A prediction model of loess collapsibility base on the method of fuzzy information optimization processing

For establishing a prediction model of collapsibility loess for central region of Shanxi Province, 48 groups of loess collapsibility test data have been collected from Taiyuan-Jinzhong area. The mechanism of collapsibility were discussed and based on that the moisture content,void ratio and compressibility were chosen form the test data to be the basic parameter for establishment of prediction model by the method of fuzzy information optimization processing. 40 groups of test data were used to establish the model and 8 groups of test data were used to validate the correctness and applicability of the prediction model. From the validation, the correctness and applicability loess collapsibility prediction model has been confirmed. This method is simple, efficient and practical, which can be applied to evaluate the collapsibility of the loess in Taiyuan-Jinzhong area for future construction engineering.