Aiming at the problem that the crucial biological parameters（ such as substrate concentration

biomass concentration and product concentration） are difficult to measure on-line in marine biologic enzyme fermentation process

according to the characteristics of its fermentation process

the method combining with mean influence value（ MIV） and Elman dynamic regression neural network is adopted. Firstly

variates are filtrated through the MIV method combined with Elman neural network

input variables which have great influence on the key biological parameters are found. Secondly

these variables are used as the input of Elman neural network and crucial biological variates are used as the output of Elman neural network

then the soft measurement model based on MIV-Elman neural network is established. Take marine protease enzyme as the research object which is a typical marine biological enzyme. Using its experimental data

the feasibility of soft measurement model based on MIVElman neural network is validated in the prediction of key biological parameters. Compared with Elman neural network

RBF neural network and BP neural network

the results show that the prediction performance of MIVElman neural network soft measurement model is better than the above three models.