Knowledge Discovery In Time Series Research Articles

Towards a systematical approach for wear detection in sheet metal forming using machine learning

Wear is one of the decisive factors for the economic efficiency of sheet metal forming processes. Thereby, progressive wear phenome lead on the one hand to a poor workpiece quality and on the other hand to tool failure resulting in high machine downtimes. This trend is intensified by processing high-strength materials and the reduction of lubricant up to dry forming. In this context, data-driven monitoring methods such as machine learning (ML) provide the potential of detecting wear at an early stage to overcome manual and cost-intensive process inspections. The presented study aims to provide a ML based inline quantification of wear states within sheet metal forming processes. The development of this monitoring approach is based on a procedure model the Knowledge Discovery in Time series and image data in Engineering Epplications (KDT-EA) which is validated on two forming processes, blanking and roll forming, that strongly differ in their physical process behavior and their acquired process data. The presented inline quantification allows an estimation of wear states with a deviation of less than 0.83% for the blanking process and 2.21% for the roll forming process from the actual wear state. Furthermore, it is shown that combining different feature extraction methods as well as a compensation of unbalanced data using data augmentation techniques are able to improve the performance of the investigated ML models.

Knowledge Discovery in Engineering Applications Using Machine Learning Techniques

Abstract Sensorial acquired process data combined with machine learning (ML) algorithms are fundamental for mastering the challenges of modern production systems, however, their potential is rarely exploited in real-world manufacturing applications. In this context, the literature presents systematic procedure models to generate knowledge from data, such as the cross industry standard process for data mining (CRISP-DM) model, which is used as a standard methodology for conducting data mining in industrial applications. However, these models do not take into account boundary conditions of manufacturing processes as well as the characteristics of the sensorial acquired data within these systems to generate knowledge. Therefore, this work presents a novel procedure model for knowledge discovery in time series and image data in engineering applications (KDT-EA). A holistic view of knowledge discovery in manufacturing processes becomes feasible with a strong focus on data acquisition, data preprocessing, and data transformation to generate reliable input data for ML models estimating the actual state of manufacturing processes. The process model supports operators in industry setting up a suitable measurement chain acquiring high-quality data and selecting preparation techniques depending on superimposed disturbances. Furthermore, it suggests data transformation techniques reducing the amount of data without losing informational value and establishing a basis for product-related inline monitoring. To quantify the benefits of KDT-EA and the impact of its phase on the quality of the generated knowledge, the novel procedure model is applied to an application in the field of inline wear detection on a sheet metal forming tool.

Empirical study of symbolic aggregate approximation for time series classification

Symbolic Aggregate approximation (SAX) has been the de facto standard representation methods for knowledge discovery in time series on a number of tasks and applications. So far, very little work has been done in empirically investigating the intrinsic properties and statistical mechanics in SAX w ords. In this paper, we applied several statistical measurements and proposed a new statistical measurement, i.e. information embedding cost (IEC) to analyze the statistical behaviors of the symbolic dynamics. Our experiments on the benchmark datasets and the clinical signals demonstrate that SAX can always reduce the complexity while preserving the core information embedded in the original time series with significant embedding efficiency, as well as robust to missing values and noise. Our proposed IEC score provide a priori to determine if SAX is adequate for specific dataset, which can be generalized to evaluate other symbolic representations. Our work provides an analytical framework with several statistical tools to analyze, evaluate and further improve the symbolic dynamics for knowledge discovery in time series.

A chaotic map algorithm for knowledge discovery in time series: a case study on biomedical signals

A chaotic map algorithm is proposed to study similarity-based knowledge of temporal patterns. Several information sources can be regarded as time series, both in scientific and technological fields such as nuclear physics, computer network, biomedical signals, and many others. The application of an automatic knowledge discovery mechanism has a strong impact on system science and engineering. The advantage of the proposed algorithm is due to its capability to extract meaningful features from complex data sets, as temporal patterns, without teacher. A case study has been carried on biomedical signals, such as electroencephalographic records, to recognize patterns affected by the Huntington's disease, one of the most dangerous pathology of the central nervous system. The chaotic map algorithm succeeds in distinguishing between pathological and normal patterns, with high values of both sensitivity and specificity.