Advanced process control techniques for water treatment using artificial neural networks

Abstract

Virtually all water utilities are looking at improving the operation of their plants to keep control of costs and to meet stringent water quality regulations. Better process control and automation of the plants can help achieve these goals. However, traditional control techniques such as proportional-integral-derivative (PID) can be inadequate when automating certain water treatment processes such as turbidity, organics, or hardness removal in a clarification process. Advanced process control techniques are alternatives to mitigate this impediment. At the cornerstone of many advanced process control techniques is a model of the process being controlled, which can be developed using artificial neural networks (ANNs). This paper describes various advanced process control techniques, the potentially large role of ANN models in implementing these techniques, and issues and solutions when using ANN in a real-time control system.Key words: artificial neural networks, model-based control, proportional-integral-derivative control, forward and inverse models, direct and indirect control.Résumé : Presque tous les services publics d'eau cherchent à améliorer le fonctionnement de leurs stations pour contrôler leurs coûts et rencontrer les exigences réglementaires sévères de la qualité de l'eau. Une amélioration du contrôle des procédés et de l'automatisation des stations peut permettre d'atteindre ce but. Cependant, les techniques de commande traditionnelles, telles que la commande proportionnelle, intégrale et dérivée (PID), peuvent être inadéquates lors de l'automatisation de certains procédés de traitement de l'eau tels que l'élimination de la turbidité, des matières organiques ou de la dureté dans un procédé de clarification. Des techniques avancées de contrôle des procédés sont un autre choix pour atténuer cet obstacle. La base de plusieurs techniques avancées de contrôle des procédés est un modèle du procédé qui est contrôlé, ce qui peut être développé en utilisant des réseaux neuronaux artificiels (RNA). Cet article décrit diverses techniques avancées de contrôle des procédés, le rôle potentiellement important des modèles RNA dans l'implantation de telles techniques et les questions et solutions qui surviennent lors de l'utilisation des RNA dans un système de contrôle en temps réel. Mots clés : réseaux neuronaux artificiels, commande basée sur un modèle, commande proportionnelle, intégrale et dérivée, modèles directs et indirects, commandes directes et indirectes. [Traduit par la Rédaction]

Effect of Different Defuzzification methods in a Fuzzy Based Liquid Flow control in Semiconductor Based Anemometer

“…In fuzzy logic control, above concepts is equivalent to the term is rules. Fuzzy Logic Rules are linguistic in nature and precise the operator to develop a control decision in a familiar human environment [4]. A typical rule can be written as follows:…”

“…fuzzy control technique receives many attentions due to its resemblance to human-like characteristics. In this paper the reservoir operation is modelled by the fuzzy logic based intelligent control approach, which operates by 'if-then' knowledge base logic, where the flow rate & change in flow-rate are the 'if' fuzzy explanatory variables while opening the control valve is a 'then' is a fuzzy consequence [11].Here in this paper various defuzzification methods are implemented in a tank water flow control system [4] & results are compared and optimization is achieved.…”

Design and Implementation of a Neural Control System and Performance Characterization with PID Controller for Water Level Control

“…Satean Tunyasrirut chose PID-fuzzy cascade as the model structure for a linear model based predictive control of the liquid level [15]. Riyaz Shariff utilized artificial neural network (ANN) as advanced process control technique for water treatment [3]. Corneliu Lazar Showed a selflearning PID control in the application of level control [5].…”

Modelagem e simulação de um sistema inteligente para controle de dosagem da pós-cloração em estações de tratamento de água

“…L -1 .Como exposto, a concepção do sistema necessita de um controle multivariável que possibilite prever as alterações das vazões da ETA e do sistema de compensação, além do complemento dos CRL da água filtrada e do sistema de compensação, para que se alcance o valor de set-point na saída do reservatório de água tratada. tratada, e por esse motivo restringem a utilização com o controle clássico PID, como mencionado emShariff et al (2004).O emprego de recursos de inteligência computacional se torna favorável, uma vez que abarca os sinais de vazão dos processos que impactam diretamente na dosagem e no set-point, bem como os pontos críticos de medição do CRL.…”