Multi-Stage Stochastic Planning Of Regional Integrated Energy System Based On Scenario Tree Path Optimization Under Long-Term Multiple Uncertainties

Planning a regional energy system based on the advantages of different types of energies and building an economic and efficient regional integrated energy system (RIES) are hot research subjects in the energy planning field. The call for using clean energy and attaining carbon neutrality worldwide has also attracted research attention in the field of low-carbon and clean planning. Because of the growth of the energy system construction cycle together with the utilization of various types of energies, multi-stage planning considering the uncertainty and construction time sequence in a long time scale has gradually become the primary research direction for planning RIES. The uncertainty of multi-energy load growth and the energy price fluctuation in a long time scale considerably affects the construction and economy of the final planning scheme. Aiming to solve the problems of incomplete load data, low accuracy of single load forecasting, and weak regularity of the energy price fluctuation, this study proposes a multi-stage scenario tree generation method based on the conditional generative adversarial network-random forest-Markov chain. And a method of energy price upper and lower boundary determination is based on the multi-index artificial neural network method for analyzing and solving energy price fluctuation. A multi-stage stochastic planning model of RIES is built herein to minimize the comprehensive cost, including the construction investment and regional operation costs, energy hub income, and carbon emission reduction. Finally, the Beichen district of Tianjin, China is considered a case for verifying the effectiveness and applicability of the proposed model and method. Results show that the construction time sequence of the energy hub substantially affects the planning scheme. Energy hub construction in advance benefits the energy hub income and carbon emission reduction; however, the initial stage requires more investment, which affects the planning and construction of renewable energy. Compared with a single-stage deterministic planning scheme, the optimized multi-stage stochastic planning scheme reduces the number of idle facilities and benefits cost recovery, energy hub income, coverage scenarios, and carbon emission reduction.