D. D'Addona

Evaluation of neural models applied to the estimation of tool wear in the grinding of advanced ceramics

Tool condition monitoring in grinding of advanced ceramics using neural networks.Acoustic emission and power signals were used in several statistical parameters.Results showed that the ANN were highly successful in estimating tool wear.Errors was less than 4%.The models will help to improve product quality and increase productivity. Grinding wheel wear, which is a very complex phenomenon, causes changes in most of the shapes and properties of the tool during machining, reducing the efficiency of the grinding operation and impairing workpiece quality. Therefore, monitoring the condition of the tool during the grinding process plays a key role in the quality of workpieces being manufactured. In this study, diamond tool wear was estimated during the grinding of advanced ceramics using intelligent systems composed of four types of neural networks. Experimental tests were performed on a surface grinding machine and tool wear was measured by the imprint method throughout the tests. Acoustic emission and cutting power signals were acquired during the tests and statistics were obtained from these signals. Training and validating algorithms were developed for the intelligent systems in order to automatically obtain the best estimation models. The combination of signals and statistics along with the intelligent systems brings an innovative aspect to the grinding process. The results indicate that the models are highly successful in estimating tool wear.

Tool-wear prediction and pattern-recognition using artificial neural network and DNA-based computing

Managing tool-wear is an important issue associated with all material removal processes. This paper deals with the application of two nature-inspired computing techniques, namely, artificial neural network (ANN) and (in silico) DNA-based computing (DBC) for managing the tool-wear. Experimental data (images of worn-zone of cutting tool) has been used to train the ANN and, then, to perform the DBC. It is demonstrated that the ANN can predict the degree of tool-wear from a set of tool-wear images processed under a given procedure whereas the DBC can identify the degree of similarity/dissimilar among the processed images. Further study can be carried out while solving other complex problems integrating ANN and DBC where both prediction and pattern-recognition are two important computational problems that need to be solved simultaneously.

Grinding Wheel Management through Neuro-Fuzzy Forecasting of Dressing Cycle Time

Abstract A multiple supplier tool management system based on reliable tool delivery forecasting is developed for optimum tool inventory sizing. The system is built up on the basis of historical data on tool management comprising the series of tool shipment and delivery dates between one manufacturing company and several external tool manufacturers in a supply network. If historical data are highly variable due to the incidence of uncertain but not random factors, classical time series analysis cannot measure the Imprecision deriving from events that are neither stochastic nor casual. An alternative is given by the analysis through intelligent computing techniques capable to deal with deterministic events but also take into account unpredictable factors for better results in prediction and forecast. A neuro-fuzzy system is used in this paper to approach optimum tool inventory sizing of CBN grinding wheels for nickel base alloy turbine blade fabrication.

An Intelligent Computation Approach to Process Planning in Multiple-Step Cold Forging

Abstract An intelligent computation approach to time and cost reduction in process planning of cold forging operations is illustrated. The problem taken into consideration is the generation of optimized working sequences in the fabrication of multi-diameter shafts through multiple-step cold forging. A supervised learning neural network paradigm was employed in order to identify the technologically feasible working sequences to be considered for process planning decision making. The process planner can then select the appropriate solution according to his experience or resort to further methods of detailed analysis (e.g. FEM analysis), with the advantage of applying time consuming numerical investigations only to a small number of cases suggested by the intelligent computing system. Neural network training and testing allowed to verify the system performance in classifying working sequence feasibility and its computational speed in providing technologically acceptable working sequences for process planner consideration.

Assessment of laser-based reverse engineering systems for tangible cultural heritage conservation

The process of acquiring the geometry and shape of a part and reconstructing its digital model is known as reverse engineering (RE). This approach is usefully employed in fields as diverse as product design, design modification, geometrical inspection, worn or damaged parts repair or remanufacturing, when physical object drawings, documentation or computer models are not available. The recent scientific and technical developments of RE methods and tools have broadened the possibilities of applications in the field of cultural heritage conservation ranging from reproduction (e.g. via rapid prototyping), maintenance (e.g. computer-aided repair), multimedia tools for education and dissemination (e.g. virtual museums), to artefact condition monitoring (e.g. computer-aided inspection) and many more. The first stage of the RE procedure is digital data acquisition that can be carried out by means of several different tools. The selection of the 3D digitising system is crucial as it directly affects the process time and the quality of the point cloud, which determines the final digital model. In this research work, following the EC FP7 open topic on ‘Equipment assessment for laser based applications’ compiled in Horizon 2020, two non-contact laser-based RE systems, respectively, based on a coordinate measuring machine and a portable 3D scanning equipment, are utilised for the digitisation and reconstruction of a free-form tangible cultural heritage artefact to comparatively assess the RE systems performance in terms of process time, accuracy and ease of use.

Intelligent Tool Management in a Multiple Supplier Network

The development and implementation of an intelligent Flexible Tool Management Strategy (FTMS), based on Fuzzy Logic (FL) theory and integrated in a Multi-Agent Tool Management System (MATMS) for automatic tool procurement in a supply network, is presented. The MATMS operates in the framework of a negotiation-based, multiple-supplier network where a turbine blade producer requires dressing jobs on worn-out CBN grinding wheels from external tool manufacturers. The main characteristics of the intelligent FTMS approach is the use of fuzzy set theory to model the uncertainty associated with tool demand rate and the employment of fuzzy reasoning to expand the strategy operational flexibility in comparison with traditional tool management and previously developed crisp FTMS paradigms.

Queuing network modelling techniques for response time enhancement in electronics assembly

The concept of quick response manufacturing points up that industrial manufacturing companies pressed to compete on speed by emerging competitor must change the management mind-set and reduce lead times throughout the organisation. As a factorys work-in-process, lead times and capacity are the result of complex dynamics in the production system, to predict whether a decision will improve lead times, it is necessary to be able to forecast both process and queue times. In this paper, a novel approach based on analytical techniques for queuing network modelling, called rapid modelling technology (RMT), is implemented via the MPX© software package for complex shop floor operations modelling. The main scopes of the MPX© utilisation are to evaluate the human and economic resources of a printed circuit board (PCB) producer company and to find solutions for the optimisation of the PCB manufacturing system performances.

Neural network based system for decision making support in orthodontic extractions

Publisher Summary An artificial Neural Network (NN) is a computational model of the human brain that assumes that computation is distributed over several simple interconnected processing elements, called neurons or nodes, which operate in parallel. This chapter illustrates the application of an intelligent computation approach to the support of the clinical decision-making on orthodontic extractions. Artificial neural networks trained using cephalometric and orthodontic cast measurements can provide a valuable support in the extraction therapeutical option in orthodontics. The chapter illustrates that neural network based decision-making support systems may be trained on the basis of clinical data and can be used where “rule based” decision making is unreliable or impossible. Therefore, neural network systems may become an important decision making support tool in orthodontics and find applications both in improving the clinical treatment and in maximizing the cost benefit of the treatment.

Intelligent classification of neural network models for mild steel behaviour in hot forming

Abstract The rheological behaviour of mild steel subjected to hot forming was modelled through a parallel distributed processing paradigm based on the artificial neural network prediction of the metal material response. The evaluation of different feedforward back-propagation neural networks for the flow stress prediction was carried out on the basis of laboratory data of the stress-strain behaviour of mild steel subjected to compression tests with different temperature and strain rate conditions. The results obtained consist of a number of neural network models capable of describing the material flow stress under the considered processing conditions with diverse levels of agreement with the experimental data. The availability of a range of neural network models for the prediction of mild steel rheological behaviour under hot forming conditions evidenced the need for optimum model identification. The latter was carried out through the development and application of selected intelligent computing methods: a supervised neural network technique, an unsupervised neural network procedure and a neuro-fuzzy system. Each of the developed intelligent computation paradigms provided for satisfactory results in the selection of an optimum neural network model for the flow stress prediction. The choice of the intelligent computing approach could therefore be determined by the complexity of the tool development phase and the ease in the resulting evaluation.

Spectra Measurements Using Piezoelectric Diaphragms to Detect Burn in Grinding Process

Researchers have evaluated a great number of monitoring techniques in order to control the surface condition of ground parts. Piezoelectric diaphragms of lead zirconate titanate are used in many fields, but these sensors are not common in the monitoring of the machining processes. This paper proposes a method for monitoring the workpiece surface condition (normal grinding and burn) by using a piezoelectric diaphragm and feature extraction techniques. A comparison is made with a conventional acoustic emission sensor, which is a traditional sensor in the monitoring of the machining processes. Grinding tests were performed in a surface-grinding machine with Society of Automotive Engineers (SAE) 1045 steel and cubic boron nitride (CBN) grinding wheel, where the signals were collected at 2 MHz. The workpieces were thoroughly analyzed through visual inspection, surface roughness and hardness measurements, and metallographic analyses. Study on the frequency content of both signals was carried out in order to select bands closely related to the workpiece surface condition. Digital filters were applied to the raw signals and features were extracted and analyzed. The root mean square values filtered in the selected bands for both sensors presented a better fitting to the linear regression, which is highly desirable for setting a threshold to detect burn and implementing into a monitoring system. Also, the basic damage index results show an excellent behavior for grinding burn monitoring for both sensors. The method was verified by using a different grinding wheel, which clearly shows its effectiveness and demonstrates the potential use of the low-cost piezoelectric diaphragm for grinding burn monitoring.