Search

This paper presents the development and evaluation of three fuzzy neural network (FNN) models for a full-scale anaerobic digestion system treating paper-mill wastewater. The aim was the investigation of feasibility of the approach-based control system for the prediction of effluent quality and biogas production from an internal circulation (IC) anaerobic reactor system. To improve FNN performance, fuzzy subtractive clustering was used to identify model's architecture and optimize fuzzy rule, and a total of 5 rules were extracted in the IF-THEN format. Findings of this study clearly indicated that, compared to NN models, FNN models had smaller RMSE and MAPE as well as bigger R for the testing datasets than NN models. The proposed FNN model produced smaller deviations and exhibited a superior predictive performance on forecasting of both effluent quality and biogas (methane) production rates with satisfactory determination coefficients greater than 0.90. From the results, it was concluded that FNN modeling could be applied in IC anaerobic reactor for predicting the biodegradation and biogas production using paper-mill wastewater.

Papir je jedan od najzastupljenijih proizvoda u svijetu pa su tako potrebne velike zalihe da bi se osigurali i zadovoljili razni potrošači i prerađivači papira. Tvornica papira u Belišću kao primarnu sirovinu za proizvodnju papira koristi stari reciklirani papir te u manjim količinama drvnu sirovinu. Potrebno je jako puno proizvedenog papira kako bi se pokrili troškovi proizvodnje te potrošnje električne energije zbog njihovog cjelodnevnog rada. Kao u svakom poslu pa tako i u proizvodnji papira normalno je da dolazi do raznih kvarova i ostalih čimbenika koji utječu na proces proizvodnje i njegovo usporavanje. Zbog toga se automatizacijom i modernizacijom strojeva pokušava što više smanjiti zaustavljanje i usporavanje proizvodnje. Sustav kontrole kvalitete papira i upravljanja proizvodnim procesom omogučava da se iz kontrolne sobe sa računala daljinski upravlja pojedinim djelovima stroja i vrši pregled trenutnog stanja proizvodnje. Mjerna platforma u sebi sadrži razne senzore koji održavaju vrijednosti postavljene od strane odgovorne osobe i tako održava kvalitetu proizvoda u cjelini. Paper is one of the most prominent products in the world and so large quantities are needed to provide and meet the various consumers and paper makers. The paper mill in Belišće uses recycled paper as a primary raw material for paper production and in smaller quantities of wood raw material. A lot of paper is needed to cover the cost of production and operation of machines and electricity consumption due to their full-day work. As in any business, so in paper production, it is normal to have various faults and other factors that affect the production process and its slowdown. As a result, the automation and modernization of the machines themselves are trying to reduce production stoppage and deceleration as much as possible. The Paper Quality Control and Production Process Management system allows remote control of the individual parts of the machine from the computer control room and reviews the current state of production. The measuring platform contains various sensors that maintain the values set by the responsible persons and thus maintains the quality of the product as a whole.

Automation is being accepted for control systems onboard ships in view of the shortage of skilled manpower in marine sector. Control theory has long been applied to maneuvering problems and at present this trend is continuing at an increased rate. This is for speed control, course control and path keeping. Heading control and course keeping are very important for surface ships while they enter shallow water regions. A typical situation is the entry of a large commercial ship into the harbour basin. The ship faces a sudden change of forces and moments around it due to the change in the hydrodynamics. The Master of the ship regulates the speed while entering the basin. Forces and moments, due to the hydrodynamic flow around the moving hull, are balanced by the rudder behind it. The feed back from the heading angle is taken and the gain in the control system prompts the steering gear to turn the rudder. The conventional control algorithm based on PID is attempted in the first part of the paper and case studies are shown for a Mariner class ship whose hydrodynamic derivatives are known. Displacement, velocities and accelerations are determined for short duration from the simulation of a voyage in calm water. The proposed system can be implemented into autopilot systems. The codes developed in MATLAB can accommodate wind and wave forces as well. The simulation is of a general type and can be used for other vessels with a change in the constants of P (Proportional), I (Integral) and D (Derivative) which can be arrived at by trial and error. The design of the control system depends on the choice of the three control constants Kp, Ki and Kd. These will change as per sea state and extra loads. The control of motions in shallow water and deep water cases are discussed in the paper. In the next part of the paper, fuzzy logic control is discussed. Fuzzy logic is applied to control the heading and bring the path of the vessel to the desired level. The models discussed are versatile because of the provision for add-on facility for wind, wave or any other environmental loads. The rudder turning rate is taken as that for commercial ships. In the near future, a combined control system for ship maneuvering can be possible by considering both the PID and Fuzzy based algorithm. The two algorithms with suitable modifications are useful for non-linear cases. It is also tried to compare the two methods and an attempt is made to find out which is better and acceptable. � 2008 Elsevier Inc. All rights reserved.