• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.4
• /
• pp.1982-1987
• /
• 2011

In this study, we proposed a new method that can estimate viscosity curves of unknown samples or high viscous resins like LFT(Long Fiber Thermoplastics). First, we built the system that could detect the pressure of melt during filling the cavity in a mold. It consists of both pressure sensors which are installed in a mold and the Kit which can convert analog signal to digital signal. The kit measures the melt pressure in mold cavity. We could also simulate the cavity pressure during filling process with commercialized CAE softwares(ex, Moldflow). If the viscosity data in CAE Database were correct, the simulated pressure profile coincided with the measured one. According to our proposed algorithm, we obtained correct viscosity data by iterating the process of comparing the simulated profile with the measured one until both coincided each other. In order to verify this algorithm, we selected well-defined PP resin and concluded that the experimental profile comply with the CAE profile. We could also estimate the optimized viscosity curves for PP-LFT by applying our method.

• Journal of Korean Society of Transportation
• /
• v.23 no.2
• /
• pp.143-150
• /
• 2005

Estimation of trip distribution, estimated O-D matrix must satisfy the condition that the sum of trips in a row should equal the trip production, and the sum of trips in a column should equal the trip attraction. In most cases the iterative calculation for convergence is needed to satisfy this condition. Most of all present convergence of iterative methods may results a big difference between estimated value and converged value, and from this, the trip distribution patterns may be changed. This paper presents a new convergence of iterative method that comes closer to meeting the convergence condition and gives the maximum likelihood estimation for calculating a distribution patterns from the trip distribution estimation model. The newly developed method differs from existing methods in three important ways. First, it simultaneously considers both the convergence condition and the distribution patterns. Second, it computers simultaneous convergence of rows and columns instead of iterating respectively. Third, instead of using the growth rates to the trip production, trip attraction, it uses the differences between trip production and sum of trips in a row, and trip attraction and sum of trips in a column. Using 38 by 38 O-D matrix, this paper compared the Fratar method and the Furness method to the newly developed method and found that this method was superior to the other two methods.

• Nuclear Medicine and Molecular Imaging
• /
• v.43 no.5
• /
• pp.459-467
• /
• 2009

Purpose: The maximum likelihood-expectation maximization (ML-EM) is the statistical reconstruction algorithm derived from probabilistic model of the emission and detection processes. Although the ML-EM has many advantages in accuracy and utility, the use of the ML-EM is limited due to the computational burden of iterating processing on a CPU (central processing unit). In this study, we developed a parallel computing technique on GPU (graphic processing unit) for ML-EM algorithm. Materials and Methods: Using Geforce 9800 GTX+ graphic card and CUDA (compute unified device architecture) the projection and backprojection in ML-EM algorithm were parallelized by NVIDIA's technology. The time delay on computations for projection, errors between measured and estimated data and backprojection in an iteration were measured. Total time included the latency in data transmission between RAM and GPU memory. Results: The total computation time of the CPU- and GPU-based ML-EM with 32 iterations were 3.83 and 0.26 see, respectively. In this case, the computing speed was improved about 15 times on GPU. When the number of iterations increased into 1024, the CPU- and GPU-based computing took totally 18 min and 8 see, respectively. The improvement was about 135 times and was caused by delay on CPU-based computing after certain iterations. On the other hand, the GPU-based computation provided very small variation on time delay per iteration due to use of shared memory. Conclusion: The GPU-based parallel computation for ML-EM improved significantly the computing speed and stability. The developed GPU-based ML-EM algorithm could be easily modified for some other imaging geometries.