Biogas Intelligence - operate biogas plants using Neural Network and Fuzzy logic

Biogas-plants are one part of using renewable energies and so a part of the chance to get over the energy crisis. Running biogas-plants needs some optimization. Because of microorganisms are not unique populations there will be indifferent demands on feeding and environmental conditions for the fermentation process. Living conditions are not well defined. Each plant in industrial scale has an own behavior, not comparable to another. At time the experience of the plant operator and watching the process by help of measuring data is the only one guaranty for successful results. Owing to this ambiguity of fermentation the biological processes are only limited controllable and presentable by functional equations. These conditions in mind, a controller is selected, who is able to find solutions. Our Neural Network Predictive Controller, a combination of Neural Network, Fuzzy logic and optimization seems to be the best solution to build a driver for biogas production. Fuzzy control is applied to identify diffuse trends and evaluate linguistic expert knowledge. So the whole complexity can be reduced.