A novel parameter estimation in dynamic model via fuzzy swarm intelligence and chaos theory for faults in wastewater treatment plant

• Published in: Soft computing (Berlin, Germany) Vol. 24; no. 1; pp. 111 - 129
• Main Authors: Anter, Ahmed M., Gupta, Deepak, Castillo, Oscar
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.01.2020; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Artificial Intelligence; Chaos theory; Classification; Computational Intelligence; Control; Cost function; Datasets; Dynamic models; Engineering; Exploitation; Fault detection; Faults; Feature selection; Focus; Fuzzy logic; Fuzzy sets; Intelligence; Mathematical Logic and Foundations; Mechatronics; Optimization algorithms; Optimization techniques; Parameter estimation; Robotics; Swarm intelligence; Uncertainty; Wastewater treatment; Water treatment; Whale optimization algorithm; Fuzzy; Fault detection; Chaos theory; Wastewater treatment; means algorithm
• ISSN: 1432-7643; 1433-7479
• DOI: 10.1007/s00500-019-04225-7

Faults during a wastewater treatment for plant (WWTP) are critical issue for social and biological. Poorly treated wastewater may achieve dangerous effect for human as well as nature. This paper proposed a novel model based on a binary version of whale optimization algorithm (WOA), chaos theory and fuzzy logic, namely (CF-BWOA). CF-BWOA is applied in the application of WWTP to find out the more relevant attributes from the whole dataset, reducing cost and validation of decision rules, and helping to identify a non-well-structured domain. CF-BWOA attempts to reduce the whole feature set without loss of significant information to the classification process. Fast fuzzy c -means is used as a cost function to measure the fuzzification and uncertainty of data. Ten different chaos sequence maps are used to estimate and tune WOA parameters. Experiments are applied on a complex real-time dataset with various uncertainty features and missing values. The overall result indicates that the CWOA with the Sine chaos map shows the better performance, lower error, higher convergence speed and shorter execution time. In addition, the proposed model is capable of detecting sensor process faults in WWTP with high accuracy and can guide the operators of these systems to control decisions.