Existing actively trained meta-models, which present a new promising direction in reliability analysis, are not applicable to networks due to the inability of these methods to handle high-dimensional problems as well as discrete or mixed variable inputs. However, analyzing systems’ reliability using computationally expensive flow-based models faces substantial challenges, especially for rare events. The numerical results highlight the effectiveness of the proposed approaches in improving the accuracy and efficiency of the reliability analysis.įlow network models can capture the underlying physics and operational constraints of many networked systems including the power grid and transportation and water networks. The proposed approaches are then applied to a simulation-based study of the two-terminal connectivity in a benchmark transportation network subject to an extreme earthquake event. The gain in accuracy and efficiency is achieved by a more effective exploration of the sample space lead to an accelerated convergence of Monte Carlo simulations. To overcome this challenge, this paper presents fast approaches for: (i) uncertainty quantification in the modeling of natural disasters, in order to improve the accuracy of system response calculations, and (ii) infrastructure system model reduction, based on the principal component analysis, in order to speed-up computations of the system response. While numerous research efforts have addressed and quantified the impact of natural disasters on transportation systems, they still suffer from high computational cost, impairing the accuracy and efficiency of the application of these approaches on large networks. Therefore, it is essential to develop tools that provide accurate and efficient evaluation of the system reliability to facilitate optimal decision making for mitigation, preparedness, response, and recovery practices. I strongly recommend carry at least 30 infiltration trials to evaluate Ksat mean and variance.Natural disasters can have catastrophic impacts on the functionality of transportation systems and cause severe physical and socio-economic losses. I have calculated nearly 1200 samples to evaluate Ksat within 95% confidence and 10% of error in Loamy Oklahoman Soils. Unfortunately the number of trials to conducto with a good degree of precision is quite large. To evaluate the mean and varianceof Ksat go tothe literature for the soil type you are trying to evaluate or conduct a set of experiments to evaluate these parameters. Just follow the preliminar recommendation and then you will have random Ksat numbers. I guess SAS can generate Log-Normal random number with a given Mean and a Variance but I have not done so. You can use one of the several routines Excell or SAS has integrated into the system. Re-transform back to the original units and there you have Log-Normal Ksat data. Generate uniform random numbers and transform them into the normal distribution function by using the mean and variance of the logarithmic data. Once the data is normalized evaluate the mean and variance. By transforming the Ksat data into a logarithmic function you will normalize the data. The Log-Normal distribution function is a special case of the Normal distribution. The Saturated Hydraulic Conductivity, Ksat, parameter of soil follows usually a Log-Normal Distribution.
0 kommentar(er)
