Dynamic simulators for improved plant performance

Paul Boughton

Niclas Krantz focuses on some very useful and recently developed simulator applications from power utilities. He examines three different simulator applications, very useful in their applications respectively. They are all developed for power utilities and power houses in process industries like pulp mills with their own power generation.

Computer simulation of technical systems has a long history. The military, automotive, aircraft and power industry have all been using advanced simulators in design and training for decades. Earlier, these applications have been subject to use only of an exclusive group of people, though. The development of personal computers and simulator platforms has made it possible to extend the use of simulation. Many more engineers can gain the benefits of this fantastic technology and more applications can help increase the engineering efficiency. Advanced simulators can be built up even without computer programming expertise.

These days it is quite easy to develop advanced simulator applications, even with complex dynamic behaviour. Many modelling tools are very intuitive and often contain advanced models with default settings. This gives tremendous possibilities to achieve advanced 'easy to use' simulators in a short time.

On the other hand there is a great risk that a simulator that has a professional appearance has no correlation with the plant or system it represents, unless the right parameters are put into the model. Another danger is if the user does not know in what operating range the model is valid.

The engineers developing the simulators have a big responsibility to achieve the proper accuracy by using correct parameters, by validating the models, and by describing the accuracy and valid operating range of the simulator.

Due to these pitfalls it is also essential that every buyer of a simulator defines detailed requirements for the simulator including requested operating range and accuracy in all interesting aspects.

When an industrial plant is being modernised and vital systems like overall control systems, turbines or boilers are replaced, it is essential that the new equipment is carefully designed and integrated in the existing plant. The experience from several such projects has proven that the investment in a simulator as a development tool has been very profitable.

The benefit of using dynamic simulator technology is that it can be used through the whole project from the design phase to commissioning and training:

- Conceptual design of components and control.

- Analysis of extreme operating conditions and faults.

- Tuning of control systems to ensure proper function before commissioning.

- All involved professions, from automation engineers to operators can gather around the simulator and gain common understanding of the project.

- The simulator can be used for training of operators and maintenance personnel.

An example is given here from one of many projects where a simulator has been used. The work was initiated when new turbines were purchased and the steam control system was modernised at a major pulp and paper mill. A simulator was developed in order to get the best possible plant performance. The model of the existing system was first developed so that validation could be made by means of measurements and co-operation with the plant operators. After that, the new components were added to the model. Control optimisation was performed in two steps.

1. An efficient control strategy was developed by conceptual studies.

2. Primary control loops were tuned in the simulator to handle all the operational conditions set up in the requirements. The events included turbine trips, island operation, instant changes in steam consumption etc.

The results of these efforts were:

- The commissioning and the plant startup to full production became very short.

- Major parts of the control system changes could also be made without plant shutdown, since the control was already optimised in the simulator.

- Stable operation with very few outages and large amounts of energy savings.

Furthermore the simulator was used for operator training so that the plant staff knew what to expect when the new systems were started. The payback time for the simulator and engineering was estimated to a few months.

Simulator based operator training

The simulator technology is well suited for operator training. If a simulator has been developed in the engineering process, it can be used also for training.

Specific applications can also be developed only for training and at a reasonable price. In the Fig.3 such a simulator is shown, where the power system dynamic behaviour and interaction is trained.

The benefit with this technology is that you can mix staff with a variety of backgrounds and knowledge and still be able to give all a good training and understanding of the physical behaviour of the system. The user meets a realistic user interface due to the following:

- Operating screens can be made identical to the real control room screens

- Simulator runs in real time

- Teacher station enables introduction of faults and trips.

Plant testing with 'real time' simulator

Simulators are very useful in engineering, but in some cases it is very hard to make models that are accurate enough. After all, a model still represents an ideal world of known circumstances. In order to represent the reality a model needs to be carefully validated. Only real tests can reveal deficiencies of a model. On the other hand, testing is not always possible due to risks of failure and high costs.

One situation, where tests are hard to make and a simulator is insufficient, is island operation of power plants.

The evaluation of the island operation performance cannot be trusted using only simulation because every plant responds individually. Testing of island operation is often hard to perform due to safety reasons and high costs.

An excellent method is to use simulation in combination with testing. By using a real time simulator connected to a power plant HardWare In the Loop, these tests can be made safely and with very accurate results. A model with high accuracy is used in combination with the real plant, which is the unit to be evaluated. An example of such test equipment is SolvSim Power Station, which is connected to the frequency governor and simulates an island network. The governor acts as in real island mode while the machine still runs synchronised to the main grid. The plant and the simulator are running together in real-time in closed loop operation. This eliminates the risk for network (customer) disturbances and drastically improves the possibility to perform proper tuning of the control systems. Furthermore the results can be used for validation of power plant models.

Enter √ at www.engineerlive.com/ipe

Niclas Krantz is managing director, Solvina, Gothenburg, Sweden. www.solvina.se

Recent Issues