Real-time voltage control method for distributed expandable quantum deep width learning

Abstract

The invention provides a real-time voltage control method for distributed expandable quantum deep width learning. According to the method, a distributed structure and an expandable quantum depth-widthlearning method are combined for voltage control of a power system. Firstly, the proposed method combines the ideas of deep learning and width learning, introduces a density matrix in quantum mechanics, and proposes an expandable quantum deep-width neural network. Secondly, according to the method, four networks of a depth deterministic strategy gradient method structure are fitted by using an expandable quantum deep width neural network, and an expandable quantum deep width learning method is provided. And finally, the voltage of the power system is controlled in real time through a distributed structure. According to the method, real-time global optimal control of the voltage of the power system can be achieved, on the basis of guaranteeing the control precision, the calculation burdenof a controller is relieved, the calculation process is accelerated, the requirement of the voltage control process for the reliability of the communication technology is lowered, and the privacy of power system information of all regions is kept.

Description

technical field [0001] The invention belongs to the field of power system voltage control, relates to a real-time voltage control method of distributed artificial intelligence technology, and is suitable for real-time control of power system voltage. Background technique [0002] As more and more intermittent energy sources are added to the power system, the uncertainty of the reactive power output of a large number of intermittent energy sources increases the risk of the grid voltage exceeding the limit, and also increases the power system's demand for real-time voltage control . The existing grid voltage control method is mainly based on the centralized three-layer voltage control method, and the voltage control instructions need to be transmitted layer by layer before they can be executed, which makes it difficult for the existing voltage control method to be applied to power systems containing a large number of intermittent energy sources . Therefore, it is necessary t...

AI Technical Summary

Problems solved by technology

[0004] At present, deep learning methods have attracted the attention of many scholars in many research fields, but the problem of slow learning and training caused by the multi-hidden layer structure of deep learning methods has not been particularly effectively solved.

Since the width learning method does not have a multi-hidden layer structure, its learning and training speed is not limited by the multi-hidden layer structure, but its training accuracy needs to be improved

Images