A data-driven adaptive optimization control method and medium for random disturbance system

Abstract

The invention discloses a data-driven adaptive optimization control method and medium for a random disturbance system. The method includes a problem description part, a design part of a data-driven optimal state observer, and a different-strategy data-driven ADP control part of a random disturbance system; for The above three parts, the present invention has been described in detail. The invention implements different-strategy data-driven ADP control of a random disturbance system by designing a data-driven optimal state observer. For the first time, the data-driven ADP method is used for the system whose state is completely unpredictable; the model-free LQG control is extended to the continuous time system; the non-matching noise outside the control signal channel is considered in the ADP design, and the independent state and control signal are not dependent Noise: A new type of different strategy data-driven ADP control method and medium for random disturbance systems is proposed, which avoids the burden of repeatedly reading and updating control signals, and significantly reduces the amount of calculation.

Description

technical field

[0001] The invention relates to a random noise disturbance system, in particular to a model-free random optimal control. Random noise disturbance systems are used in many fields such as industrial and agricultural production, power system, chemical technology, machinery manufacturing, transportation, aerospace, artificial intelligence and so on. Background technique

[0002] Uncertainty in real systems can come from noise in signals such as input and state. Therefore, the optimal control problem of stochastic noise disturbance system has been paid much attention. In traditional literature, this kind of problem usually adopts H 2 or H ∞ The mainstream implementation method is to adjust the disturbance input with a certain deterministic model, and then design the state feedback control. But in engineering practice, it is often unrealistic to make external disturbances update in the way that people expect. On the other hand, the existing H 2 and H ∞ The r...

Images