The present study proposes a new feature extraction method based on non-stationary conductivity probe signals. Two types of discriminative network models, i.e., the probabilistic neural network (PNN) and nonlinear support vector machine (SVM), are established for flow regime identification using small sample sets. The eigenvectors are composed of 16 feature quantities obtained by wavelet packet decomposition (WPD) and 8 feature quantities in the time-domain derived from the reconstructed low-frequency signals. The 8 features include maximum, minimum, standard deviation, arithmetic mean, kurtosis, peak factor, impulse factor and margin factor. The signals are normalized based on features rather than samples before flow regime identification. In the current study, WPD results show that the conductivity probe signals in two-phase flow are mostly in low frequency. The identification accuracy of the nonlinear SVM is 90.47%, which is better than 83.33% by the PNN method. This study verifies the superiority of nonlinear SVM in solving small samples and nonlinear flow regime classification problems. However, the accuracy of flow regime identification near flow regime transitional boundaries still remains questionable and needs further improvement.

The floating reactor system will be rolling, heaving and other movements affected by waves, wind, etc. These motions will introduce additional inertial force field into the rod bundle, thus affecting the flow characteristics in the rod bundle channel. In order to study the influence of rolling motion on the flow characteristics of rod bundle, a visualization research of the flow field of rod bundle channel with a pitch-to-diameter ratio of 1.326 was carried out under rolling motion. The results show that under a small driving force, the rolling motion has a significant effect on the flow field in the rod bundle, affecting the velocity distribution in different sub-channels, and there is transverse mixing between adjacent sub-channels. With the increase of driving force, the influence of rolling motion is gradually weakened. The flow field distribution under rolling motion is significantly different from that under pulsating flow. The experimental results show that the influence of rolling motion on the middle sub-channel of the rod bundle channel is small, and the influence on the edge sub-channel is large. The velocity field of the subchannels on both sides of the edge fluctuates periodically, and the wave phase is inverse. This study shows that the flow field variation caused by rolling motion is different from that caused by pulsating flow, and the flow field of fuel assembly in floating reactor system under ocean condition is further studied.

Ocean conditions will cause ships to be affected by rolling, heaving and other motions, which introduce additional inertial forces to the marine nuclear power system, introduce additional pressure drop to the fluids in each loop, bring errors to the measurement of thermal hydraulic parameters such as pressure difference and flow rate. Therefore, the measurement of pressure difference and flow rate under ocean conditions is studied and corrected in this paper. We built a rolling platform and a heaving platform and measured the differential pressure and flow rate under various working conditions. Through analysis, the relationship between differential pressure, flow fluctuation and swaying, undulating and other movements under moving conditions was obtained. Besides, the calculation method of attached pressure drop was analyzed, and the following conclusions were obtained: The influence of motion on differential pressure and flow rate is mainly composed of two parts: The additional pressure drop caused by the swinging motion of working medium in the pipeline and the pressure lead-in pipe, and the periodic fluctuation of friction pressure drop caused by the periodic fluctuation of flow velocity caused by swinging motion. The attached pressure drop caused by motion can be calculated through theoretical analysis. Through the analysis of experimental data and the comparison of theoretical calculation methods, it is found that the pressure drop measured by the experiment is in good agreement with that fitted by the theoretical calculation, and the modification scheme of pressure difference under ocean conditions is feasible to some extent.

The load distribution and coupling feedback between parallel loops is one of the issues in the multi-loop nuclear reactor system. In the present study, a dual-loop reactor natural circulation experimental system with typical structural features was designed and built based on the characteristics of a floating nuclear power plant. The system is used to study the system response characteristics under the conditions of heat load disturbance, resistance disturbance and system response time. The results show that the flow of the primary loop will be redistributed when the load disturbance was introduced into the natural circulation loop. The primary flow rate changes quickly when the heat load disturbance reaches the critical point. When the heat load disturbance is lost on one side, the total natural circulation flow rate of the loop decreases to 60% of the value in the steady-state flow. The response of the system after introduction of resistance disturbance is similar to the introduction of the load disturbance. The response of the system can be divided into two stages, i.e., the rapid response period and the slow adjustment period after the introduction of the disturbance. With the increase of disturbance, the negative slope of stage 1 increases, and the slope of stage 2 gradually changes from positive to negative. With the experimental, a model is developed to predict the response of the system. A characteristic number S is defined to characterize the symmetry of the system. Based on the experimental data, it is found that when S is between 0.5–2, the external disturbance will not cause asymmetric response of the system.

The rod bundle fuel is characterized by compact structure and narrow flow passage. The fragments and corrosion products, flowing with the coolant, can cause local blockage accident, threaten the integrity of the fuel cladding. Therefore, it is necessary to use the Particle Image Velocimetry (PIV) to visualize and measure the flow fields downstream of the blockages. The results show that partial blockages will cause flow reversal. In the backflow zone, vortices are generated downstream of the blockage, causing increase in the resistance. The length of backflow zone increases with the increase of the Reynolds number. The wake area formed downstream of the blockage presented periodic changes with the time and the period is about 0.8s. For the blockage of the interior subchannels, in the backflow zone, two rows of asymmetrically distributed vortices, and the vortices interfere with each other and cause squeeze deformation. For the blockage of the side and corner sub-channels, the formed vortices have irregular shape and nonuniformed distribution, and the flow field is more complex and changeable. This is believed to be caused by the high intensity turbulence and the influence of the wall.

The present study investigates the air-water two-phase flow across a 90-degree vertical-upward elbow with the computational fluid dynamics (CFD) simulation. The Eulerian-Eulerian two-fluid model and the Multi Size Group (MUSIG) model are used to predict the development of the detailed interfacial structures between the two phases. The axial development of the void fraction and the interfacial area concentration are investigated and benchmarked with the experimental data measured using the four-sensor conductivity probe. It is concluded that CFD simulation can predict the characteristics distributions of void fraction and interfacial area concentration and their development downstream of the elbow. The double-peaked void fraction distribution is found to be caused by the secondary flow induced by the elbow. The liquid phase on the outer curvature moves to the inner curvature and forms a double counter rotating vortex, entraining the bubbles to form a double-peaked distribution. The elbow effects become dissipated between 33 and 63 hydraulic diameters. The simulation results of liquid-phase and gas-phase parameters can be used to develop the theoretical two-phase flow models for the elbow region.