Pedro Pablo González Pérez

Net of multi-agent expert systems with emergent control

Abstract An aperture to expertise as a social outcome is to minimize supervised control: control of an expert system society and control of an agents society in multi-agent expert systems. The present paper starts this aperture with autonomous behavioral agents within Expert Systems that exhibit bottom-up emergent control. The behavior-agents are either reasoning agents or communicating agents. The reasoning agents have a blackboard Domain Working Memory or a Meta Working Memory. The Expert Systems communicate with each other in a net for the solution of a problem. A practical development has been made in the medical domain of interconsultation that shows expedient problem-solving behavior.

Redsiex: a cooperative network of expert systems with blackboard architectures

The blackboard architecture, originally developed for the system that permits the comprehension of language, HEARSAY II, has later been used in a great variety of domains and in various environments for the construction of systems. From the classic architecture of HEARSAY II, many applications, generalizations, extensions and refinements have been developed. In this paper we present REDSIEX, (RED de SIstemas EXpertos) which is a network of expert systems within a blackboard architecture, for the cooperation solution of distributed problems. The REDSIEX system inherits various of the elements defined by the architecture of HEARSAY II and incorporates new components and organization. These produce a very characteristic and exclusive global work style in the solution of problems, within a conceptual framework of emergent control. The main structural and functional characteristics of REDSIEX are discussed.

A Model for Combination of External and Internal Stimuli in the Action Selection of an Autonomous Agent

This paper proposes a model for combination of external and internal stimuli for the action selection in an autonomous agent, based in an action selection mechanism previously proposed by the authors. This combination model includes additive and multiplicative elements, which allows to incorporate new properties, which enhance the action selection. A given parameter α, which is part of the proposed model, allows to regulate the degree of dependence of the observed external behaviour from the internal states of the entity.

Optimization Using Neural Network Modeling and Swarm Intelligence in the Machining of Titanium (Ti 6Al 4V) Alloy

The process of titanium’s machining in the aerospace industry today is by trial and error, it produce non efficient results, because this material is classified by the high chemical reaction with other materials and its low thermal conductivity such as a difficult to machine, so the process of finding the correct parameters for machining are hard to determine, and today researchers are looking to develop new models to predict and optimize these parameters. A recently developed optimization algorithm called particle swarm optimization is used to find optimum process parameters.Accordingly, the results indicate that a system where neural network is used to model and predict process outputs and particle swarm optimization is used to obtain optimum process parameters can be successfully applied to multi-objective optimization of titanium’s machining process

A Comparison between Back Propagation and the Maximum Sensibility Neural Network to Surface Roughness Prediction in Machining of Titanium (Ti 6Al 4V) Alloy

Titanium alloys are attractive materials due to their unique high strength, excellent performance at elevated temperatures and exceptional resistance to corrosion. The aerospace and military industries are the main users of this material. Titanium alloys are classified as materials difficult to machine. The correct parameters for machining are a hard to determine, and today researches are looking to develop new models to predict and optimize these parameters. The surface roughness (Ra) in turning of a titanium alloy machining Ti 6Al 4V predicted using neural and maximum sensitivity network is shown. The machining tests were carried out using PVD (TiAIN) coated carbide inserts under different cutting conditions. Confidence intervals were estimated in the model to get correct results. There are various machining parameters and they have an effect on the surface roughness. A set of initial parameters in finished turning of Ti 6Al 4V obtained from literature have been used. These parameters are cutting speed, feed rate and depth of cut. This paper shows the results obtained using these neural networks approaches to analyze the variables to model the machining process.

Preliminary Meanline Design for Gas Turbines Using Multi-objective Optimization

Designing gas turbines is a very complex task. It is not a linear procedure but an iterative one, composed by several phases. In the initial phase, the general geometric characteristics and estimate efficiency of the turbine are determined. This phase is known as the meanline design, and it is very important because it determines the starting point for more complex analysis. In this work we use a multi--objective evolutionary algorithm to calculate the meanline design. We consider two conflicting objectives: the number of stages of the turbine, and the efficiency of the stages.

Development and Application of an Intelligent System to Predict and Optimize the Surface Roughness of 1018 and 4140 Steel

The aim of this research is to present a new methodology for predicting and optimizing the surface roughness during machining of 1018 and 4140 Steel. There is particular interest in finding the best machining value parameters that should be used to achieve good surface roughness. These parameter values can be found by this neural intelligent approach. This methodology analyzes and identifies the parameters involved in the machining process; with this information the model is able to predict the surface roughness value in different conditions and then optimize the results with different intelligent heuristics. The experimental results show that we may conclude that this intelligent system is a suitable methodology for predicting and optimizing surface roughness during the machining of 1018 and 4140 Steel.