RESUMOA região oeste do estado do Paraná, que abrange a Bacia Hidrográfica Paraná III, possui um sistema agrícola ainda muito dependente das condições climáticas. Diante disto, os aspectos que envolvem a construção de modelos probabilísticos, capazes de determinar parâmetros de ocorrência e quantificação da precipitação pluvial diária, têm-se destacado como fator necessário e fundamental para consecução dos projetos regionais, em todas as áreas. Portanto, o objetivo deste trabalho foi realizar a construção dos referidos modelos considerando-se a ocorrência dos fenômenos ENOS e se utilizando, para tal, os dados de 34 estações meteorológicas com séries superiores a 21 anos de registros diários, obtidos junto à Agência Nacional de Águas. A modelagem da ocorrência da precipitação diária foi tratada por meio dos processos de Markov enquanto a modelagem da quantidade precipitada foi feita por meio do ajuste à função gama de probabilidade. Para a validação do ajuste utilizou-se o teste de Kolmogorov-Smirnov e a partir dos resultados obtidos foi possível constatar que a metodologia desenvolvida pode ser aplicada para a simulação de séries sintéticas de precipitação pluvial diária para a área de estudo. Palavras-chave: recursos hídricos, séries sintéticas, distribuição gama de probabilidadeModeling of daily intra-annual precipitation of the Paraná III Basin associated with events ENSO ABSTRACT The western region of Paraná state, covering the Paraná III Basin, has an agricultural system still largely dependent on climatic conditions. Thus, the aspects that involve the construction of probabilistic models, able to determine parameters of occurrence and quantification of the daily precipitation have emerged as a necessary and fundamental factor for the achievement of regional projects in all areas. Therefore, the objective of this work was the construction of such models considering the occurrence of ENSO phenomena, using for this purpose data from 34 weather stations with series of more than 21 years of daily records, which were obtained from the ANA (Agência Nacional de Águas). The modeling of the occurrence of daily precipitation was treated through the Markov processes, while the modeling of the precipitated amount was performed by adjustment of the gamma function of probability. For validation of the adjustment, the Kolmogorov-Smirnov test was used. From the results it was established that this methodology can be applied to the simulation of synthetic series of daily rainfall for the study area.
In this paper we used WPT (Wavelet Packet Transform) and neural classifier SVM (Support Vector Machine) to recognize spoken digits from 0 to 9 in Brazilian Portuguese. The main objective this work is to find out the Wavelet mother that better represents the speech signal in Brazilian Portuguese. The results obtained were compared with MFCC (Mel frequency cepstral coefficients). We carried out sixteen experiments with different Wavelets in dependentcase and four experiments in independent-case. The database was recorded in three months with 82 eighteen-toforty years old male speakers. The SVM was used as a classifier in a "one vs. all" strategy. Best results have been obtained using Wavelets Daubechies 5, Meyer and Coiflet 5. Finally, we used a neural network MLP (Multi Layer Perceptron) in order to improve the SVM results.
Digital signal processing techniques have been used by acoustic analysis to evaluate the voice quality of the patient, due to the simplicity and non-invasive procedures for measurements.
This study presents a comparative analysis of wavelets, in order to find a descriptor that provides a better classification of voice pathologies. Different types of Wavelet Packet Transform were used as a tool for feature extraction and Support Vector Machine (SVM) to classify vocal disorders. Tests were conducted with 23 wavelets types in two SVMs, the first using the strategy "one vs. all" to classify normal and pathological voices and the second, using the strategy "one vs. one" to classify pathologies: edema and nodules. The best results were obtained using Daubechies family, especially Daubechies 5 (db5) wavelet.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.