Active flow controller for inhibiting vortex-induced vibration based on deep reinforcement learning and control method

Abstract

The invention provides an active flow controller for suppressing vortex-induced vibration based on deep reinforcement learning and a control method. According to the active flow controller disclosed by the invention, a deep reinforcement learning decision-making agent is established based on a Soft Actor-Critic algorithm, and a robust real-time control strategy is output through continuous interaction with a flow environment. A reward and punishment function related to a vortex-induced vibration state in a cylinder transverse flow direction and cylinder surface resistance is established, an artificial neural network weight in a decision-making agent is dynamically learned and adjusted, and a mapping relation from flow states such as flow environment speed and pressure to control actions is established, so that the active flow controller is obtained. The active flow controller is used for controlling the air suction and blowing devices symmetrically installed on a cross flow pole of a cylinder, and the two control targets of vibration suppression and resistance reduction of the cylinder can be achieved.

Description

technical field [0001] The invention relates to an active flow control method for suppressing vortex-induced vibration based on deep reinforcement learning. Through the continuous interaction between the decision-making agent and the flow field environment, the parameters of the artificial neural network are dynamically adjusted, and according to the flow state observation of the numerical simulation environment, the control The blowing and suction device on the cross-flow pole of the cylinder affects the shedding of the vortex on the surface of the cylinder and the fluid-solid coupling process, thereby achieving the effects of suppressing vortex-induced vibration and reducing drag, which belongs to the field of active flow control. Background technique [0002] Designing active flow control strategies is a tedious task. Compared with passive flow control, the actuating mechanism of active flow control is often more complex. Therefore, when designing an active flow control ...

AI Technical Summary

Problems solved by technology

Compared with traditional wind tunnel experiments, the emergence of computational fluid dynamics technology has provided convenience for the exploration of control strategies, but due to the high dimensionality and strong nonlinear characteristics of fluid mechanics, the exploration of strategies requires a large amount of computing costs. Active flow control strategies in most studies are limited to simple constant or harmonic inputs, therefore, there is a need to develop an effective control strategy exploration method for active flow control mechanisms that takes full advantage of the control possibilities of active control

Images