With the great variety of complicated characteristics of modern power systems

plus the fact that power grid interactive operation has become a research hotspot in the new situation

improvement of safe operation of the system in the many kinds of complex operating characteristics has become the research focus and problem. Such being the case

a power flow balancing analyzing method based on the entropy theory is proposed. By studying the characteristics of power flow variation

new methods for system safe operation improvement are explored. By proceeding from the mechanism of influence of load variation on power flow

the power flow transfer entropy is established to describe the balance of branch power flow variation. On this basis

an optimal power flow balancing model is proposed with the power flow transfer entropy as the objective function. By generation rescheduling

the load security margin of power system could be improved. The conventional operation mode and source-load interactive operation mode are set in an IEEE 300-bus system case study. The results have validated the effectiveness of the power flow transfer entropy index and the optimal power flow balancing model under discussion. Additionally

compared with the optimization method with load margin as the objective direct

the method proposed has much higher efficiency

which provides a reference for daily scheduling and day-ahead scheduling in the power system.