Chatter identification and detection in machining processes has been an active area of research in the past two decades. Part of the challenge in studying chatter is that machining equations that describe its occurrence are often nonlinear delay differential equations. The majority of the available tools for chatter identification rely on defining a metric that captures the characteristics of chatter, and a threshold that signals its occurrence. The difficulty in choosing these parameters can be somewhat alleviated by utilizing machine learning techniques. However, even with a successful classification algorithm, the transferability of typical machine learning methods from one data set to another remains very limited. In this paper we combine supervised machine learning with Topological Data Analysis (TDA) to obtain a descriptor of the process which can detect chatter. The features we use are derived from the persistence diagram of an attractor reconstructed from the time series via Takens embedding. We test the approach using deterministic and stochastic turning models, where the stochasticity is introduced via the cutting coefficient term. Our results show a 97% successful classification rate on the deterministic model labeled by the stability diagram obtained using the spectral element method. The features gleaned from the deterministic model are then utilized for characterization of chatter in a stochastic turning model where there are very limited analysis methods.
Probabilistic conceptual network is a knowledge representation scheme designed for reasoning about concepts and categorical abstractions in utility-based categorization. The scheme combines the formalisms of abstraction and inheritance hierarchies from artificial intelligence, and probabilistic networks from decision analysis. It provides a common framework for representing conceptual knowledge, hierarchical knowledge, and uncertainty. It facilitates dynamic construction of categorization decision models at varying levels of abstraction. The scheme is applied to an automated machining problem for reasoning about the state of the machine at varying levels of abstraction in support of actions for maintaining competitiveness of the plant.
In recent years, much of the progress in Computer-Aided Manufacturing has emphasized the use of simulation, finiteelement analysis, and other science-based techniques to plan and evaluate manufacturing processes. These approaches are all based on the idea that we can build sufficiently faithful models of complex manufacturing processes such as machining, welding, and casting. Although there has bcen considerable progress in this area, it continues to suffer from difficulties: the first of these is that the kind of highly accurate models that this approach requires may' take many person months to construct, and the second is the large amount of computing resources needed to run these simulations. Two design advisors, Near Net-Shape Advisor and Design for Machinability Advisor, are being developed to explore the role of heuristic, knowledge-based systems for manufacturing processes, both as an alternative to more analytical techniques, and also in support of these techniques. Currently the advisors are both in the prototype stage. All indications lead to the conclusion that the advisors will be successful and lay the groundwork for additional systemsuch as these in the future.
This analysis step extracts a set of planning goals as well as a set of geometrical and technological constraints, e.g. the technological requirement that a groove can only be machined if the outline which it is contained in, has been already machined. These constraints are represented by ordering relations between goals representing the planning problem (Figure 1). In classical planning, orderings are only defined between steps. In this sense the orderings between goals are used to imply orderings between the subplans to reach these goals. The effect of these orderings is that the size of the search space can be significantly reduced without any further reasoning or representation effort, because processing of these constraints is completely done by the analysis step and by internal consistency mechanisms of the planner.
Computer Aided Process Planning (CAPP) as a significant part of the overall automation of manufacturing activities has received much research attention in both academia and industry in the past twenty years. There are two approaches to the automation of process planning: variant and generative. The variant approach is older and is based on the assumption that for a given part design there may be some similar parts that have been produced in the past. The logical way to produce the new process plan is to look at a similar one and modify it to satisfy the new design requirements. The generative approach uses a different method.