Dan Stefanoiu

Holonic self-organization of multi-agent systems by fuzzy modeling with application to intelligent manufacturing

Holonic manufacturing aims to design standardized, modular manufacturing systems made of interchangeable parts, to enable flexibility, online reconfigurability and self-organizing capabilities for the production systems. Recent advances in distributed artificial intelligence and networking technologies have proven that theoretical multi-agent systems (MAS) concepts are very suitable for the real life implementation of holonic concepts. Building on our recent results in the design and implementation of holonic reconfigurable architectures, the paper introduces a novel approach to the online self-organization of distributed systems. By using fuzzy set and uncertainty theoretical concepts, we construct a mathematical foundation for modeling MAS, where appropriate holonic structures are identified for each particular application. This approach opens new possibilities for the design of any distributed system that needs self-organization as an intrinsic property.

A genetic matching pursuit algorithm

Advanced signal processing (SP) methods often lead to greedy algorithms that are time consuming, so useless in real time applications. However, nowadays, efficient implementations of such algorithms become more and more feasible, eventually with the help of concepts withdrawn from fields outside SP. This paper introduces an implementation for one of the most appealing signal analysis procedures based on a method devised about 10 years ago: the matching pursuit algorithm (MPA).

Fault Detection and Diagnosis Using Parameter Estimation with Recursive Least Squares

The computational simplicity of Fault Detection and Diagnosis (FDD) schemes based on parameter estimation, as well as their suitability for integrated design with Fault Tolerant Control (FTC), make them attractive from the point of view of their implementation within real-time practical applications. This paper focuses on the advantages that these methods can offer as opposed to state estimation and parity equations based system diagnosis, while also highlighting the disadvantages of the parameter estimation approach in view of identifying the problems that could benefit from a simultaneous state and parameter estimation based scheme. In order to achieve this objective, a system diagnosis procedure is proposed, which makes use of the Recursive Least Squares (RLS) method with exponential weighting and constant forgetting factor in view of detecting the occurrence of a fault within the system, as well as providing the post-fault system model. An application example from the aircraft control field is also presented in order to illustrate the efficiency of the proposed approach.

ASTANK2: Analytical Modeling and Simulation

The paper focuses on building the mathematical model of a filling/drainage process (namely ASTANK2), by analytical approach. Models are subsequently tested by simulation. The process is versatile: it consists of two tanks with different geometric shape, which could be considered in independent or coupled setup and are supplied by an adjustable water distribution pipelines network. The proposed model is nonlinear and can be defined in various configurations, by connecting the desired component models, from the available ones. A simulator that integrates these components is implemented in MATLAB-SIMULINK environment. Its goal is twofold: to simulate the plant dynamics according to several scenarios, on one hand, and to validate the proposed analytical model, on the other hand. Some simulation results are presented in the end.

Fuzzy-Statistical Reasoning in Fault Diagnosis

When searching for faults threatening a system, the human expert is sometimes performing an amazingly accurate analysis of available information, frequently by using only elementary statistics. Such reasoning is referred to as “fuzzy reasoning,” in the sense that the expert is able to extract and analyse the essential information of interest from a data set strongly affected by uncertainty. Automating the reasoning mechanisms that represent the foundation of such an analysis is, in general, a difficult attempt, but also a possible one, in some cases. The chapter introduces a nonconventional method of fault diagnosis, based upon some statistical and fuzzy concepts applied to vibrations, which intends to automate a part of human reasoning when performing the detection and classification of defects.

Faults Diagnosis through Genetic Matching Pursuit

Signals that carry information regarding the existing defects or possible failures of a system are sometimes difficult to analyze because of various corrupting noises. Such signals are usually acquired in difficult conditions, far from the place where defects are located and/or within a noisy environment. Detecting and diagnosing the defects require then quite sophisticated methods that are able to make the distinction between noises encoding the defects and another parasite signals, all mixed together in an unknown way. Such a method is introduced in this paper. The method combines time-frequency-scale analysis of signals with a genetic algorithm.

A New Challenge in Ecological Process Control Based on PLCs with Profinet Communication Protocol

The current situation of environmental resources requires their efficient usage through efficient process control. But, process control can not be achieved without an essential component: the communication between all devices, such as sensors, controllers, actuators, programming stations etc. The aim is to achieve the irrigation process of plants in greenhouse conditions. Therefore, it is submitted a comprehensive application that proposes to be a distributed control system including data acquisition and processing, both local and remote monitoring of the process parameters. The distributed system is based on modern equipment: Siemens Simatic PLCs and HMI, Memsic e̅Ko Pro wireless sensors network, level sensors for irrigation tanks and actuators such as on/off valves and pumps. In terms of communication between equipments, an industrial bus and protocols were used - Profinet and wireless communication. The software part is provided by TIA Portal with STEP 7 and WinCC flexible RT and LOGO! Soft Comfort for programming the automation devices. ModbusTCP was used to develop the communication protocol between central devices.

A WSN Based Monitoring System for Oil and Gas Transportation through Pipelines

Abstract This article introduces some results obtained in monitoring of long pipelines employed in oil and gas transportation. A mobile unit for signals acquisition, processing, identification and diagnosis (MUSAPID), based on wireless sensors networks (WSN) technology was employed. The MUSAPID architecture was adapted to match the configuration of monitoring system, inspired by the national energy dispatcher. Thus, the system includes four hierarchical levels assigned to: the central dispatcher, the zonal dispatcher, the local dispatcher and the sensor nodes. The sensors are acquiring the flow and temperature of oil from the monitored pipeline. Sensors distribution along the pipeline is of in-line type. Nevertheless, a cross-link schema between sensor nodes is proposed, in order to increase the exploitation safety and to reduce false alarms sending. Two independent monitoring approaches based on comparison between values of sensors in pairs of successive nodes allow controlling the occurrence of rifts and leaks in the pipeline network. One of the approaches is unconventional and uses inter-correlation between the data acquired from two successive sensor nodes, as well as auto-correlation of temperature data. A simple logic to avoid sending false alarms was also implemented.

A fuzzy-statistical reasoning model for bearings fault diagnosis

When searching for faults threatening a system, the human expert sometimes performs an amazingly accurate analysis of available information, eventually by only using some elementary statistics. Automating the reasoning mechanisms founding such an analysis is, in general, a difficult attempt, but also a possible one, in some cases. The paper aims to introduce a nonconventional method of bearings faults diagnosis, based upon some statistical and fuzzy concepts.

Environment Communication and Control Systems Integrated on Teaching Platforms: Case Study: Double Water Tank System

This paper introduces a solution for integrating environment systems on teaching platforms using two components: communication and primary control. The communication relies on a bus and specific protocol. The primary control algorithms were implemented in MATLAB-SIMULINK environment, while using Siemens S7-1200 PLC as acquisition server and some regular PC as a client, communicating to each other through the Mudbugs-TCP protocol. The teaching platform is represented by the double water tank system ASTANK2, after being equipped with sensors and actuators.