Data-driven flexibility evaluation methodology for community integrated energy system in uncertain environments

Abstract

Flexibility operation and evaluation have become central and critical for grid integration of high percentage of renewable energy generations (REGs). However, the interactions between multiple energy forms in community integrated energy system (CIES) and the uncertainty of REG significantly complicate the quantification of flexibility. To this end, a novel two-stage data-driven flexibility evaluation methodology is proposed for CIES. The first stage performs optimal scheduling of multiple flexible resources, where the flexible thermal comfort requirements of users in integrated response demand are considered by introducing a predicted mean vote index. The second stage refines the flexibility supply model for flexible resources, and carves out the system flexibility supply domain by enveloping the net load. To overcome the uncertainty of REG, a data-driven scenario analysis approach is developed, which fully captures the renewable output characteristics under different climate environments and avoids complicated explicit feature model construction. Furthermore, a set of evaluation metrics is proposed from the aspect of direction, amount, and frequency to quantify the flexibility. The effectiveness and accuracy of the proposed methodology are verified by applying it to a modified CIES located in North China. The sensitivity analysis of multiple time scales and flexible resources to operational flexibility and economy of CIES is presented, which provides guidelines for the efficient allocation of flexible resources and expand planning of CIES.