Accelerated Dynamic Modeling for Complex Networks

Abstract

New developments in today's energy market have been trending towards large, complex production systems consisting of thousands of wells tied to one or more long trunklines; such developments include unconventional resources and geographically large-scale developments.During production, transient problems can occur in these complex systems that can have a negative impact on performance. These transient problems are best predicted and detected using transient multiphase flow simulator software. However, simulating these large complex networks in a transient simulator typically results in extended processing times. These large models often run slower than real-time and become impractical to use in online surveillance and monitoring applications or in forecasting scenarios. Engineers will not have enough time to react and mitigate the transient problems.To reduce the simulation time for these models, an integrated solution is required that captures the major transients of interest occurring in the complex networks while reducing the computational load. An integrated solution combines steady state and dynamic models building a representative asset model of the complex network.Breaking up the complex network into subnetworks where parts are solved with steady state simulators and others are solved with dynamic simulators is determined by observing the time constants of various parts of the network. The time constant is a measure of how quickly the system responds to changes in inputs. Selecting the time constant of interest determines the speed of solution. For all faster transients or smaller time constants, these subnetworks can be solved with steady state simulation. The subnetwork(s) of interest are modeled with a dynamic simulator.To combine the two simulation models into a single integrated asset model, boundary conditions between the transient and the steady state simulators should be in communication and synchronized at specific intervals to maintain the accuracy and fidelity of the overall model.Using an integrated solution for the simulation of large, complex networks can improve simulation time by an order of magnitude faster than a standalone transient simulation, while maintaining the accuracy of the results.This new integrated solution has proven to be very beneficial and practical for engineers since it allows fast, dynamic runs for complex networks, giving the user the ability to make timely decisions to mitigate operational problems and to run multiple, fast, sensitivity runs for different production scenarios.