Oil and Gas

Process Simulation

Since the middle of the 1980s the Institute has developed and delivered simulation models and simulation technology for processing facilities in the petroleum industry. This work is performed in the Process and Fluid Flow Department in the Petroleum Sector.

Sendstad, Olav

Specialist Group Manager


The most important application areas for process models are described here. Information on the methods and technologies for process simulation are given under section Simulation Technology.

The Production Process
Oil and gas flow from the reservoirs through production wells and subsea installations to facilities for treatment and transport to the market, either via production platforms or production ships, or directly to onshore processing plants. The facilities have equipment for removal of water and chemicals, separation of gas and oil, heating, cooling, pumping and compression, as well as auxiliary systems for this main process. Due to the complexity of such plants, it is very useful to apply mathematical simulation models for various types of analyses as described below.

Models are used for operator training
Good training of the personnel to operate the processes is very important. A training simulator is an efficient tool to teach the operators how to handle the process. In addition to learning in the regular plant operation, one can train especially for more demanding operations like startup and shutdown. Malfunctions and accidents that occur seldom in real life can be initiated in the simulator, so that the operator is trained to act correctly should the event really happen. The consequences of an operator error in the real process may be serious, but in the simulator it is without cost.

Models are used for planning and design
The simulation models are sufficiently accurate and detailed to be very useful during the planning and design of the facilities. Possible bottlenecks in the plant can be detected, and the control system can be verified and tuned before the plant startup. The simulator may uncover problems and poor design at an early stage. When such problems are detected later, when the plant has been built, the cost to remedy the problems will be much, much higher. The simulator may be used to prepare for the plant startup, in order to save time and avoid unforeseen events.

Models are used for operational support
A model is a tool for achieving the best possible operation of the facilities. Detailed process models, which traditionally are used for engineering and for building training simulators, are more often being used also in the daily operations. The so-called “E-field concept” means to make measured process information available at the right place to the right time for the persons needing them, but just as important are efficient tools to organize and evaluate this large flow of information. By feeding data into models and by comparing the simulation results with the measurements, the operators and engineers will have a better basis for their decisions.

Models are used to compute information that cannot be measured
In spite of improvements in process instrumentation, there are many process data that still cannot be measured. Instrumentation may be too expensive or not sufficiently robust. The mathematical model may save the situation. Some data are given as input to the model, other data may be used to calibrate the model, and the model responds by calculating many quantities that cannot be directly measured. An example is the computation of the gas and liquid flow in long transport pipelines, where normally only measurements at the pipeline inlet and outlet are available.

Models warn about abnormal process conditions
A model that is directly connected to the plant instrumentation, can warn about abnormal process behavior. The model describes the relation between measurements and the physical laws that apply to the process, and will inform when the instruments give inconsistent information and when the model predicts a different result than what is observed. Deviations between the measurements and the model can give important information about the need for maintenance and repairs.

Models are used for predictions
A model that is initiated with a known process state can be used to simulate how the process will behave in the future, from seconds to hours and days. In this way one will be able to foresee undesirable developments, and one may use the simulator to evaluate various actions before they are applied.

Models are used for advanced process control
Advanced control systems can use built-in mathematical models. An example is the so-called “Model Predictive Control”, where the model itself is a central element in the controller’s calculations. Such methods have already been tested in systems for stabilization of oscillating production wells.

Models are used to optimize production
The offshore production systems are becoming more complex all the time, and its operation calls for better coordination. Simulation models of the total systems assist in keeping track of the system state and may be used for various types of optimization tasks.