Controle Estatistico de Processo (CEP) aplicado na avaliação de uma estação de tratamento de efluentes

Abstract

Contamination of water bodies has been a growing problem in recent decades. The industries are responsible for a large part of this contamination and it is up to them to decontaminate the waste generated before the final disposal. In this context, it is necessary to carry out the treatment of the effluent generated to guarantee the quality of the water in the rivers. It is known that the effluents received at the WWTP depend on several characteristics of the factory's production line and the manufacturing process, making them often unstable, leading to problems in terms of treatment efficiency, so it is important to use techniques for identification failures in the process in order to guarantee the stability of the effluent. In this context, the performance and stability evaluation of an effluent treatment plant is presented, using statistical process control techniques. Shewhart individual measure plots and weighted exponential moving average (MMEP) plots were used. The studied ETE has as main characteristic the treatment of the effluent generated in the potato snack manufacturing process. Three ETE points were evaluated, raw effluent, effluent after physical-chemical treatment and final effluent, the parameters measured were biochemical oxygen demand (BOD5,20), chemical oxygen demand (COD), total phosphorus (P) and potential hydrogen (pH), over a period of two years, monthly temporal space. A descriptive statistical analysis of the data was performed to identify the outliers and their causes. The Shewhart control charts indicated that the parameters DBO5,20 and COD of the raw effluent were shown to be out of statistical process control. In the effluent after physical-chemical treatment, all variables are under statistical process control. In the treated effluent with the exception of COD, all variables are under statistical process control. The MMEP graphs showed that the raw effluent and after physical chemical treatment have all variables under statistical process control. In the final effluent, the only variable under control was pH. It was possible to identify that there is a difference in the use of the graphs, the Shewhart individual measurement graph being more suitable when there are large changes in concentration in the variables, such as in the gross effluent. MMEP, on the other hand, was more efficient in evaluating the final treatment, since concentrations vary succinctly and are not detected by the Shewhart graph. The process capacity index showed that the ETE is not able to meet the limits required by Conama 357/05 legislation regarding the total phosphorus parameter. However, for the other parameters analyzed, the ETE has the capacity to treat effluent in order to reduce the polluting load as required by federal and state legislation.