The Online Journal on Electronics and Electrical Engineering (OJEEE) Vol. (4) – No. (2) Reference Number: JO-0018 517 Mathematical Modeling and Neural Network Control for Dissolved Oxygen of Aquaculture Pond Aeration System Doaa M. Atia 1 , Faten H. Fahmy 1 , Ninet M. Ahmed 1 , and Hassen T. Dorrah 2 1 Electronics Research Institute,, Cairo, Egypt 2 Faculty of Engineering, Cairo University Email : doaa_matia@yahoo.com Abstract -Dissolved oxygen (DO) is considered as one of the most important aspect of aquaculture. It is needed by fish to respire and perform metabolic activities. Thus low levels of dissolved oxygen are often linked to fish kill incidents. On the other hand, optimum levels can result to good growth, thus result to high production yield. To increase pond production it is necessary to control dissolved oxygen. Artificial intelligence (AI) techniques are becoming useful as alternate approaches to conventional techniques or as components of integrated systems. They have been used to solve complicated practical problems in various areas and are becoming more and more popular nowadays. This paper presents a new design of diffused aeration system using fuel cell as a power source. Mathematical modeling of dissolved oxygen presents in the pond over the day and the fuel cell stack modeling are also presented. Artificial intelligence control techniques are used for controlling the speed of air flow rate from the blower to air piping connected to the pond through control blower speed. MATLAB SIMULINK results show the neural network control (NNC) offer high performance of the system. Keyword: aeration system, fuel cell, artificial intelligence (AI) techniques, neural network control. I. INTRODUCTION Oxygen is the first limiting factor for growth and well- being of fish. Fish require oxygen for respiration, which physiologists express as mg of oxygen consumed per kilogram of fish per hour (mg O 2 /kg/h). Aerators are of two basic types: splashers and bubblers. Splasher type splashes water into the air to affect aeration. Splashing action also causes turbulence in the body of water being aerated. Bubbler aerators rely upon release of air bubbles near the bottom of a water body to affect aeration. A large surface area is created between air bubbles and surrounding water. Rising bubbles also create turbulence within a body of water. Circulation of pond water by aerators is an additional benefit of aeration. NNC and FLC are the two major branches of intelligence control, which is based on the concept of AI. AI can be defined as computer emulation of the human thinking process [1, 2]. In the last years, an interest toward Fuel Cell (FC) studies has grown, as FC is a clean and efficient source of electricity, and has a wide range of transportation and stationary applications. FCs are electrochemical devices that convert the chemical energy of a gaseous fuel directly into electricity and are widely regarded as a potential alternative stationary power source [3]. They complement heat engines and reduce the dependence on fossil fuels and thus have significant environmental implications. Owing to their attractive characteristics, FCs are recognized as one of the most promising technologies to meet the future power generation requirements and likely to be widely used for the stationary production of electric energy. During the last ten years, there has been a substantial increase in the interest on artificial neural networks. During the last ten years, there has been a substantial increase in the interest on artificial neural networks. Specifically, they are good for tasks involving incomplete data sets, fuzzy or incomplete information and for highly complex and ill-defined problems, where humans usually decide on an intuitional basis [4, 5]. This paper presents the mathematical modeling of fuel cell and dissolved oxygen for aquaculture pond aeration system. Also control of dissolved oxygen of aquaculture pond is achieved using NNC technique. II. DIFFUSED AERATION SYSTEM Diffused-air system aerators use a low pressure, high volume air blower to provide air to diffusers positioned on the pond bottom or suspended in the water. A variety types of diffusers have been used, including ceramic dome diffusers, porous ceramic tubing, porous paper tubing, perforated rubber tubing, perforated plastic pipe, packed columns. Most diffused-air aerators release a large volume of air at low pressure. The minimum permissible system pressure becomes greater with increasing depth of water above diffusers, because enough pressure must be available to force air through the piping system and cause the air to exit from the diffuser against the hydrostatic pressure at the discharge point. Diffused-air systems that release fine bubbles usually are more efficient than those that discharge coarse bubbles. This results because fine bubbles present a greater surface area to the surrounding water than larger bubbles. Oxygen diffuses into water at the surface, so a large surface area facilitates greater oxygen absorption. Diffused-air systems also are more efficient in deep ponds than in shallow ponds. Diffused aeration devices are usually classified as either fine or coarse bubble referring to the relative diameter of the bubble produced. Oxygen can be supplied by means of air or pure oxygen bubbles introduced to the water to create additional gas-water interfaces. Submerged bubbles aeration is most frequently accomplished by dispersing air bubbles in