Search

This paper presents an improved long short-term memory (LSTM) neural network based on particle swarm optimization (PSO), which is applied to predict the closing price of the stock. PSO is introduced to optimize the weights of the LSTM neural network, which reduces the prediction error. After preprocessing the historical data of the stock, including opening price, closing price, highest price, lowest price, and daily volume these five attributes, we train the LSTM by employing time series of the historical data. Finally, we apply the proposed LSTM to predict the closing price of the stock in the last two years. Compared with typical algorithms by simulation, we find the LSTM has better performance in reliability and adaptability, and the improved PSO-LSTM algorithm has better accuracy.

Demands for small-scale and low-cost robots have witnessed a great increase in recent years [1–5]. This paper introduces the design and fabrication of a novel, simple, low-cost and designer-friendly locomotive robot. The materials and tools to build the robot originate from everyday life. The robot is pneumatically powered and manually controlled by simply pumping and vacuuming the syringe repeatedly, which realizes reliable locomotion by folding and opening of the planes. In order to realize this complicated motion, a “3D Sticky Actuator” is developed. The motion and force analysis of actuator are then modelled by the numerical method to develop the relations between design parameters. This suggests a systematic and user interactive way of manufacturing various shapes of the actuator, depending on user-defined road condition (e.g. obstacles and slopes) and other constraints. One key advantage of the paper-based robot is suggested by its high feasibility.

University staff and students typically spend most of their time indoors. This paper evaluates the thermal environment of an air-conditioned scientific laboratory in a tertiary educational institution in Malaysia using Computational Fluid Dynamics (CFD). This computational technique has been used in analysing the indoor environments and has been found to be useful in aiding facilities management. A pilot survey was conducted to collect the required information such as indoor parameters and boundary conditions for the setting up of a CFD model of the laboratory. The model was then simulated based on the data obtained from field observations. Results indicate that the laboratory users sitting at different rows and work desks would experience different thermal sensations. The mean air temperature was below the recommended comfort zone specified in the local energy standard, but the air velocities were generally within the acceptable range. Based on the calculated predicted mean vote (PMV) and predicted percentage dissatisfied (PPD) indices, most of the users would be thermally uncomfortable, and a warmer environment was preferred. Recommendations were made to regulate the inlet air temperature of the laboratory to improve thermal comfort of laboratory users and for energy saving purposes.