Search

A mixed-signal binarized neural network (BNN) processor for the MNIST image classification is demonstrated based on analogue circuit networks. BNN algorithm for training neural networks with binary weights and activations reduces power consumption and memory size. This algorithm is a design to perform core operations of multi-layer perceptron (MLP) to reduce complexity and power consumption using analogue circuits. The mixed-signal BNN processor employs a current mirror neuron to perform multiply-and-accumulate (MAC) operations and sign activation functions. The near-threshold current mirror neuron that computes the key operations of the BNN algorithm is used to achieve low power consumption. The design occupies 0.065 mm2 in 28-nm CMOS with 560B of on-chip SRAM. The 28-nm CMOS test-chip achieves energy efficiency of 24.1 TOPS/W and 94% accuracy on the MNIST image classification. This design with binary weights and activations exhibits only 5% degradation in accuracy compared with the models with floating-point precision. [ABSTRACT FROM AUTHOR]

Liquid hydrogen has been studied for use in vehicles. However, during the charging process, liquid hydrogen is lost as gas. Therefore, it is necessary to estimate and reduce this loss and simulate the charging process. In this study, the initial charging process of a vehicle liquid hydrogen tank under room temperature and atmospheric pressure conditions was numerically investigated. A transient thermal-fluid simulation with a phase-change model was performed to analyze variations in the volume, pressure, mass flow rate, and temperature. The results showed that the process could be divided into three stages. In the first stage, liquid hydrogen was actively vaporized at the inner wall surface of the storage tank. The pressure increased rapidly, and liquid droplets were discharged into the vent pipe during the second stage. In the third stage, the mass flow rates of liquid and hydrogen gas at the outlet showed significant fluctuations, owing to complex momentum generated by the evaporation and charging flow. The temperatures of the inner and outer walls, and insulation layer, decreased significantly slower than that of the gas region because of its high heat capacity and insulation effect. The optimal structure should be further studied because the vortex, stagnation, and non-uniform cooling of the wall occurred near the inlet and outlet pipes. [ABSTRACT FROM AUTHOR]