The invention provides an intelligent treatment method for
mariculture tail water, and belongs to the technical field of
machine learning. By constructing a multi-mode
tail water environment-evolution
data set, comprehensive
perception of a
tail water visual
image sequence, physicochemical index
time sequence data and operation
control parameters is realized, and dynamic evolution prediction of a tail water
system is carried out based on a virtual environment generation model. Through the
steady state identification and mode mining module, the stable operation cluster, the critical boundary and the abrupt change mode of the tail water
system can be quantified, control-abrupt change response knowledge is extracted, and a structured basis is provided for optimization
decision making. On the basis, a multi-
target control optimization model is constructed, efficient regulation and control over the
microalgae growth rate, the suspended particle
sedimentation rate and the nutritive salt
absorption efficiency are achieved, meanwhile, the potential
mutation risk is avoided, and steady-state operation of the
system is guaranteed. According to the method, intelligentization, precision and multi-objective optimization of the tail
water treatment process can be achieved, and efficient and reliable
technical support is provided for
mariculture tail water ecological management.