Outline
- Abstract
- Highlights
- Keywords
- 1. Introduction
- 2. Review of Available Relationships
- 3. Database Development and Influential Parameters
- 4. Modeling Using Polynomial Neural Networks (pnn)
- 4.1. Data Division
- 4.2. Optimized Model Architecture
- 4.3. Weight Optimization (training)
- 4.4. Model Validation
- 5. Results and Discussion
- 5.1. the Proposed Equation
- 5.2. Parametric Study
- 6. Comparison with the Previous Models
- 7. Application of the Proposed Model in Liquefaction Analysis
- 7.1. Example 1
- 7.2. Example 2
- 7.3. Example 3
- 8. Summary and Conclusions
- References
رئوس مطالب
- چکیده
- کلیدواژه ها
- 1. مقدمه
- 2. بررسی نسبت های موجود
- 3. توسعه داده های پایه و مولفه های موثر
- 4. مدل سازی با استفاده از شبکه عصبی چند جمله ای (PNN)
- 4.1 بخش داده ها
- 4.2. معماری مدل بهینه
- 4.3. بهینه سازی وزن (آموزش)
- 4.4 اعتبار سنجی مدل
- 5.1 نتایج و بحث
- 5.2. مطالعه پارامتریک
- 6. مقایسه با مدل های قبلی
- 7. استفاده از مدل ارائه شده در تجزیه و تحلیل گداخته
- 8. خلاصه و نتیجه گیری
Abstract
Geophysical and geotechnical field investigations have introduced several techniques to measure in-situ shear wave velocity of soils. However, there are some difficulties for the easy and economical use of these techniques in the routine geotechnical engineering works. For the soil deposits, researchers have developed correlations between shear wave velocity and SPT-N values. In the present study, a new database containing the measured shear wave velocity of soil deposits have been compiled from the previously published studies. Using polynomial neural network (PNN), a new correlation has been subsequently developed for the prediction of shear wave velocity. The developed relationship shows an acceptable performance compared with the available relationships. Three examples are then presented to confirm accuracy and applicability of the proposed equation in the field of liquefaction potential assessment.
Conclusions
A new correlation was presented for shear wave velocity, Vs, of soil deposits as a function of corrected SPT blow counts, N1,60, and effective overburden stress, Vs0v . The correlation was developed using polynomial neural network (PNN) and a newly compiled database of shear wave velocity measurement including 10 different sites, 80 boreholes, and totally 394 data pairs. Prior to the model development, numerous existing equations, which were previously proposed for specified site and soil conditions, were examined via the compiled database. Significant scatter and errors were observed for the previous models because majority of them were just dependent to penetration resistance and ignore effective overburden pressure. In addition, most of the previous validated in depth by the comparisons made between the model predictions and measured shear wave velocity of some boring logs.
More precision of the PNN-based Vs model, compared with the previous equations, comes from the facts that SPT-N value cannot be sufficient to determine shear wave velocity and it should be accompanied with effective overburden pressure to enhance accuracy of the prediction. In fact, effective overburden pressure was found to be an important parameter for the estimation of shear wave velocity because the developed model, which considers this parameter, yields superior performance compared with the previously published equations. This finding is in accord with the small strain laboratory tests that have shown significant dependency of small strain shear modulus (Gmax) to effective stress; considering the fact that ð Þ Gmax and shear wave velocity are rigorously dependent together.