Estimação da densidade de solos utilizando sistemas de inferência fuzzy
Carregando...
Data
2007-12-03
Autores
Benini, Luiz Carlos [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Este trabalho tem por objetivo principal apresentar o desenvolvimento de um sistema inteligente, utilizando a Teoria Fuzzy, para estimar valores aproximados da densidade do solo a partir de medidas diretas (campo) sem a necessidade de ensaios laboratoriais e, consequentemente, identificar a compactação do solo por meio destes valores estimados. A densidade do solo é um dos principais parâmetros utilizado para a identificação do grau de compactação do solo, e está relacionada com outros parâmetros tais como a resistência à penetração do solo, o teor de água e a textura do solo. Para o desenvolvimento do trabalho foram considerados três parâmetros do solo: a resistência à penetração representado pelo índice de cone (em kPa), o teor de água dado pela umidade do solo (em porcentagem, %), e a textura dada pela quantidade de argila presente no solo (em porcentagem, %). Foram, ainda, considerados solos preparados (passagem de arado, de grade, de escarificador, e outros) e solos não preparados (nenhum tipo de preparado ou em solo de plantio direto). Segundo a porcentagem de argila no solo, estes foram divididos em solo tipo I (teor de argila menor que 30%), solo tipo II (teor de argila entre 30% e 50%), solo tipo III (teor de argila maior que 50%) para o solo não preparado, e solo tipo I (teor de argila menor que 30%) e solo tipo III (teor de argila maior que 50%) para o solo preparado. O modelo matemático proposto para determinar as estimativas da densidade do solo foi desenvolvido com base em dados experimentais representados pelas três características do solo: índice de cone, umidade e argila. Utilizando os dados experimentais os modelos foram identificados por meio de um algoritmo neuro-fuzzy, em função da resistência à penetração, teor de água e textura do solo, onde se pode analisar a densidade do solo para os distintos valores das variáveis de entradas...
The present work aims to develop a intelligent system using fuzzy theory in order to estimate approximate values for the soil density taking in account direct measurements (in loco) disregarding laboratorial essays and, consequently, to identify the compactation of the soil through those estimated values. The soil density is one of the main parameters used to identify the soil compactation level, and it is also related to other parameters such as resistance to the soil penetration, water content and soil texture. Three soil parameters were considered for the development of this work: resistance to the soil penetration represented by the cone index (in kPa), the water content given by the soil humidity (percentage, %), and the texture given by the quantity of clay present in the soil (percentage, %). Also, prepared soils were considered (plough step, grid, disk harrow, and others) as well as non prepared soils (no kind of soil preparation or direct planted soil). According to the percentage of clay in the soil, they were classified as soil type I (clay content less than 30%), soil type II (clay content between 30% and 50%), soil type III (clay content higher than 50%) for the case of non prepared soil. For the case of prepared soil it was considered only soils type I (clay content less than 30%) and type III (clay content higher than 50%). The mathematical model considered to estimate the soil density was developed on the basis of given experimental data having the three soil characteristics: Cone index, humidity and clay content. Using the experimental data the models were identified by means of a neuro-fuzzy algorithm in function of the resistance to the penetration, water content and soil texture, through which one can analyze the soil density for different values of the model entrance variables. The experimental data and the estimated ones by the model...(Complete abstract click electronic access below)
The present work aims to develop a intelligent system using fuzzy theory in order to estimate approximate values for the soil density taking in account direct measurements (in loco) disregarding laboratorial essays and, consequently, to identify the compactation of the soil through those estimated values. The soil density is one of the main parameters used to identify the soil compactation level, and it is also related to other parameters such as resistance to the soil penetration, water content and soil texture. Three soil parameters were considered for the development of this work: resistance to the soil penetration represented by the cone index (in kPa), the water content given by the soil humidity (percentage, %), and the texture given by the quantity of clay present in the soil (percentage, %). Also, prepared soils were considered (plough step, grid, disk harrow, and others) as well as non prepared soils (no kind of soil preparation or direct planted soil). According to the percentage of clay in the soil, they were classified as soil type I (clay content less than 30%), soil type II (clay content between 30% and 50%), soil type III (clay content higher than 50%) for the case of non prepared soil. For the case of prepared soil it was considered only soils type I (clay content less than 30%) and type III (clay content higher than 50%). The mathematical model considered to estimate the soil density was developed on the basis of given experimental data having the three soil characteristics: Cone index, humidity and clay content. Using the experimental data the models were identified by means of a neuro-fuzzy algorithm in function of the resistance to the penetration, water content and soil texture, through which one can analyze the soil density for different values of the model entrance variables. The experimental data and the estimated ones by the model...(Complete abstract click electronic access below)
Descrição
Palavras-chave
Solos - Compactação, Solos - Densidade, Sistema de Internferencia fuzzy, Fuzzy inference systems, Soil density, Cone index, Water content, Soil texture, Soil compactation
Como citar
BENINI, Luiz Carlos. Estimação da densidade de solos utilizando sistemas de inferência fuzzy. 2007. xv, 194 f. Tese (doutorado) -Universidade Estadual Paulista, Faculdade de Ciências Agronômicas de Botucatu, 2007.