Abstract:

In response to the current lack of effective error prevention verification methods and low level of intelligence in operation tickets, a knowledge graph based intelligent error prevention technology for substation switching is proposed by combining deep reinforcement learning algorithms. Firstly, the data such as power grid equipment topology and scheduling error prevention regulations is utilized to constructe a physical entity graph and an error prevention semantic graph of the equipment, and a scheduling knowledge graph is automatically fused to form. Then, based on intelligent error prevention algorithms, a graph of error prevention regulations is constructed to automatically generate the optimal switching operation sequence that complies with the error prevention regulations, and the intelligent error prevention verification is achieved. Finally, the practicality of knowledge graph intelligent error prevention, the performance of deep reinforcement learning, and the efficiency of intelligent error prevention are analyzed through examples. The results show that the proposed method has certain advantages in improving the efficiency and accuracy of intelligent error prevention for switching.

Key words: switching operation, intelligent error prevention, deep reinforcement learning, knowledge graph