Abstract
Based on a Double Deep-Q Network with deep ResNet (DDQN-ResNet), this paper proposes a novel method for transmission network expansion planning (TNEP). Since TNEP is a large scale and mixed-integer linear programming (MILP) problem, as the transmission network scale and the optimal constraints increase, the numerical calculation and heuristic learning-based methods suffer from heavy computational complexities in calculation and training. Besides, due to the black box characteristic, the solution processes of the heuristic learning-based methods are inexplicable and usually require repeated training. By using DDQN-ResNet, this paper constructs a high-performance and flexible method to solve large-scale and complex-constrained TNEP problem. Firstly, we form a two-objective TNEP model, in which one objective is to minimize the comprehensive cost, and another objective is to maximize the transmission network reliability. The comprehensive cost takes into account the expansion cost, the network loss cost, and the maintenance cost. The transmission network reliability is evaluated by the expected energy not served (EENS) and the electrical betweenness. Secondly, from the TNEP model, the TNEP task is constructed based on the Markov decision process. By abstracting the task, the TNEP environment is obtained for DDQN-ResNet. In addition, to identify the construction value of lines, an agent is establish based on DDQN-ResNet. Finally, we perform the static planning and visualize the reinforcement learning process. The dynamic planning is realized by reusing the training experience. The validity and flexibility of DDQN-ResNet are verified on RTS 24-bus test system.
Highlights
In terms of dynamic transmission network expansion planning (TNEP) problems, the heuristic learning-based algorithm is still powerful, and the dynamic programming genetic algorithm has been proven to be effective in the dynamic planning of large-scale power distribution networks [25]
For the first time, this paper proposes a novel method to solve TNEP tasks through the deep reinforcement learning DDQNResNet
The model contains the improved energy not served (EENS) based on equal dispersive sampling, the improved electric betweenness based on Gini coefficient and the comprehensive cost of transmission network, and we use them to measure the reliability and economy of the transmission network
Summary
Power system component failures are often considered to be independent, which is inconsistent with the real situation For this reason, the Markov cut set method based on DC-OPF was proposed in reference [10] to calculate the PSRE under independent faults. The following three methods are adopted to simplify this dynamic complexity: 1) STATIC PLANNING APPROACH This type of method only considers the grid structure of the target year and calculates the total cost during the planning period. In terms of numerical calculation methods, reference [18] constructed the multi-level equilibrium model through the classification of TNEP constraints, which greatly improves the solution of MILP problems. In terms of dynamic TNEP problems, the heuristic learning-based algorithm is still powerful, and the dynamic programming genetic algorithm has been proven to be effective in the dynamic planning of large-scale power distribution networks [25].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
