In laboratory experiments, students apply basic control principles in order to enhance their understanding of the techniques taught in the lectures and tutorials. In particular, students learn how to use and configure standard sensors, control devices, programmable logic controllers (PLC), and small process control systems.

A list of all available experiments is given below.

The basics of feedback control are investigated for a system often occurring in chemical plants, a series of buffer tanks. The control objective is to adjust the level in the tanks to desired values, and to avoid overflow.

After manually controlling the levels in a first step, standard PI controllers have to designed. The advantages and drawbacks of the choice of parameters have to be explored.

This experiment serves to exercise the task of creating theoretical and experimental models of the heating system of a chemical reactor. These models are used to design a temperature controller. In order to check the performance of the controller, the latter is first tested by simulating the models, and then by implementing it on the real plant.

Based on dynamic modelling, a set of P- and PI-controllers is developed for a pressure control system. The positive and negative characteristics of the different controllers are investigated, and the most suitable selection has to be identified.

The scope of this experiment is to learn how to control a discontinuous production system via a process control system. The essential steps of the experiment are to precisely formulate the desired production sequence, to design an appropriate sequential (logic) controller, to implement the latter on the control system, and to run the controlled plant. The plant is equipped with a Siemens S7 Programmable Logic Controller, a system that is very common in industry.

The scope of this experiment is to learn how to control a discontinuous production system via a process control system. The essential steps of the experiment are to precisely formulate the desired production sequence, to design an appropriate sequential (logic) controller, to implement the latter on the control system, and to run the controlled plant. The plant is equipped with a Siemens S7 Programmable Logic Controller, a system that is very common in industry.

By usage of various flow measurement devices, the difficulties of accurate and reliable measuring are explored. Measurements are never exactly precise, but are afflicted with errors depending on the method of measurement, the selected measurement scale and other random and systematic factors. These effects are investigated for the procedure of filling a vessel exactly with a specified amount of water.

Chemical engineering processes often encounter numerous situations where the correct value of parameter estimation plays a very important role in the operation. This parameter may often be dependent on different physical and chemical properties that keep changing during reactions. The aim of this experiment is to model a simple non-isothermal reactor and then use temperature measurements to estimate different parameters of the process. Two approaches for designing observers, namely 'Calorimetry' and 'Luenberger observer' are used and compared in this experiment.

In the chemical industry, scheduling plays a major role in production planning and the organization of material flows. In production planning the objectives are the achievement of a high plant efficiency and the keeping of delivery times. In multi-product plants this leads to complex planning problems. The goal of this experiment is to illustrate fundamental concepts of scheduling problems by means of a small pipeless plant. In the plant, the "reactors" consists of movable tanks which are transported between a storage area and stations, where filling, mixing, and cleaning tasks take place by a single transport unit. Every station executes one production step. The plant is complex enough, so that the scheduling problems cannot be solved trivially. The scheduling problem at hand is to execute several production orders at the same time under the objective of completion time minimization.

In this computer experiment, a tubular reactor is simulated with a focus on the change of concentration of the reacting component and the change of temperature. The model involves partial differential equations since changes in time as well along the reactor have to be considered. For numerically solving the differential equations, the tool Matlab/Simulink is used. Since Simulink is restricted to the solution of ordinary differential equations (ODEs; only time derivatives), the partial derivatives have to be discretised before the model can be simulated with standard algorithms for ODEs. The result of the simulation are plots of the evolution of concentration and temperature over time and length of the reactor.

Learn2control is a joint project of five German universities. The intention of this project is to teach control techniques based on 6 laboratory experiments that are accessible over the internet. Three of the examples are currently available:

This experiment (contributed by the chemical engineering department of the University of Dortmund) comprises the design of a controller for the heating system of a reactive distillation column. Rigorous process models have to be developed for this system and serve as basis for the design of suitable controllers. The rigorous process models are also used for the validation of the controllers.

The goal of this project is to provide an opportunity to apply theoretical knowledge and to gain experience in basic principles of methods for the analysis of dynamic systems. An illustrative example of a simple continuous stirred reactor system is considered in which an equilibrium reaction of two species takes place. The first task is to describe the dynamic system by means of three differential equations for the state variables for the concentrations for both substances and for the volume of the content of the reactor. The next step is to calculate the steady state of the system as a prerequisite of a subsequent linearisation at this point. By consideration of the eigenvalues of the resulting Jacobian matrix, a statement about the stability of the system can be made. Thereafter a transfer function for the system at hand must be computed from the previous linearisation. For the resulting transfer function the root locus diagram has to be computed. The graph of the root locus and step responses can also be plotted. In the given example a zero-pole cancellation can be regarded.