Monitoramento de uma estação de tratamento de esgoto por meio de técnicas estatísticas de controle de qualidade.

Abstract

The shortage or lack of water affects more than 40% of the world, due to political, economic, and weather forecast reasons. Thus, sanitation services as collection and sewage treatment become very important to ensure the quality of people s lives. It is known that, generally, there is only one variable responsible for the quality and variability of an effluent, which depends on the changes: in the affluent load, environmental conditions in the reactors, the nature of sewage to be treated, presence of toxic substances, variability inherent to biological treatment processes, as well as mechanical and human failures in the system. All these factors can lead to problems and instability in the processes, so, techniques and methods should be used, since they are effective in identifying failures of treatment processes. The statistical methods of quality control are techniques applied to check if sewage treatment is being effective or not. Thus, this study applied statistical methods of quality control to evaluate the performance of a sewage treatment plant in order to identify failures in the process and discuss about their possible causes. The studied plant of sewage treatment is in Cascavel, Paraná. It is considered a Ete Oeste. The plant consists of a grid system with grids of bars and manual cleaning, a gravitational tangential flow with removal of sand retained by air-lift, Parshall gutter flow meter, anaerobic biological treatment, by two reactors of up-flow sludge manta in parallel, complementary treatment by physicochemical process of coagulation, flocculation and sedimentation laminar. It was used ferric chloride as a coagulant, Parshall gutter and the final effluent disinfection by chlorination. The parameters were: hydrogen-ion potential, sediment solids, total suspended solids, chemical oxygen demand and biochemical oxygen demand of five days. The statistical analysis was carried out through Shewhart individual charts, control graphs of the moving average, exponentially weighted and indices of ability process. According to Shewhart charts, it was observed that in raw sewage, variables as total suspended solids, COD and BOD5,20 were out of statistical control. On Ralf I effluent, variables as pH, sediment solids, total suspended solids and COD were out of the statistical control. For Ralf II effluent, variables as sediment solids, total suspended solids, COD and BOD5,20 were out of the statistical control. Only the variable BOD5,20 was under statistical control in the treated effluent. EWMA graphs show that all variables were under statistical control for the raw sewage. For Ralf I effluent, all variables were out of statistical control, but for BOD5,20. For Ralf II effluent, only pH was under statistical control. On the other hand, for the treated sewage, only the sediment solids were out of statistical control. Shewhart control charts of individual measure seemed to be a good alternative to the statistical process control regarding a plant of sewage treatment, since they help on visualizing the failures during the process, especially in unforeseen changes. The EWMA control charts are also supposed to be a good alternative, because they have a good run for data without normality. However, since this study has widely long time information (monthly samples), the situations of unforeseen changes are greater and may not be detected by the charts of exponentially weighted moving average. According to the process ability indices and performance index of process, it was observed that, except for pH, the sewage treatment of this plant was not able to meet the specifications or the standard discharges required by environmental legislation.