New approach for data evaluation of industrial wastewater treatment plant: methodology and case study for a coke and steel-processing plant.

Environmental monitoring of biological wastewater treatment plants (BWWTP) treating industrial effluents produces large amount of data. Frequent sampling is done in the influent and effluent but also in intermediate points. Samples are analyzed for classical and specific contaminants and physical-chemical parameters are monitored. In this paper data from a BWWTP treating the effluents of a coke and steel-processing factory are analyzed. Due to a complex situation, this BWWTP gave poor performances that did not match environmental regulations, meanwhile upgrading proved to be uneasy. Data analysis using principal component analyses (PCA) or kinetic modeling with a Haldane model was unsuccessful in handling these data, which was attributed to undetermined toxic effects. A new methodology is reported, that allowed to identify a kinetics for thiocyanate degradation and a relation between pH and toxic effects. This analysis of the plant data allowed to make hypothesis on the process control parameters and to recommend management modifications, allowing a further increase of the performances.