The control and optimisation of large-scale environmental facilities is gaining more and more in importance as a consequence of continually increasing demands. New, lower limiting values on discharges within wastewater treatment and the falling payment for renewable forms of energy such as biogas, solar thermal energy and photovoltaics make efficient plant operation absolutely essential. The use of modern control and optimisation strategies offers a very good alternative compared to conversion and redevelopment measures, which are often expensive or traditional control processes.
Intelligent processes from the field of computational intelligence (CI) offer, above all, the possibility of integrating existing expert knowledge (fuzzy logic) and various plant states (state-recording machines) in a new control strategy. In addition, successful optimisation strategies from the natural world such as the behaviour of flocks of birds or shoals of fish or even evolutionary biology using the processes of genetic algorithms (GA) and particle swarm optimisation (PSO) are imitated. Agent systems represent another possibility, where independent control components are responsible for different parts of the system that communicate and deal with one another in order to achieve the best possible condition for the plant.
The use of these methods on complex, non-linear processes has proved extremely useful in practice, as good to very good solutions for the most widely varying plant conditions can be quickly found.
In this connection, the iPCOIN software developed by the GECO►C team itself has been successfully used for several years for the process optimisation of wastewater treatment plants and biogas plants (cf. process-optimising software). Currently, both wastewater treatment plants and sewerage systems as well as biogas plants are being operated using these control and optimisation processes, as a result of which a clear increase in efficiency is achieved with a considerable saving in costs.
