Abstract:

As large energy users, the short-process steel enterprises have great potential for peak-shaving, which provides an important resource for improving the peak-shaving state of the power grid. However, their production processes are closely linked and orders fluctuate greatly, resulting in irregular electricity consumption, which makes it difficult for steel enterprises to participate in power grid peak-shaving. To this end, this paper proposes a bi-level optimization peak-shaving strategy for short-process steel enterprises considering maximum demand based on the improved resource-task network (RTN) model, so as to help short-process steel enterprises participate in power grid peak shaving. Firstly, an improved RTN with time window nodes was designed to accurately characterize the coupling relationships of materials and temporal resources between devices within a single production line when processing different types of orders, ensuring the feasibility of order allocation and scheduling strategies. Secondly, enterprise orders were allocated based on the actual multi-production-line scenarios, and a supply-demand interaction bi-level optimization peak-shaving model considering maximum demand was proposed, which was solved using a hybrid algorithm combining adaptive particle swarm optimization (APSO) and the Cplex solver. Finally, according to the data from an actual short-process steel enterprise, three simulation scenarios were set up to verify the proposed scheduling strategy. The results show that the proposed strategy can effectively smooth the load curve while reducing the enterprise’s electricity costs.

Key words: short-process steel enterprises, bi-level optimization, maximum demand, order allocation, peak-shaving strategy, time window node