Abstract The shearing behavior of discontinuities is one of the factors with major influence on rock slope stability analysis, which is predictable when adopting different analytical models. However, ...these methodologies, generally of a purely deterministic nature, do not allow an assessment of the influence of the variability of the input parameters of the models on the shear behavior of unfilled rock discontinuities and, consequently, on the risk involved in the rock slope stability analyses. The purpose of this article is to present a methodology for rock slope risk assessment based on the development of a model to predict the shear behavior of unfilled rock discontinuity considering the variability of its input parameters, using a Mamdani fuzzy controller. The input variables of the model are the boundary normal stiffness and initial normal stress acting on the discontinuity, its roughness, the uniaxial compressive strength, the basic angle of friction of the intact rock and the shear displacement imposed on the discontinuity. The model outputs are the membership functions for the shear strength and dilation of the unfilled rock discontinuity, and from which the membership function can be defined for the factor of safety of the rock slope considering the failure mechanism governed by discontinuity. The results reveal the use and importance of fuzzy logic and fuzzy number operations in assessing the risk in rock slopes and may be used even in situations where there are major uncertainties on the existing information of the characteristics of the unfilled discontinuities.
Water supply systems risk collapsing during droughts, which can affect millions of people. To mitigate these risks, we developed a proactive drought management system that integrates climate, ...hydrological variables, and mathematical modeling. The proposed Integrated Information and Early Warning System for Drought (IIEWSD) includes three main components: monitoring, prediction, and action, which trigger short- and long-term mitigating actions. By analyzing historical and forecasted time series on precipitation, flow, and water volume, the IIEWSD provides a notion of the tendency to drought worsening. The prediction component uses dynamic and statistical methods and artificial neural networks to monitor and predict the drought status of each water system, allowing the estimation of transition probabilities and the anticipation of actions. With predicted information, decision-makers can anticipate actions before the worsening of the drought state. A risk aversion planning matrix was proposed to help decision-makers trigger mitigation actions and responses, avoiding regret in anticipation of actions. This matrix restricts the use of forecasted information to intensify mitigation and response actions while using monitored information to reduce those actions. The IIEWSD was applied to the urban supply system of Fortaleza, Brazil, a 4.1 million metropolitan region whose main water resources are in the semi-arid region. The results revealed that the system effectively estimated the probabilities of future drought states. Our approach enhances proactive drought planning and management by integrating the assessment of current and future drought states.
Aiming at ensuring the quality of the product and reducing the cost of steel manufacturing, an increasing number of studies have been developing nonlinear regression models for the prediction of the ...mechanical properties of steel rebars using machine learning techniques. Bearing this in mind, we revisit this problem by developing a design methodology that amalgamates two powerful concepts in parsimonious model building: (i) sparsity, in the sense that few support vectors are required for building the predictive model, and (ii) locality, in the sense that simpler models can be fitted to smaller data partitions. In this regard, two regression models based on the Least Squares Support Vector Regression (LSSVR) model are developed. The first one is an improved sparse version of the one introduced in a previous work. The second one is a novel local LSSVR-based regression model. The task of interest is the prediction of four output variables (the mechanical properties YS, UTS, UTS/YS, and PE) based on information about its chemical composition (12 variables) and the parameters of the heat treatment rolling (6 variables). The proposed LSSVR-based regression models are evaluated using real-world data collected from steel rebar manufacturing and compared with the global LSSVR model. The local sparse LSSVR approach was able to consistently outperform the standard single regression model approach in the task of interest, achieving improvements in the average R2 from previous studies: 5.04% for UTS, 5.19% for YS, 1.96% for UTS/YS, and 3.41% for PE. Furthermore, the sparsification of the dataset and the local modeling approach significantly reduce the number of SV operations on average, utilizing 34.0% of the total SVs available for UTS estimation, 44.0% for YS, 31.3% for UTS/YS, and 32.8% for PE.
Abstract It is well known that learning about the mechanical behavior of a fractured rock mass depends on the shear behavior of its discontinuities. Several studies have shown that the shear behavior ...of unfilled rock discontinuities depends on their boundary conditions, roughness characteristics and the properties of the joints walls. Currently, there are several analytical models that can be used to predict the shear behavior of clean rock joints. However, they are all purely deterministic in nature because their input variables are defined without considering the uncertainties inherent in the formative process of the rock mass and the discontinuity itself, i.e., they need an auxiliary tool to consider the variability of their parameters such as the Monte Carlo or Point Estimation Methods. Therefore, the purpose of this article is to present a model to predict the shear strength of clean rock joints incorporating uncertainties in the variables that govern their shear behavior with a zero-order Takagi-Sugeno fuzzy controller. The model is developed based on the results of 44 direct shear tests carried out on different joints. The model input variables are the normal boundary stiffness and initial normal stress acting on the joint, its roughness expressed by the JRC value, the uniaxial compressive strength and basic friction angle of the intact rock, as well as the shear displacement imposed on the joint. The results showed that the predicted shear strength of clean rock joints obtained by the fuzzy model fit the experimental data satisfactorily and helped define the shear behavior of the discontinuity.
In this work, a fuzzy voltage controller design of a 1 kW high-gain, high-efficiency direct current converter operating in discontinuous conduction mode is developed. In this condition, the design of ...a conventional controller is more challenging. This converter is part of an autonomous photovoltaic pumping system without batteries consisting of four photovoltaic modules, a variable frequency inverter and an induction motor coupled to a pump. The boost converter is responsible for the voltage elevation of photovoltaic modules to a 311V direct current bus, which must not have oscillations for a good operation of an algorithm to track the maximum power point of the modules. The fuzzy controller was implemented in a digital signal processor device to control the boost converter, producing a response with an overshoot of 5.78%, settling time of 3.8s and zero error in steady state.
This paper reports results from a comprehensive performance comparison among standalone machine learning algorithms (SVM, MLP and GRNN) and their combinations in ensembles of classifiers when applied ...to a medical diagnosis problem in the field of orthopedics. All the aforementioned learning strategies, which currently comprises the classification module of the SINPATCO platform, are evaluated according to their ability in discriminating patients as belonging to one out of three categories: normal, disk hernia and spondylolisthesis. Confusion matrices of all learning algorithms are also reported, as well as a study of the effect of diversity in the design of the ensembles. The obtained results clearly indicate that the ensembles of classifiers have better generalization performance than standalone classifiers.
This paper introduces a novel forecastings technique based on randomized fuzzy cognitive maps (FCM), called LRHFCM (or large reservoir of randomized high-order FCM) for predicting univariate time ...series. LR-HFCM is a hybrid method combining fuzzy time series (FTS), FCMs, and reservoir computing. It is a type of echo state network (ESN) consisting of the input layer, intermediate (or large reservoir) layer, and output layer, where LASSO regression is applied to train the output layer. The novelty of this approach is that the internal layer includes a very large reservoir, considering different combinations from the sets of concepts and order using a certain number of sub-reservoirs to capture different dynamics of input time series. It is important to highlight that the weights within each sub-reservoir are chosen randomly and remain constant throughout the training process. The validity of the LR-HFCM approach is evaluated across 15 different time series datasets. The results highlight the outperformance of the LR-HFCM technique in comparison to various baseline models.
A pesquisa em redes neurais artificiais (RNAs) está atualmente experimentando um crescente interesse por modelos que utilizem a variável tempo como um grau de liberdade extra a ser explorado nas ...representações neurais. Esta ênfase na codificação temporal (temporal coding) tem provocado debates inflamados nas neurociências e áreas correlatas, mas nos últimos anos o surgimento de um grande volume de dados comportamentais e fisiológicos vêm dando suporte ao papel-chave desempenhado por este tipo de representação no cérebro BALLARD et al. (1998). Contribuições ao estudo da representação temporal em redes neurais vêm sendo observadas nos mais diversos tópicos de pesquisa, tais como sistemas dinâmicos não-lineares, redes oscilatórias, redes caóticas, redes com neurônios pulsantes e redes acopladas por pulsos. Como conseqüência, várias tarefas de processamento da informação têm sido investigada via codificação temporal, a saber: classificação de padrões, aprendizagem, memória associativa, controle sensório-motor, identificação de sistemas dinâmicos e robótica. Freqüentemente, porém, não fica muito claro até que ponto a modelagem dos aspectos temporais de uma tarefa contribui para aumentar a capacidade de processamento da informação de modelos neurais. Esta tese busca apresentar, de uma maneira clara e abrangente, os principais conceitos e resultados referentes à proposição de dois modelos de redes neurais não-supervisionadas (RNATs), e como estas lançam mão da codificação temporal para desempenhar melhor a tarefa que lhes é confiada. O primeiro modelo, chamado rede competitiva Hebbiana temporal (competitive temporal Hebbian - CTH), é aplicado especificamente em tarefas de aprendizagem e reprodução de trajetórias do robô manipulador PUMA 560. A rede CTH é uma rede neural competitiva cuja a principal característica é o aprendizado rápido, em apenas uma época de treinamento, de várias trajetórias complexas contendo ) elementos repetidos. As relações temporais da tarefa, representadas pela ordem temporal da trajetória, são capturadas por pesos laterais treinados via aprendizagem hebbiana. As propriedades computacionais da rede CTH são avaliadas através de simulações, bem como através da implementação de um sistema de controle distribuído para o robô PUMA 560 real. O desempenho da rede CTH é superior ao de métodos tabulares (look-up table) tradicionais de aprendizagem de trajetórias robóticas e ao de outras técnicas baseadas em redes neurais, tais como redes recorrentes supervisionadas e modelos de memória associativa bidirecional (BAM). O segundo modelo, chamado rede Auto-Organizável NARX (Self-Organizing NARX-SONARX), é baseado no conhecido algoritmo SOM, proposto por KOHONEN (1997). Do ponto de vista computacional, as propriedades de rede SONARX são avaliadas em diferentes domínios de aplicação, tais como predição de séries temporais caóticas, identificação de um atuador hidráulico e no controle preditivo de uma planta não-linear. Do ponto de vista teórico, demonstra-se que a rede SONARX pode ser utilizada como aproximador assintótico de mapeamentos dinâmicos não-lineares, graças a uma nova técnica de modelagem neural, chamada Memória Associativa Temporal via Quantização Vetorial (MATQV). A MATQV, assim como a aprendizagem hebbiana da rede CTH, é uma técnica de aprendizado associativo temporal. A rede SONARX é comparada com modelos NARX supervisionados, implementados a partir das redes MLP e RBF. Em todos os testes realizados para cada uma das tarefas citadas no parágrafo anterior, a rede SONARX tem desempenho similar ou melhor do que o apresentado por modelos supervisionados tradicionais, com um custo computacional consideravelmente menor. A rede SONARX é também comparada com a rede CTH na parendizagem de trajetórias robóticas complexas, com o intuito de destacar as principais diferenças entre os dois ) tipos de aprendizado associativo. Esta tese também propõe uma taxonomia matemática, baseada na representação por espaço de estados da teoria de sistemas, que visa classificar redes neurais não-supervisionadas temporais com ênfase em suas propriedades computacionais. Esta taxonomia tem como principal objetivo unificar a descrição de modelos neurais dinâmicos, facilitando a análise e a comparação entre diferentes arquiteturas, contrastando suas características representacionais e operacionais. Como exemplo, as redes CTH e a SONARX serão descritas usando a taxonomia proposta.
Neural network research is currently witnessing a significant shift of emphasis towards temporal coding, which uses time as an extra degree of freedom in neural representations. Temporal coding is passionately debated in neuroscience and related fields, but in the last few years a large volume of physiological and behavioral data has emerged that supports a key role for temporal coding in the brain BALLARD et al. (1998). In neural networks, a great deal of research is undertaken under the topics of nonlinear dynamics, oscillatory and chaotic networks, spiking neurons, and pulse-coupled networks. Various information processing tasks are investigated using temporal coding, including pattern classification, learning, associative memory, inference, motor control, dynamical systems identification and control, and robotics. Progress has been made that substantially advances the state-of-the-art of neural computing. In many instances, however, it is unclear whether, and to what extent, the temporal aspects of the models contribute to information processing capabilities. This thesis seeks to present, in a clear and collective way, the main issues and results regarding the proposal of two unsupervised neural models, emphasizing how these networks make use of temporal coding to perform better in the task they are engaged in. The first model, called Competitive Temporal Hebbian (CTH) network, is applied specifically to learning and reproduction of trajectories of a PUMA 560 robot. The CTH model is a competitive neural network whose main characteristic is the fast learning, in just one training epoch, of multiple trajectories containing repeated elements. The temporal relationships within the task, represented by the temporal order of the elements of a given trajectory, are coded in lateral synaptic trained with hebbian learning. The computational properties of the CTH network are assessed through simulations, as well ) as through the practical implementation of a distributed control system for the real PUMA 560 robot. The CTH performs better than conventional look-up table methods for robot trajectory learning, and better than other neural-based techniques, such as supervised recurrent networks and bidirectional associative memory models. The second model, called Self-Organizing NARX (SONARX) network, is based on the well-known SOM algorithm by KOHONEN (1997). From the computational view-point, the properties of the SONARX model are evaluated in different application domains, such as prediction of chaotic time series, identification of an hydraulic actuator and predictive control of a non-linear plant. From the theoretic viewpoint, it is shown that the SONARX model can be seen as an asymptotic approximator for nonlinear dynamical mappings, thanks to a new neural modelling technique, called Vector-Quantized Temporal Associative Memory (VQTAM). This VQTAM, just like the hebbian learning rule of the CTH network, is a temporal associative memory techniques. The SONARX network is compared with supervised NARX models which based on the MLP and RBF networks. For all simulations, in each one of the forementioned application domains, the SONARX network had a similar and sometimes better performance than those observed for standard supervised models, with the additional advantage of a lower computational cost. The SONARX model is also compared with the CTH network in trajectory reproduction tasks, in order to contrast the main differences between these two types of temporal associative learning models. In this thesis, it is also proposed a mathematical taxonomy, based on the state-space representation of dynamical systems, for classification of unsupervised temporal neural networks with emphasis in their computational properties. The main goal of this taxonomy is to unify the description of dynamic neural models, ) facilitating the analysis and comparison of different architectures by constrasting their representational and operational characteristics. Is is shown how the CTH and SONARX models can be described using the proposed taxonomy.