Parallel Traffic Flow Simulation of Freeway Networks: Phase 2
Anthony Theodore Chronopoulos
Abstract
Open-access reader
Anthony Theodore Chronopoulos
Abstract
Open-access reader
Explicit and implicit numerical methods for solving simple macroscopic traffic flow continuum models have been \nstudied and efficiently implemented in traffic simulation codes in the past. We have already studied and \nimplemented explicit methods for solving the high-order flow conservation traffic model. Implicit methods allow \nmuch larger time step size than explicit methods, for the same accuracy. However, at each time step a nonlinear \nsystem must be solved. We use the Newton method coupled with a linear iterative method (Orthomin). We \naccelerate the convergence of Orthomin with parallel incomplete LU factorization preconditionings. We \nimplemented this implicit method on a 16 processor nCUBE2 parallel computer and obtained significant execution \ntime speedup.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Explicit and implicit numerical methods for solving simple macroscopic traffic flow continuum models have been \nstudied and efficiently implemented in traffic simulation codes in the past. We have already studied and \nimplemented explicit methods for solving the high-order flow conservation traffic model. Implicit methods allow \nmuch larger time step size than explicit methods, for the same accuracy. However, at each time step a nonlinear \nsystem must be solved. We use the Newton method coupled with a linear iterative method (Orthomin). We \naccelerate the convergence of Orthomin with parallel incomplete LU factorization preconditionings. We \nimplemented this implicit method on a 16 processor nCUBE2 parallel computer and obtained significant execution \ntime speedup.
Key concepts: Flow (mathematics), Traffic flow (computer networking), Computer science, Transport engineering, Environmental science, Simulation, Engineering, Computer network