Control Of Discrete Processes Research Articles

Solving the Problem of Discrete Process Control Synthesis Using Optimization on a Sliding Interval

Solving the Problem of Discrete Process Control Synthesis Using Optimization on a Sliding Interval

Output feedback control of discrete processes under time constraint: application to cluster tools

Practical control of timed behaviours in industrial applications is difficult. One of the few efficient modelling approaches to deal with time constraints on discrete event systems is to use timed event graph (TEG), a subclass of Petri nets. The dynamic behaviour of these graphs is represented by a linear equations system over the max-plus algebra. Up to date, these models were assumed to be fully observable. In this paper, it is demonstrated that if a TEG is strongly connected, one can derive a realisable output feedback control. In other words, the control law depends solely on the control itself and the system’s outputs, considered to be the only observable events. The satisfaction of a set of constraints is addressed in the case of a single input model (i.e. having only one input transition) and the procedure to derive a realisable output feedback control law is provided. Finally, the present control method is applied to a dual-armed cluster tool, a well-known industrial practical application.

Symbolic Verification of Control Systems and Operating Procedures

In this paper, we provide a review of Professor Powers’s and his students’ work on connecting fault analysis, discrete process control, human operating procedures, and symbolic model checking. In recent years, this type of research is placed under the banner of “cyber-physical systems research”. Some of the techniques and procedures Powers and his students developed can be found in the open literature and conference proceedings. However, they have not been published broadly due to the untimely passing of Professor Powers. A complete overview of the methods are not available, and the cap-stone results obtained in the two last Ph.D. theses have not been published.

Optimal stationary control of discrete processes and a polynomial time algorithm for stochastic control problem on networks

Abstract The stochastic version of classical discrete optimal control problems with a finite set of states and infinite time horizon is studied. An approach for determining the optimal stationary control in discrete processes with deterministic and random states’ transitions of the dynamical system is described. The main problem is formulated on networks and a polynomial time algorithm for its solving is proposed. An application of the proposed algorithm for a class of stochastic decision problems is described.

Generation and manipulation of “smart” droplets

We report the generation and manipulation of electrorheological (ER) droplets that exhibit the giant ER effect. The experiments were carried out on specially designed microfluidic chips, in which the ER droplets were generated by using the microfluidic flow-focusing approach. Both the size and formation rate of these droplets can be controlled through digitally applied electrical signals. The principle of droplet manipulation is based on the electrical responsiveness of ER droplets and hence the denotation of “smart” when the electrical signals can be triggered by sensing/control devices. Due to the unique characteristics of the GER effect, the smart droplets can deform and even stop the microfluidic channel flow under an applied electric field. The pressure difference induced by the smart droplets inside the micro-channel is controllable by varying the field strengths, droplet sizes and particle concentrations in the GER suspension. By trapping and timed release of smart droplets in different micro-branch channels, we demonstrate that the smart droplets generated upstream cannot only be stored or displayed in the desired downstream channel(s) and thereby offer the potential of micro-droplet display, but also be useful in counting, flow directing and sorting the desired number of passive droplets sandwiched between two smart droplets. Such capabilities of smart droplets will enable the programmable control of discrete processes in bio-analysis, chemical reactions, digital microfluidics, and digital droplet display.

SD Builder ®: A production rules-based tool for on-line simulation, decision making and discrete process control

The paper presents and discusses a structure and components of a distributed decision-making system for complex discrete systems (CDSs) and processes control. The decision-making nucleus of the system—system dynamics builder (SD Builder)—is described. The SD Builder is based on a hierarchical object-based representation of a CDS and on a production rules-based representation of discrete processes. It allows the using of on-line simulations, state-graph search, expert systems and other methods for decision making. A conceptual prototype of a sawmill in a timber factory illustrates the concepts described in the paper. In a case of a manufacturing system, the decision-making system makes a part of a manufacturing execution system (MES).

A generic production rules-based system for on-line simulation, decision making and discrete process control

The paper presents and discusses a structure and components of a distributed decision-making system for complex discrete systems and processes control. The decision-making nucleus of the system—system dynamics Builder (SD Builder)—is described. The SD Builder is based on a hierarchical object-based representation of a complex discrete system and on a production rules based representation of discrete processes. It allows the using of on-line simulations, state-graph search, expert systems and other methods for decision making. A conceptual prototype of a sawmill control system in a timber factory illustrates the concepts described in the paper. In a case of a manufacturing system, the decision-making system makes a part of a manufacturing execution system (MES).

In and Out of the Loop: Strategic Skill and Windows of Opportunity in Discrete Process Control

A long-standing question in cognitive engineering is the effect of computer aids on work skills. We examined the performance of human operators in a discrete process control environment. During an extended training period, humans controlled production under one of four different conditions (manual, hybrid, supervisory, and automatic) that varied the degree of in-the-loop control. After a training period all subjects transferred to a supervisory condition with two scheduling logic errors to compensate for. Training results showed that production was best overall with supervisory control, intermediate with hybrid control, and worst with manual control. Supervisory performance did not differ across the two scheduling aids. When scheduling logic errors were introduced, subjects who had trained in the automatic condition (simply observing) took longer to respond and let production drop most. Overall, performance depends on an interplay between (1) ability to identify goal-relevant cues and (2) ability to connect cues to action. Moreover, whether out-of-the-loop unfamiliarity is seen depends on how tight the “windows of opportunity” for action are in the domain.

Sharing out control in distributed processes

In this paper, we discuss the problem of distributed control of discrete processes. Given n sites from which only partial information is available about a process, we describe how to share out controllable events in order that the process meets a given global specification. As with many problems involving partial observation, the solution relies on inefficient algorithms. The second part of the paper treats the question of identifying problems that can be solved efficiently.

Programmable Intelligent Sequential Controllers and Real-Time Languages

The paper presents descriptions of sequential real-time control of discrete processes. Some problems concerned with process model for sequential control are presented. A microcomputer realisation of sequential control system is proposed. Finally proposals of three real-time sequential control languages which have been implemented in INTEL 8080 microcomputer are presented.

Functional Programming for Discrete Process Control

Manufacturing process can be described by transfer functions, which act on the manufactured objectʼs attributes, in the same way as computing functions act on data. Machines are regarded as interpretation resources, with the same conceptual function as processors, and a factory is seen as an object-flow interpreter. This analogy is explored in several ways, and it can be considered a starting point for the design on the process controller. The resulting methodology should contribute towards better communication between computer control users and suppliers, leading to faster and more reliable control software design.

An Analysis of Discrete Process Control Languages According to the Mealy Machine

The aim of this paper is to present a unified tool for Che analysis of real time controller specification and programming languages, according to Che MEALY model of sequential machines. The controller model is constituted of two parts: the observer, which reconstitutes the controlled process's state vector, and the combinatorial, which generates the commands as a function of this vector. Some significant aspects, such as parallelism, and the shared resource management problem are considered with respect to this model and to programming practices, from relay ladder diagrams up to procedural languages.

Optimal Control of Discrete Processes with Complete or Incomplete Information

N stage decision processes described by state vectors, control vectors and parameter vectors are considered. The objective function is given by an additive function of real-valued or vector-valued functions. In the oase of complete information, we formulate Bellmanean functional equations with parameters as sufficient optimality conditions. The case of incomplete information (also for multi-criterion optimization) is formulated by an operator in such away that in our theory stochastic control, adaptive control, and games (minimax control) are special cases. Extended Bellmanean functional equations are given as sufficient optimality conditions, if the operator has the properties multiplicativity and mo no tonicity. As an example of a class of applied problems for our theory, we formulate a transport control problem in the building industry.An approximative algorithm for the solution of optimal control problems with incomplete information together with error estimations is given as an example of suitable computational methods. The error bound functions are determined as solutions of new functional inequalities.

Singular controls in discrete systems

WE shall discuss singular controls in certain problems of the optimal control of discrete processes.