An optical dissolved oxygen sensor nonlinear temperature compensation method based on a neural network L-M algorithm

The invention belongs to the technical field of environmental monitoring, and relates to a nonlinear temperature compensation method for an optical dissolved oxygen sensor. The nonlinear temperature compensation method for the optical dissolved oxygen sensor comprises the following steps: constructing L-M algorithm-based neural network structure by a large amount of actually measured calibration point sample data and subjecting the neural network structure to network training learning; inputting measured data and environment temperature parameters obtained through experiments into a neural network completing network training learning, and carrying out network operation; Error calculation is carried out on a result of neural network operation, and correction compensation and output are carried out. Neural network L- And the M algorithm has good training learning efficiency and compensation correction effect on the basis of solving the nonlinear temperature compensation problem of the optical dissolved oxygen s