Energy storage plays an important role in establishing a modern energy system with clean energy as the core. In the new power system, the energy storage technology is mainly applied to promote the consumption of new energy, participate in the auxiliary service of the power market and support the construction of grid-load storage system. Firstly, the structure and classification of energy storage are summarized, and the application characteristics of energy storage are introduced according to power side, grid side and user side respectively. Then, it discusses the way of energy storage promoting new energy consumption on the grid side, and introduces the technology of "electric hydrogen production" to promote new energy consumption. Secondly, the construction of grid-load storage and its typical projects are introduced. The diversification of source grid load storage subjects greatly promotes the development of new energy and realizes the maximization of energy utilization. Finally, the development of power auxiliary services at home and abroad and typical energy storage cases are summarized, and the future development trend of energy storage is prospected.

Objectives Compressed air energy storage (CAES) is a new type of energy storage system that utilizes the mutual conversion of electrical energy and compressed air potential energy to balance the fluctuation of power grid. The accumulation of relevant experience in the construction and operation of CAES power station is of great significance to the development of CAES technology. In view of the low efficiency of information transfer between the designer and the site, the difficulty of construction control, and the difficulty of overall project process supervision, a whole-process intelligent system suitable for CAES power station was proposed. Methods Firstly, the construction process of CAES power station was analyzed and its engineering characteristics were found out. Secondly, based on the whole process from design to operation and maintenance of CAES power station, the whole process intelligent construction system of CAES power station was constructed in space and time dimensions. Finally, in the design stage, the key techniques such as forward design drawing were proposed. In the stage of equipment manufacturing, the key technologies such as virtual pre-assembly of equipment were proposed. In the construction stage, the key technologies such as 5D construction management were proposed. In the operation and maintenance stage, the key technologies such as data delivery oriented to operation and maintenance were proposed. Results The verification results of a 300 MW CAES demonstration project show that the construction of the whole process intelligent system of CAES power station is reasonable, and the application of key equipment and software provides technical support for the project. Conclusions Through the intelligent system of the whole process of CAES power station and its key technologies, the interworking chain of each stage of power station construction is opened, and the information management of the whole life cycle of CAES power station is realized.

Objectives Compressed air energy storage is a type of energy storage technology with large capacity, long cycle, low cost and high efficiency. Due to the strict requirements of gas storage chambers, gaseous compressed air energy storage cannot be widely promoted and applied in multiple scenarios and on a large scale. Therefore, a non-supplementary combustion liquid compressed air energy storage system was proposed. Methods A theoretical calculation model was constructed to conduct sensitivity analysis on key parameters such as compressor interstage temperature, number of compressor stages, and turbine inlet temperature within the system. The results were compared with those of a non-supplementary combustion gaseous compressed air energy storage system. Results Too low or too high interstage temperature in compressors will restrict the improvement of electric-electric conversion efficiency of the system. The number of compressor stages is positively correlated with compressor power consumption, and negatively correlated with the turbine power generation. Under the same inlet pressure, the higher the inlet air temperature of the turbine is, the larger the power generation is, and the higher the electric-electric conversion efficiency is. Compared with the non-supplementary combustion gaseous energy storage system, the density of non-supplementary combustion liquid energy storage system is increased by 3.7 times, and the volume of the storage chamber is decreased by 9/10. Conclusions The non-supplementary combustion liquid compressed air energy storage system effectively solves the problem of gas storage chambers, enabling compressed air energy storage technology to be promoted and applied in multiple scenarios and on a large scale. It is of great significance for deep peak shaving of thermal power units and large-scale energy storage in power grids.

In order to achieve the strategic goals of "carbon peak" and "carbon neutral", China's power grid will gradually be built into a green smart grid with new energy as the main power source and multiple types of power sources coexisting. However, the limited peak regulation capacity of traditional conventional power sources is difficult to meet the peak regulation demand of the future power system after accessing high proportion of new energy, which restricts the absorption capacity of new energy and reduces the safety and economy of system operation. Therefore, it is necessary to build multiple types of energy storage models, such as pumped storage, electrochemical energy storage, and electric vehicle virtual energy storage. Combined with four typical scenarios and extreme scenarios of a provincial power system, an optimal peak regulation efficiency model from the perspective of dispatching agency is proposed based on the existing energy storage peak regulation auxiliary service compensation mechanism. Through simulation verification using historical data from a provincial power grid, it has been demonstrated that this model plays a positive role in reducing frequent start-stop cycles for thermal power units and improving economic efficiency in peak regulation. This promotes both economic viability and safe operation for future advanced electricity systems.

Although distributed photovoltaics (DPVs) are increasingly being integrated into power grids, the problems of daytime overvoltage and nighttime low voltage in the distribution the network persist, which affect network economy. To overcome this problem, a two-stage optimization operation strategy involving DPVs and energy storage systems (ESSs) was proposed. In the first stage of the strategy, the position and amount of charge and discharge of the ESSs were determined by calculating the voltage sensitivity of the node and the adjustable photovoltaic reactive power node. Next, the optimization operation model was established. The model minimizes the sum of dispatching cost of ESSs, power exchange cost, and network loss cost. The constraints of the study include the distribution network, node voltage, state of charge (SOC) of ESSs, and adjustable reactive power capacity of photovoltaic inverters. In this model, particle swarm optimization was used to devise the daily scheduling output strategy of DPVs and ESSs. Finally, the effectiveness of the proposed method was verified using a 31-node actual distribution network. The simulation results revealed that the proposed strategy can effectively solve the voltage violation problem through the cooperative dispatching of DPVs and ESSs. Thus, the operating cost of the distribution network can be optimized while ensuring voltage safety.

In order to meet the challenges brought by the high proportion of new energy access in the future, it is necessary to fully tap the adjustable potential of different types of scheduling resources. Therefore, a multi-time scale optimal scheduling strategy of source-load-storage considering demand response was proposed to improve the economy and reliability of system operation by participating in the coordinated optimal scheduling of power grid. Firstly, the characteristics of different types of adjustable resources were analyzed, and the overall framework of multi-time scale rolling scheduling was constructed. The overall scheduling was divided into two stages: day-ahead scheduling and intra-day scheduling. Secondly, based on the multi-scenario stochastic programming method, the day-ahead and intra-day optimal scheduling models with the goal of minimizing the total operating cost of the system were established, and the models were solved under the premise of ensuring the reliable operation of the system. Finally, the improved IEEE-30 node system was used for simulation analysis to verify the feasibility and effectiveness of the proposed strategy.