Aiming at the problem of insufficient stability of gas sensor in complex environment

a sensor compensation algorithm based on improved back propagation（ BP） neural network is proposed. Firstly

the BP neural network structure based on temperature and humidity compensation is established

and the number of nodes in each layer network is determined. Then

particle swarm optimization（ PSO） is proposed to optimize the initial weight and threshold of BP neural network. Finally

the particle survival value is introduced and combined with the simulated annealing to improve the traditional PSO algorithm survival particle swarm optimization

survival PSO（ SPSO） to improve the global extremum optimization ability of the model. The experimental results show that the improved SPSO algorithm is more powerful than the traditional PSO algorithm. The SPSO and BP neural network are combined to improve the temperature and humidity compensation accuracy of the gas sensor.