Energy as a scarce strategy resource is the physical base for national economic development and social progress. Along with the fossil fuel depletion and environmental problems induced by energy utilization, the early-warning of energy security has gradually attracted more attention and thereby emerged as a rising discipline of energy sciences, covering the energy security theory, energy demand forecasting models and energy security early warning index systems. After reviewing the evolution of energy security theory, we summarized the merits and limitations of three kinds of commonly used forecasting models including top-down models, bottom-up models and hybrid energy models. Top-down models such as CGE could give a detailed description about the interaction among different economic sectors. However, these models fail to provide a concrete description for energy technologies. For bottom-up model, although a detailed consideration of technologies related to energy production and consumption has been incorporated into modeling, few economies can be described adequately based on such an engineering perspective. In order to integrate the advantage of both modelling approaches, more and more hybrid models have been constructed to overcome the weakness in recent years. Although more dimensions such as environmental, social, and economic aspects have been integrated to the hybrid model, the simulation for water environment and biodiversity is still insufficient. In terms of the modelling technique, the issues like boundary difference, database difference for hybrid model need to be concerned. Additionally, the current energy demand forecasting models can only predict the future energy demand without consideration of energy security. It is therefore necessary to develop the energy security evaluation indicator system. The existing energy security evaluation indicator systems can be divided into two categories: individual indicator-based system and aggregate indicator-based system. Individual indicator system attempts to employ various individual indicators such as reserve to production ratio, strategic fuel stocks, and net energy import dependency to quantitatively assess the energy security level, while aggregated indicator system as a combination of individual indicators usually develops a composite index to evaluate the energy security. Despite many efforts have been devoted to energy demand forecasting models and energy security evaluation indicator system, a universal early-warning framework of energy security, however, has not been constructed. The limitation for current early-warning system of energy security can be summarized as follows: 1) Interaction mechanism among the dimension of energy security, environmental protection, and economy for energy security evaluation indicator is still unclear. 2) Theoretical research and practical application for existing early-warning system of energy security remain seperate. 3) Weights of the indicators are still determined by subjective judgment rather than objective methods. According to the summary of advances in various energy forecasting and early warning studies and their advantages and limitations as well, it is suggested that in the future research, a chain-based early warning system should be established in perspective of supply chain considering the interactions between internal and external factors of the energy system so as to promote the current energy early-warning system.