Search

Cracking is one of the main problems in concrete structures and is affected by various parameters. The step-by-step laboratory method, which includes casting specimens, curing for a certain period, and testing, remains a source of worry in terms of cost and time. Novel machine learning methods for anticipating the behavior of raw materials on the ultimate output of concrete are being introduced to address the difficulties outlined above such as the excessive consumption of time and money. This work estimates the splitting-tensile strength of concrete containing recycled coarse aggregate (RCA) using artificial intelligence methods considering nine input parameters and 154 mixes. One individual machine learning algorithm (support vector machine) and three ensembled machine learning algorithms (AdaBoost, Bagging, and random forest) are considered. Additionally, a post hoc model-agnostic method named SHapley Additive exPlanations (SHAP) was performed to study the influence of raw ingredients on the splitting-tensile strength. The model’s performance was assessed using the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). Then, the model’s performance was validated using k-fold cross-validation. The random forest model, with an R2 of 0.96, outperformed the AdaBoost models. The random forest models with greater R2 and lower error (RMSE = 0.49) had superior performance. It was revealed from the SHAP analysis that the cement content had the highest positive influence on the splitting-tensile strength of the recycled aggregate concrete and the primary contact of cement is with water. The feature interaction plot shows that high water content has a negative impact on the recycled aggregate concrete (RAC) splitting-tensile strength, but the increased cement content had a beneficial effect.

A kind of beam-splitting-based single-shot phase retrieval is proposed, where the transmitted field of a sample to be observed is diffracted into many replicas by a Dammann grating and then incident on a weakly scattering phase plate with a known structure, and the exiting beams propagate roughly along their original directions and form a diffraction pattern array on the detector. All sub-diffraction patterns isolated with each other were recorded with a single measurement, and the complex amplitude of the radiation incident on the grating and accordingly transmitted from the sample can be iteratively reconstructed from the recorded diffraction pattern array. Since the weakly scattering plate is accurately measured in advance, fast convergence and high accuracy can be achieved with a small number of sub-diffraction patterns. The feasibility of this proposed method is verified with experiments using visible light.

A special optical alignment is adopted and corresponding reconstruction algorithm is developed to reduce the reconstruction error induced by the hysteresis or backlash error of the translation stage in Ptychographical Iterative Engine (PIE) imaging with weak scattering specimen. In this suggested method, the positions of the scanning probe are determined directly from the recorded diffraction patterns rather than from the readout of the stage meter. This method not only remarkably improves the reconstruction quality, but also completely lowers the dependency of PIE on the device accuracy and accordingly enhances its feasibility for many applications with weak scattering specimen.

With the rapid development of the modern city, the urban waterfront landscape has become the window of the urban development and epitome of the society. This thesis is focused on the participation and experience of people. Based on the landscape conceptual design of the city center of Liaobin Water Town in Panjiang City, Liaoning Province, this thesis analyzes the enlightenment and significance of Riverside Scene at Qingming Festival to modern urban waterfront landscape through various researches and discussions.

The recently developed Light Field Moment Imaging (LMI) is adopted to show the stereoscopic structure of the sample studied in Coherent Diffractive Imaging (CDI), where 3D image were always generated with complicated experimental procedure such as the rotation of the sample and time-consuming computation. The animation of large view angle can be generated with LMI very quickly, and the 3D structure of sample can be shown vividly. This method can find many applications for the coherent diffraction imaging with x-ray and electron beams, where a glimpse of the hierarchical structure required and the quick and simple 3D view of object is sufficient. The feasibility of this method is demonstrated theoretically and experimentally with a recently developed CDI method called Ptychographic Iterative Engine.