Reactivity transients have been analyzed with an updated RELAPS-3D (ver. 2.4.2) system model of the pin core design for the 2400MWt gas-cooled fast reactor (GCFR). Additional reactivity parameters were incorporated in the RELAP5 point-kinetics model to account for reactivity feedbacks due to axial and radial expansion of the core, fuel temperature changes (Doppler effect), and pressure changes (helium density changes). Three reactivity transients without scram were analyzed and the incidents were initiated respectively by reactivity ramp, loss of load, and depressurization. During the course of the analysis the turbine bypass model for the power conversion unit (PCU) was revised to enable a better utilization of forced flow cooling after the PCU is tripped. The analysis of the reactivity transients demonstrates the significant impact of the PCU on system pressure and core flow. Results from the modified turbine bypass model suggest a success path for the GCFR to mitigate reactivity transients without scram.

Numerical Modeling of Solid Gas Flow, System Identification for purposes of modeling and control, and Wireless Sensor and Actor Network design were pursued as part of this project. Time series input-output data was obtained from NETL's Morgantown CFB facility courtesy of Dr. Lawrence Shadle. It was run through a nonlinear kernel estimator and nonparametric models were obtained for the system. Linear and first-order nonlinear kernels were then utilized to obtain a state-space description of the system. Neural networks were trained that performed better at capturing the plant dynamics. It is possible to use these networks to find a plant model and the inversion of this model can be used to control the system. These models allow one to compare with physics based models whose parameters can then be determined by comparing them against the available data based model. On a parallel track, Dr. Kumar designed an energy-efficient and reliable transport protocol for wireless sensor and actor networks, where the sensors could be different types of wireless sensors used in CFB based coal combustion systems and actors are more powerful wireless nodes to set up a communication network while avoiding the data congestion. Dr. Ahmadi's group studied gas solid flow in …