Fuzzy distributed artificial intelligence systems

In this paper, we introduce a DAI approach called hereinafter Fuzzy Distributed Artificial Intelligence (FDAI). Through the use of fuzzy logic, we have been able to develop mechanisms that we feel may effectively improve current DAI systems, giving much more flexibility and providing the subsidies which a formal theory can bring. The appropriateness of the FDAI approach is explored in an important application, a fuzzy distributed traffic-light control system, where we have been able to aggregate and study several issues concerned with fuzzy and distributed artificial intelligence. We also present a number of current research directions necessary to develop the FDAI approach more fully.

Influence of Earth's shadow on the rotational motion of an artificial satellite perturbed by solar radiation torque

A semi-analytical approach is proposed to study the rotational motion of an artificial satellite under the influence of the torque due to the solar radiation pressure and taking into account the influence of Earth's shadow. The Earth's shadow is introduced in the equations for the rotational motion as a function depending on the longitude of the Sun, on the ecliptic's obliquity and on the orbital parameters of the satellite. By mapping and computing this function, we can get the periods in which the satellite is not illuminated and the torque due to the solar radiation pressure is zero. When the satellite is illuminated, a known analytical solution is used to predict the satellite's attitude. This analytical solution is expressed in terms of Andoyer's variables and depends on the physical and geometrical properties of the satellite and on the direction of the Sun radiation flux. By simulating a hypothetical circular cylindrical type satellite, an example is exhibited and the results agree quite well when compared with a numerical integration. © 1997 COSPAR. Published by Elsevier Science Ltd.

Comparison of artificial and natural, new and reused, brackish water for the larviculture of the freshwater prawn Macrobrachium rosenbergii in a recirculating system

Differences in culture duration, metamorphosis rate and the productivity in hatchery culture of M. rosenbergii using a closed system with natural and artificial brackish water were evaluated. Reuse of brackish water in more than one hatchery cycle was also evaluated. Natural and artificial brackish water constituted the two tested treatments, which were distributed in four independent recirculating systems (tank and respective biofilter). Four batches of cultures were conducted and the 2nd and 4th reused the water from the 1st and 3rd, respectively. Mean duration of the hatchery period was 28 d in natural brackish water and 31 d in artificial brackish water. The metamorphosis rate and the average productivity for the natural brackish water treatment were 74% and 60 postlarvae/ L. respectively, and values obtained with artificial brackish water were 55% and 44 postlarvae/L. The successful hatchery culture of M. rosenbergii in this specific artificial brackish water suggests its potential use in enterprises located far from the coast. Brackish water can be used in two consecutive cultures without a negative effect on productivity.

New approach of induction machine drive using an adaptive digital PID controller with gain planning based on artificial neural networks

The present work introduces a new strategy of induction machines speed adjustment using an adaptive PID (Proportional Integral Derivative) digital controller with gain planning based on the artificial neural networks. This digital controller uses an auxiliary variable to determine the ideal induction machine operating conditions and to establish the closed loop gain of the system. The auxiliary variable value can be estimated from the information stored in a general-purpose artificial neural network based on CMAC (Cerebellar Model Articulation Controller).

Evaluation and identification of lightning models by artificial neural networks

This paper describes a novel approach for mapping lightning models using artificial neural networks. The networks acts as identifier of structural features of the lightning models so that output parameters can be estimated and generalized from an input parameter set. Simulation examples are presented to validate the proposed approach. More specifically, the neural networks are used to compute electrical field intensity and critical disruptive voltage taking into account several atmospheric and structural factors, such as pressure, temperature, humidity, distance between phases, height of bus bars, and wave forms. A comparative analysis with other approaches is also provided to illustrate this new methodology.

Impedance-based structural health monitoring with artificial neural networks

This paper presents a non-model based technique to detect, locate, and characterize structural damage by combining the impedance-based structural health monitoring technique with an artificial neural network. The impedance-based structural health monitoring technique, which utilizes the electromechanical coupling property of piezoelectric materials, has shown engineering feasibility in a variety of practical field applications. Relying on high frequency structural excitations (typically >30 kHz), this technique is very sensitive to minor structural changes in the near field of the piezoelectric sensors. In order to quantitatively assess the state of structures, multiple sets of artificial neural networks, which utilize measured electrical impedance signals for input patterns, were developed. By employing high frequency ranges and by incorporating neural network features, this technique is able to detect the damage in its early stage and to estimate the nature of damage without prior knowledge of the model of structures. The paper concludes with experimental examples, investigations on a massive quarter scale model of a steel bridge section and a space truss structure, in order to verify the performance of this proposed methodology.

Modeling and identification of fertility maps using artificial neural networks

The application of agricultural fertilizers using variable rates along the field can be made through fertility maps previously elaborated or through real-time sensors. In most of the cases applies maps previously elaborated. These maps are identified from analyzes done in soil samples collected regularly (a sample for each field cell) or irregularly along the field. At the moment, mathematical interpolation methods such as nearest neighbor, local average, weighted inverse distance, contouring and kriging are used for predicting the variables involved with elaboration of fertility maps. However, some of these methods present deficiencies that can generate different fertility maps for a same data set. Moreover, such methods can generate inprecise maps to be used in precision farming. In this paper, artificial neural networks have been applied for elaboration and identification of precise fertility maps which can reduce the production costs and environmental impacts.

Influence of artificial reefs on fish assemblage of the Barra Bonita Reservoir (São Paulo, Brazil)

This paper deals with the effects of introduced artificial reefs on the diversity of freshwater fish communities in lentic and lotic zones of a very impacted river in southeastern Brazil. To accomplish this goal, artificial reefs were introduced, in December 1997, in the Barra Bonita reservoir and in the lotic zone immediately below the dam. Fish diversity was always higher in the lotic zone than in the reservoir. Accordingly, fish diversity near the artificial reefs was consistently higher than in the control areas. We propose that the higher environmental complexity in the lotic zone, as compared to the reservoir, is incremented in areas where artificial reefs were introduced; also, we present that, in both areas, diversity is mainly affected by the introduction of artificial reefs.

Investigations into the use of the finite element method and artificial neural networks in the non-destructive analysis of metallic tubes

This work presents an investigation into the use of the finite element method and artificial neural networks in the identification of defects in industrial plants metallic tubes, due to the aggressive actions of the fluids contained by them, and/or atmospheric agents. The methodology used in this study consists of simulating a very large number of defects in a metallic tube, using the finite element method. Both variations in width and height of the defects are considered. Then, the obtained results are used to generate a set of vectors for the training of a perceptron multilayer artificial neural network. Finally, the obtained neural network is used to classify a group of new defects, simulated by the finite element method, but that do not belong to the original dataset. The reached results demonstrate the efficiency of the proposed approach, and encourage future works on this subject.

Using artificial intelligence for the control of medium voltage substations

This paper introduces a method for the supervision and control of devices in electric substations using fuzzy logic and artificial neural networks. An automatic knowledge acquisition process is included which allows the on-line processing of operator actions and the extraction of control rules to replace gradually the human operator. Some experimental results obtained by the application of the implemented software in a simulated environment with random signal generators are presented.

Artificial reefs and their effects on fish assemblages in a Brazilian reservoir and Tailrace

In this paper, we evaluate the effects of artificial reefs on fish assemblages in a hypereutrophic reservoir and in the lotic zone immediately below dam. Fish diversity was highest in the lotic zone relative to the reservoir. We also found an inverse relationship between diversity and distance from the river margin. Catches near the artificial reefs were more diverse than in control areas. A seasonal effect, possibly caused by variation in temperature, was significant in all comparisons. We argue that, in a scale of local effects, the ecological function of these structures would be similar to refuges.

The influence of fruit morphology and habitat structure on ant-seed interactions: A study with artificial fruits

Artificial fruits designed to simulate lipid-rich non-myrecochorous diaspores were used to test for the effect of fruit morphology and habitat structure on ant-seed interactions in an Atlantic Forest site in SE Brazil. The outcome of the interaction (i.e., if the fruit was removed, cleaned by ants on the spot or had no interaction with ants) and the time of ant response were the investigated variables. Models simulating drupes and arilate diaspores were used to test for morphological effects and four habitat attributes (litter depth, number of logs, number of trees, and percentage of bromeliad coverage on the forest floor), likely to be correlated with the ant diversity and abundance in the study site, were measured to test for the effect of habitat structure. The proportion of fruits removed or cleaned did not differ between the two morphological models. Sites in which fruits were cleaned had more trees than those in which no interaction occurred. This may be a result of the foraging behavior of arboreal ants that frequently descend to the forest floor to exploit fleshy diaspores. Sites in which model removal occurred had lower litter depth than both those in which models were cleaned and those in which no interaction occurred. A negative correlation was observed between litter depth and ant response time. Accumulation of leaf litter at a given point may have constrained the movements of large ants in general, and ponerine ants (that are important seed removers) in particular. We conclude that that local pattern in litter depth and tree density influence the frequency and outcome of interactions between ants and non-myrmecochorous, fleshy diaspores.

Determination of resonant frequencies of triangular and rectangular microstrip antennas, using artificial neural networks

This paper presents models that can be used in the design of microstrip antennas for mobile communications. The antennas can be triangular or rectangular. The presented models are compared with deterministic and empirical models based on artificial neural networks (ANN) presented in the literature. The models are based on Perceptron Multilayer (PML) and Radial Basis Function (RBF) ANN. RBF based models presented the best results. Also, the models can be embedded in CAD systems, in order to design microstrip antennas for mobile communications.