Wireless Sensor Layout Optimization of Raw Tobacco Pallets Based on Swarm Intelligence

Abstract

Temperature data inside pallets are used as important data indicators for raw tobacco storage and maintenance. To improve the monitoring effect of wireless sensors in raw tobacco pallets, the layout optimization model of wireless sensors in a three-dimensional complex environment is constructed, a multi-objective function with the smallest sensor layout cost and the largest monitoring range is established, and an improved particle swarm optimization (IPSO) is designed to obtain preliminary results. The layout of wireless sensors is optimized and then the long short-term memory (LSTM) neural network algorithm is used to predict the temperature data in a cigarette pallet to achieve the secondary optimization of the sensor layout. Finally, on the basis of actual temperature data in raw tobacco pallets, a simulation environment model is established and verified by simulation experiments. Simulation results show that the sensor layout optimization method proposed in this paper can effectively reduce the number of sensors arranged and, at the same time, allow enterprises to effectively minimize the cost of raw tobacco storage and maintenance.