In this study the behavior of rectangular footing on gypeous soil was studied under inclined and eccentric loading. The experiments were performed using laboratory scale rectangular footing rested on soil taken from Tikrit University site in Salah Al Din province under 3 m depth which has a gypsum content of (50.48 %). The load test was performed on rectangular footing at eccentricity of (e/B= 0, 0.1, 0.2 and 0.3) and an angle of inclination of the load with the vertical ( i= 0°, 5°, 10°, 15°and 20°). The local specifications of the soil are measured (density, moisture, maximum density and optimum moisture content), it is found that that the vertical settlement, horizontal displacement, and base inclination increases with the increasing of eccentricity and inclination of load, the values of bearing capacity that getting in this study was less than of the previous theoretical studies when the load was vertical, and is given a good agreement when load was inclined and field density and moisture of soil. The values of bearing capacity was decreased when the load eccentricity increased because of the effective area became small. It is found that a high settlement occur in footing when a water (unsaturated with gypsum salts) diffuses through the soil, then gypsum become soluble thereby the soil resistance decreases because of rupturing of chemical bond between gypsum and soil.
The mechanical behaviour of partially saturated soils can be very different from that of fully saturated soils. It has long been established that for such soils, changes in suction do not have the same effect as changes in the applied stresses, and consequently the effective stress principle is not applicable. A procedure was proposed to define the soil water characteristic curve. Then this relation is converted to relation correlating the void ratio and matric suction. The slope of the latter relation can be used to define the H-modulus function. This procedure is utilized in the finite element analysis of a footing on unsaturated coarse grained soil to investigate its bearing capacity. The finite element results demonstrated that there is a significant increase in the bearing capacity of the footing due to the contribution of matric suction in the range 0 to 6 kPa for the tested compacted, coarse-grained soil. The ultimate pressure increases from about 120 kPa when the soil is fully saturated to about 570 kPa when the degree of saturation becomes 90%. This means that an increase in the bearing capacity of about 375% may be obtained when the soil is changed from fully saturated to partially saturated at a degree of saturation of 90%. This development in the bearing capacity may exceed 600% when the degree of saturation decreases to 58%.
Collapse of gypseous soils may cause excessive settlement and serious damage to engineering structures. Various improvement approaches, such as mechanical techniques and chemical additions, have been used to reduce the collapsibility of these soils. The odometer test has traditionally been used to assess the collapsibility of the improved gypseous soils; however, because the small size of test specimens, this method may not adequately reflect field conditions. In this research, a laboratory model test of 600 x 600 x 600 mm with a model footing of 100 x 100 mm was developed to measure the collapse characteristics of a gypseous soil. The top layer underneath the footing was improved by compaction, cement kiln dust (CKD), geogrid, and a combination between CKD and geogrid. The top layer was improved at two values of thickness of 50 and 100 mm. The results obtained from this study indicate that the values collapsibility settlement reduction factor for compacted soil and the soil treated with CKD were 75 and 82%, 89% receptively. These values increased up to 95 % when a combination of CKD and geogrid was applied. As discussed herein, the aforementioned treatment methods can effectively be used to improve the collapsibility of gypseous soils.
This study was involved with the diffferent types of cracks and reptures that may occur in structures, also a practical case for a school building in Al – Ramadi city is presented in this work. Necessary and instantenuos ways for treatments were suggested in this study.It was obvious that this building needs to support footing .In addition, supporting the columns and concrete beams was included in the present work
This study was involved with the diffferent types of cracks and reptures that may occur in structures, also a practical case for a school building in Al – Ramadi city is presented in this work. Necessary and instantenuos ways for treatments were suggested in this study.It was obvious that this building needs to support footing .In addition, supporting the columns and concrete beams was included in the present work
Abstract:This study explores the potential of back propagation neural networks (BPNN) computingparadigm to predict the ultimate bearing capacity of shallow foundations on cohesionlesssoils. The data from 97 load tests on footings (with sizes corresponding to those of realfootings and smaller sized model footings) were used to train and validate the model. Fiveparameters are considered to have the most significant impact on the magnitude ofultimate bearing capacity of shallow foundations on cohesionless soil and are thus used asthe model inputs. These include the width of the footing, depth of embedment, length towidth ratio, dry or submerge unit weight and angle of internal friction of the soil. Themodel output is the ultimate bearing capacity. Performance of the model wascomprehensively evaluated. The values of the performance evaluation measures such ascoefficient of correlation, root mean square error, mean absolute error reveal that themodel can be effectively used for the bearing capacity prediction. BPNN model iscompared with the values predicted by most commonly used bearing capacity theories.The results indicate that the model perform better than the theoretical methods.KEYWORDS: Ultimate bearing capacity; Shallow foundations; cohesionless soil; backpropagation neural network (BPNN); prediction