A framework for runtime verification of industrial process control systems

Despite the significant research on automation systems verification, further work is needed on methods for the industrial scale verification of automated processes that are able to cover control software errors, sensor and actuator errors as well as design errors in the process to be controlled. In this paper, such an approach is based building on previous work on runtime verification using plant simulation models created during the plant design process. A framework for runtime verification is proposed based on keyword driven testing that is applied to the entire automated process rather than only to control software. Through the use of ontologies, the framework is made independent of any particular simulator. The approach is demonstrated with a case study from the mineral processing industry.