Research on Predictive Control Based on LongShort-Term Memory Network

Predictive control has the advantages of low model accuracy requirements, ability to handle multivariables, and ability to handle constraints. This paper expounds the basic principles of long- and short-term memory networks, and combines its advantages in modeling time series data with a recursive multi-step model to establish a prediction model. The golden section method is used to dynamically solve the control quantity. Correct the error of the prediction model in real time through feedback correction. The design of predictive control algorithm based on long and short-term memory network is completed. The simulation verifies the feasibility and antidisturbance of the algorithm.