Predicción del comportamiento de la dbo5, DQO, SST, n. total y p. total en planta piloto mediante minería de datos

Abstract

Within the sewage treatment system there are different configurations which are applicable according to the type of treatment, water quality, socioeconomic level, among others. For these treatment systems to operate and work in an adequate and efficient manner; it is necessary to carry out continuous monitoring of the behavior of different physicochemical parameters; such as BOD5, COD, SST, Ntotal, Ptotal, among others; These parameters are of great importance not only for monitoring the treatment system, but some of these are the basis for the design of such systems. The monitoring of treatment systems generates key information that generally expresses the behavior of the system in certain periods; thus, and to be able to analyze this information which is usually generated in large quantities and stored in databases or on paper; Currently, artificial intelligence has gained great importance, specifically data mining and its artificial neural network technique. This technique has been used to facilitate the analysis of a large volume of data or information, and through this technique and simulating the behavior of a biological neural network; to be able to abstract and understand said information stored in databases as well as estimate their behavior over time and be able to make decisions that contribute to the good behavior and efficiency of the treatment system.