Application of data generation model in aquaculture water quality monitoring

Abstract

In order to solve the problem of insufficient data in the process of constructing concentration monitoring model of ammonia nitrogen in intensive aquaculture, a new improved data generation model of TableGAN is proposed based on the model optimization algorithm. The method generates synthetic data with the same distribution characteristics as the original data by confrontation training, and makes the generated data more effective in the optimization model by adding classifiers and optimization functions. The field data of a breeding enterprise show that the accuracy of the ammonia nitrogen concentration soft sensing model trained by the synthetic data set is better than that of the model trained by the original data set in terms of root mean square error and maximum absolute error, and the test effect of the model is also improved significantly.