Previsão de recalques por colapso de sapatas assentes em solo natural e compactado
Carregando...
Data
2018-01-26
Autores
Soares, Fábio Visnadi Prado
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O comportamento dos solos não saturados é um tema amplamente estudado, tanto por meio de ensaios de laboratório quanto de campo. Fenômenos como o colapso dos solos são relativamente bem explicados pelos modelos elastoplásticos existentes, dos quais o Modelo Básico de Barcelona é um dos mais utilizados e difundidos no meio acadêmico. Apesar dos avanços obtidos, ainda existe certa escassez de estudos que comparem o comportamento previsto por tais modelos a resultados de situações reais de engenharia. Esta pesquisa teve como objetivo simular numericamente resultados de recalques devidos a carregamentos e inundação de provas de carga em placa e de protótipos de sapata construídas sobre solo natural e compactado em cava. Com os resultados buscou-se analisar a eficiência da técnica de compactação para mitigar o problema do colapso do solo. Para isso, ensaios de compressão triaxial e edométricos com sucção controlada, bem como ensaios de curva de retenção e condutividade hidráulica foram realizados para determinação dos parâmetros constitutivos hidráulicos e mecânicos do solo natural e compactado que foram utilizados no programa Code_Bright para simulação numérica dos resultados de campo. Os resultados demonstraram a capacidade do modelo em reproduzir o comportamento do solo no campo. Demonstraram também que o uso da compactação do solo que preconiza compactar em cava uma camada de espessura equivalente à largura da sapata é viável para redução da colapsibilidade, mas depende do nível das tensões atuantes nas fundações e da tolerância aos recalques. Tensões propagadas para a camada de solo natural superiores à tensão de pré-adensamento do solo natural saturado provocarão recalques por colapso quando o solo for umedecido.
The behaviour of unsaturated soils is a widely studied theme around the world, both by utilising laboratory and field tests. Phenomena like soil collapsibility are relatively well explained by the existing elastoplastic models, among which the Barcelona Basic Model is the most widely used in the academic field. Despite the recent advances, there is still a scarcity of papers that compare the behaviour predicted by constitutive models and results of real engineering problems. This research aims to numerically simulate the results of settlement due to loading and wetting of load tests built both on undisturbed and compacted soil. The results were analysed in order to study the efficacy of the soil compaction technique in mitigating soil collapse. Suction controlled triaxial compression and oedometer tests, as well as soil water retention curve and hydraulic conductivity experiments were conducted to determine the hydraulic and mechanic parameters used in the software Code_Bright to numerically simulate field results. The results highlight the capability of the model in reproducing the analysed field behaviour. They also show that the technique that involves compaction of the soil equal to the footing’s width is viable to reduce its collapsibility, but its efficacy will also depend on the tensions applied on the foundation and on the tolerable settlement values. Tensions that are propagated to the natural soil layer that are higher than its preconsolidation stress will cause collapse settlements and wetted.
The behaviour of unsaturated soils is a widely studied theme around the world, both by utilising laboratory and field tests. Phenomena like soil collapsibility are relatively well explained by the existing elastoplastic models, among which the Barcelona Basic Model is the most widely used in the academic field. Despite the recent advances, there is still a scarcity of papers that compare the behaviour predicted by constitutive models and results of real engineering problems. This research aims to numerically simulate the results of settlement due to loading and wetting of load tests built both on undisturbed and compacted soil. The results were analysed in order to study the efficacy of the soil compaction technique in mitigating soil collapse. Suction controlled triaxial compression and oedometer tests, as well as soil water retention curve and hydraulic conductivity experiments were conducted to determine the hydraulic and mechanic parameters used in the software Code_Bright to numerically simulate field results. The results highlight the capability of the model in reproducing the analysed field behaviour. They also show that the technique that involves compaction of the soil equal to the footing’s width is viable to reduce its collapsibility, but its efficacy will also depend on the tensions applied on the foundation and on the tolerable settlement values. Tensions that are propagated to the natural soil layer that are higher than its preconsolidation stress will cause collapse settlements and wetted.
Descrição
Palavras-chave
Solos não saturados, Solos colapsíveis, Simulação numérica, Unsaturated soils, Collapse, Strip-footing, Numerical simulation, Soil compaction