial Basis Function Neural Network
2006
In the present work, different type of artificial neural network (ANN) architectures have been used in an attempt to predict flank wear in drill bits. Flank wear in drill bit depends upon speed, federate, drill diameter and hence these parameters along with other derived parameters such as thrust force and torque have been used to predict flank wear using ANN. The results obtained from different ANN architectures have been compared and some useful conclusions have been made.
Experimental Analysis of Tool Wear in Drilling of EN-31 Using Artificial Neural Network
Tool wear in drilling is an important parameter with respect to surface quality of hole and failure of material. Operation performed with worn out tool may increase manufacturing cost. In this work, an attempt has been made to measure the wear of tool with the help of stereoscopic microscope and the results obtained have been compared with a statistical model in which tool wear is assumed as the function of thrust force, machining time, speed and feed. And also compared with ANN model in which input neurons are drill diameter, torque, thrust force, machining time, feed and speed, whereas output is tool wear. Comparison between these three results has also been made. It is found that ANN gives best result and can be used for online tool monitoring. Experiments performed from 1 to 40th hole while drilling operations have performed on EN-31.
Tool Wear Monitoring by Means of Artificial Neural Networks
The paper presents the application of multi layer perceptron artificial neural network for the tool wear monitoring in turning. To simulate factory floor conditions, six sets of cutting parameters were selected and applied in sequence. Six configurations of input parameter were tested to reveal their usability. Subsequently, the network's structure was optimised by means of an original pruning method, which makes possible an automatic network configuration. The obtained results prove the effectiveness of the studied BP neural networks for the purposes of tool wear monitoring.
Procedia Manufacturing, 2016
Nowadays, digital manufacturing systems with real-time process monitoring and control are in high demand in industries for productivity and quality improvement. During machining, such a system is anticipated to excerpt reliable data within a short time-lapse, monitor tool wear progress, anticipate its wear and breakage, alert the machinist in real time to avoid unexpected failure, and help obtaining quality products. This is vital, especially, when drilling Ni-/Ti-based superalloys as catastrophic failure and premature breakage of tools occur in random manner due to aggressive welding and chipping of the rake and flank faces. Spindle power data are easy to collect from modern machine tools and can be made available for such real-time data processing. This work aims to evaluate and analyze spindle power data for real-time tool wear/breakage monitoring during drilling of a Ni-based superalloy, Inconel 625. Experiments were performed by varying speed and feed. Power data were collected from the power meter (also called load meter) of the machine spindle to feed into the neural network (NN) for functional processing. As a counterpart, force data were also collected and processed to understand the reliability of the spindle power data. The results show that the trends of these two different types of data are similar for any feed and speed combinations. It is believed that such spindle power data integrated with the artificial intelligence (NN) system can be used for real-time tool wear/breakage monitoring and process control, thus can enhance digital manufacturing systems.
PREDICTIVE MODELLING OF DRILL WEAR: COMPARATIVE ANALYSIS OF ANN AND FUZZY LOGIC TECHNIQUES
Today's fast growing technology has raised the bar when it comes to the accuracy of machined components. The primary objective of this research is to estimate drill wear. In this study, drill wear estimation is carried out by considering Acoustic Emission (AE), Vibration Velocity and Drill Tool Chatter measured using image features obtained by Machine Vision system. In order to identify the tool wear conditions based on the signal measured, an Artificial Neural Network, using a Feed Forward -Back-Propagation algorithm, and Fuzzy Logic approach, have been adopted. The neural network is trained to estimate the average drill wear and after each drilling operation the drill wear is measured with Tool Maker's Microscope. The input parameters that are being used for estimation in this project were found to be non-linearly varying with the desired output. Due to this, the interpretation and prediction of data becomes very difficult. Hence, the two expert systems, i.e., Artificial Neural Network and Fuzzy Logic toolboxes will be used to analyse the best fit model in predicting the output of tool wear for this specific drill job. The prediction accuracy is then compared to analyse which model could give better results so that it can be recommended for machine learning and future work. When ANN and MAMDANI FIS methods were used and the actual tool wear and predicted tool wear were compared, it was observed that ANN produced better correlations and hence it is selected for predictions of tool wear for the present work conditions.
Scientia Iranica, 2017
Cutting tool wear in machining processes reduces the product surface quality, affects on the dimensional and geometrical tolerances and causes tool breakage during the metal cutting. Therefore, online tool wear monitoring is needed to prevent reduction in machining quality.An artificial neural network (ANN) model was developed in this study to predict and simulate the tool flank wear. To reach to this aim, an experiment array was provided using of full factorial method and the tests were conducted on a CNC lathe machine tool. Vibration amplitude of the cutting tool and cutting forces were considered as criterion variables in monitoring the tool flank wear. For designing the model, the cutting parameters, cutting forces and vibration amplitude were defined as model input and tool flank wear was selected as output. The model was also introduced as a simulation block diagram to be used as a useful model in online and automated manufacturing systems. The estimated and measured results were then compared with each other. Based on the comparison results, maximum squared error values are under and the R2 is 1 which it means that the designed model can predict the results with a high and reliable accuracy.
Journal of Manufacturing Systems, 2017
Digital manufacturing systems are determined to be a major key to enhance productivity and quality mainly due to real-time process monitoring and control capability with instant data processing. During machining, such systems are anticipated to excerpt reliable data within a short time-lapse, monitor tool wear progress, anticipate its wear and breakage, alert the machinist in real time to avoid unexpected failure of tool or machine, and help obtaining quality products. This is vital, especially, when drilling Ni-/Ti-based superalloys because catastrophic failure and premature breakage of tools occur in random manner due to aggressive welding and chipping of tool including the rake and/or flank faces and tool corner. Nowadays, spindle power data are easy to collect directly from modern machine tools and can be made available in production floor for such real-time data processing. This work aims to evaluate spindle power data for real-time tool wear/breakage prediction during drilling of a Ni-based superalloy, Inconel 625. Experiments were performed by varying speed and feed. Spindle power data were collected from the power meter (also called load meter) to feed into the neural network (NN) for functional processing. To understand the reliability of the spindle power data, force data were also collected and compared. The results show that the trends of these two different types of data over cutting time are similar for any feed and speed combinations. The error in NN prediction from actual wear was found to be between 0.8-18.4% with power data as compared to that between 0.4-17.9% with force data. Findings suggest that spindle power data integrated with the artificial intelligence (NN) system can be used for real-time tool wear/breakage monitoring and process control, thus appreciate digital manufacturing systems.
Prediction of Tool Wear Using Artificial Neural Networks during Turning of Hardened Steel
Materials
The ability to effectively predict tool wear during machining is an extremely important part of diagnostics that results in changing the tool at the relevant time. Effective assessment of the rate of tool wear increases the efficiency of the process and makes it possible to replace the tool before catastrophic wear occurs. In this context, the value of the effectiveness of predicting tool wear during turning of hardened steel using artificial neural networks, multilayer perceptron (MLP), was checked. Cutting forces and acceleration of mechanical vibrations were used to monitor the tool wear process. As a result of the analysis using artificial neural networks, the suitability of individual physical phenomena to the monitoring process was assessed.