Aiming at precision of infrared temperature sensor is affected by ambient temperature

a new method is put forward for temperature compensation based on particle swarm optimize（ PSO） BP（ PSO-BP） neural network.The proposed PSO-BP algorithm solves problem that iteration speed of general BP algorithm is slow

and prone to local optimum

take ambient temperature compensation of thermopile sensor for example to realize algorithm

and compare with pure BP algorithm. Simulation results show that the proposed PSO-BP algorithm has obvious compensation effect on ambient temperature

and the effect is better than that of pure BP algorithm. The constructed compensation model has high compensation precision and practical value.