الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 109 |  |  |
| | | from | 109 | | |
Abstract
The safe operation of nuclear reactors for the production of electricity and research activities is one of the main objectives of economic and sustainable development. Generally, the nuclear reactors operation requires appropriate and safe operating conditions. This is achieved through a robust application of conventional and advanced intelligent control systems. Those systems control the neutron flux changing in a regular operation and have the ability to overcome any external disturbances that occur to the reactor during the normal operation. The nuclear reactor dynamics is highly complicated due to non-linearity and time changes. The complexity of building a complete model close to reality arises from the coupling and interaction among temperature-sensitive parameters in both neutronics and thermal hydraulics. Therefore, the researchers are forced to simplify the equations of nuclear reactors in order to facilitate the use of simplified models in conducting studies and analyzing the performance of the nuclear reactor under different operating conditions, especially when applying traditional and modern control systems. Thus, the main objective of conducting this work was to study the various mathematical models of the nuclear reactor, starting from the first degree of the delayed neutrons to the sixth degree, to reach the proper degree that gives the best performance to the real reactor kinetic response. The nuclear reactor six models related to the delayed neutron precursors in six groups was introduced, starting from first degree to sixth degree, to study the performance of each group with different conventional and advanced controllers, and to tune those controllers by different kind of optimization techniques, based on investigating the power response of each module of control systems using Matlab environment. The development of the control system was based on the difference between the desired output power and the reference power, which determined the design parameter or optimization techniques, utilizing data modelling and cost function optimization procedures to optimize the designed controller parameters. These controller parameter techniques were based on a cost function that aimed to reduce the errors. Four commonly used cost functions adjusted the controller parameters settings. These cost functions were: Integral Absolute Error (IAE), Integral Time Absolute Error (ITAE), Integral Square Error (ISE), and Integral Time Square Error (ITSE). We applied conventional control systems such as the Proportional Differential controller (PD) and the Proportional Integral Differential (PID) controller to control the power level. The proposed model and its performance were compared for all degrees of delayed neutron groups, starting from the first degree to the sixth degree. Artificial intelligence, such as Genetic Algorithms (GA), was used to optimize the variables of the PD and PID controllers. In addition, Particle Swarm Optimization (PSO) algorithms were used to optimize the parameters of the controllers to reach the best performance of the operation of the nuclear reactor. Moreover, Genetic Algorithms were also used to optimize the parameters of Fuzzy logic controller such as input scaling gains, membership function types and its dimensions and the rule base. The comparison of conventional and advanced controllers was carried out for all six delayed neutron models of the nuclear reactor to investigate the proper model degree of reactor kinetics in order to facilitate the use of simplified models in the future research.