Tomoyuki Yajima

Integration of Fault Analysis and Interlock Controller Synthesis for Batch Processes

Abstract Integration amongst various decision-making processes, such as planning, design, and operation is necessary to dynamic and flexible batch production. To achieve a batch production integration, utilization of common models used for various decision-making processes is an effective approach. From this point of view, a batch system common model as described by a Petri net is proposed. In this article, a fault diagnosis technique for batch processes is presented using information about fault propagation and the possibilities of integration of fault analysis and controller synthesis are discussed on the basis of the Petri net based common models.

Integration between Scheduling and Design of Batch Systems Based on Petri Net Models

A batch process is a discontinuous and concurrent process which is suitable for multi-product, small-sized production. The distinctive feature of a batch process is that various decision making processes, such as scheduling, design, operation, etc. are strongly connected with each other. Interaction among these processes is necessary to dynamically and flexibly cope with a variety of unplanned events. This paper aims at presenting a batch scheduling technique based on Petri net models and showing the possibilities of integration between scheduling and design of batch processes. For this purpose, it first views the behavior of a batch operating system as a discrete event system and presents a Petri net model to be used for scheduling, design and operation. It next formulates batch scheduling problems based on Petri net partial languages, proposes their solution technique and last discusses the integration between scheduling and design of batch systems.

Synthesis of Operating Procedures and Procedural Controllers for Batch Processes Based on Petri Nets

Batch control is characterized by procedural control that is realized according to operating procedures to accomplish tasks. However, operating procedure synthesis takes considerable time and effort as the complexity of a batch process increases. To solve the operating procedure synthesis problem, a modeling technique is needed to represent the concurrent dynamics of a batch process. This paper aims at developing a method to synthesize operating procedures and procedural controllers for batch processes. For this purpose, it first views a batch control system as a discrete event system and next uses hierarchical Petri nets to represent information about process, plant, and schedule. It then presents a method to synthesize operating procedures. It also discusses the verification of synthesized operating procedures and last presents a method to synthesize procedural controllers.

The Lorentz gauge vector potential formulation for the boundary integral equation method

The boundary integral equation method is based on either the Poisson equation (for static problems) or the Helmholtz equation (for dynamic problems). For 3D eddy current calculations using the BIEM, the Lorentz gauge is most suitable since the Maxwell equations reduce to the Helmholtz equations under the Lorentz gauge. In this paper, the Lorentz gauge magnetic vector potential formulation, which yields a unique solution to the problem considered, is presented and numerically tested. It may be concluded from the computed results that the Lorentz gauge formulation and the Coulomb gauge formulation give almost the same computational accuracy, and the former is superior to the latter in terms of computation time and easiness of computer coding. >