1.Research significance
At present, with the rapid development of society and the continuous advancement of modernization, the demand for electricity in cities and towns is increasing day by day. Due to the harmful impact of fossil non renewable energy sources on the environment required for traditional power generation, the government vigorously promotes the policy of using new energy sources such as wind power and photovoltaic power generation. However, the random volatility of new energy power generation can pose hidden dangers to the stability of power grid operation. Since the 13th Five Year Plan, China’s energy system has undergone significant changes. As of the end of 2020, the installed capacity of renewable energy in China has reached 934000 megawatts, a year-on-year increase of 17.5%. Among them, the installed capacity of hydropower reached 370000 MW, the total installed capacity of wind power reached 281000 MW, and the installed capacity of photovoltaic system power generation reached 253000 MW.
Figure shows the total installed capacity of new energy in various countries over the past three years. It can be seen that China not only leads by far in terms of installed capacity, but also in terms of growth rate. The annual growth rates over the past three years were 12.57% and 18.02%, respectively. Compared with the almost stagnant growth rates of European countries such as Spain and Italy, it can be inferred that China’s development process of new energy power stations is still accelerating and its investment in this field will also be increasing.
Due to the fast charging and discharging characteristics of the energy storage system, it can effectively alleviate the impact of wind and light waste from new energy sources. In recent years, with the continuous updates of battery technology, electrochemical energy storage technology has also been widely applied in power systems, including auxiliary frequency regulation, load peak shaving and valley filling, power generation, distributed energy auxiliary services, and other fields.
The traditional energy storage power stations in China are mainly pumped storage power stations, which are also the most mature technology among current physical energy storage methods. They are generally owned and dispatched by power grid companies. The disadvantage is limited location selection. Electrochemical energy storage can provide various services such as peak shaving and frequency regulation assistance, backup capacity, black start, and demand side response for the operation of the power grid. It can meet the needs of upgrading and applying the “large-scale source network load storage friendly interactive system” in the power system, and has good application prospects in improving the level of accident resistance, excess wind energy, solar energy consumption, and comprehensive energy efficiency of the power grid.
(1) Ensuring the safe and stable operation of the large power grid
Electrochemical energy storage equipment can achieve rapid regulation of active and reactive power, and can quickly support power shortages in the event of grid failures, improving grid stability and risk prevention capabilities.
(2) Improve the peak shaving capacity of the power system and delay the construction of power grid supporting projects with poor economic benefits
Electrochemical energy storage power stations have the function of “peak shaving and valley filling”, which can serve as both a load and a power source, making them rare peak shaving power sources. Energy storage systems can effectively suppress the fluctuation of random power sources and loads, especially large capacity energy storage plays an important role in improving the power structure and enhancing the peak shaving capacity of the power grid, and to some extent, weakens the peak valley difference of local power grids.
(3) Large power grid frequency regulation auxiliary services
Electrochemical energy storage devices have fast and accurate power response capabilities, which can better regulate the frequency of the power grid, solve the problem of power imbalance in regional power grids in a short period of time, and improve the overall reliability of power grid operation.
(4) Improving the reactive power support capacity of regional power grids
Due to the rapid regulation ability of reactive power in electrochemical energy storage equipment, when a system malfunctions, it can quickly suppress system oscillations, stabilize voltage fluctuations, and improve the stability of power grid operation.
(5) Improve the level of new energy consumption and increase the economic benefits of source network enterprises
Electrochemical energy storage equipment can provide guarantees for the consumption of new energy on the power generation side, charging during photovoltaic power generation and discharging when there is sufficient capacity in the transmission channel. This can improve the consumption capacity of photovoltaic power stations in Qinghai that cannot be sent out due to the limitations of the transmission channel.
(6) Power grid black start power supply
In the event of a large-scale power grid accident, the restoration of power in local areas requires strong regulation ability and rapid start-up of black start power sources. Electrochemical energy storage and peak shaving power stations have strong regulation ability and fast start-up, making them ideal black start power sources and playing an important role in the rapid recovery of local power grids.
2.Research status
2.1 Current research status of large-scale battery energy storage power plants
As of the end of 2020, the proportion of new energy sources such as wind and light in the total installed capacity in China has reached 21%. In northwest regions such as Gansu and Qinghai, new energy has become the largest energy source. In the future, the scale of new energy installation will continue to grow, and large-scale energy storage stations can provide strong peak shaving and frequency regulation methods and reliable inertial support capabilities for the power grid, effectively reducing the wind and light abandonment rate of new energy and the risk of frequency instability in the power system. Therefore, in order to maintain the effective consumption of new energy and the stable operation of the main grid, it is essential to equip large-scale battery energy storage systems.
| Project Name | Energy storage configuration |
| National Wind and Solar Energy Storage and Transmission Demonstration Project (put into operation in December 2011) | Lithium ion battery: 14MW/62MW h All vanadium flow battery: 2MW/8MW-h Lead acid battery: 2MW/12MW-h Lithium titanate battery: 2MW/1MW-h |
| Golmud Era New Energy Grid Connected Photovoltaic Power Station (put into operation in June 2016) | Lithium ion battery: 15MW/18MW h |
| Dalian All Vanadium Flow Battery Energy Storage Power Station | All vanadium flow battery: 100MW/400MW h |
| Xixiantai Substation Battery System (put into operation in February 2015) | Lithium ion battery: 14MW/63MWh All vanadium flow battery: 2MW/8MWh Lead acid battery: 2MW/12MWh Lithium titanate battery: 2MW/1MWh |
Table 1.1 shows the ultra large scale battery energy storage station projects that have been built and put into operation in China in recent years. These energy storage stations are generally used for the consumption of new energy for power generation, coordination with main grid frequency regulation and peak shaving, and other purposes. Most battery types use stable and efficient lithium-ion batteries as the basic energy storage unit.
Although the technological level and large-scale commercial construction capacity of battery energy storage stations in China have significantly improved in recent years, resulting in a decreasing trend in unit cost of energy storage stations year by year, there are also three problems:
(1) Technological and economic constraints. Battery energy storage power stations have advantages such as speed and precise regulation, but their installed capacity is still relatively small compared to traditional new energy generation such as thermal power, hydropower, and even wind and solar power. Compared with pumped storage power stations using physical energy storage methods, battery based energy storage power stations have lower economic efficiency, and their common cost per kilowatt hour is 3-6 times that of pumped storage power stations. Therefore, it is difficult to participate in electricity market trading in a profit model similar to pumped storage power stations. At present, the average cost per kilowatt hour of electrochemical energy storage power stations is between 0.6 to 0.9 yuan/kilowatt hour, which is still a significant gap from the target cost of large-scale application (0.3 to 0.4 yuan/kilowatt hour).
(2) Investment costs are difficult to divert from electricity prices. Before the reform of transmission and distribution electricity prices, energy storage power stations, as part of the power grid company, provided auxiliary services such as peak shaving and frequency regulation according to dispatching instructions, and obtained service fees from the power grid company through leasing mode. However, in 2019, the National Development and Reform Commission and the Energy Administration jointly issued instructions, clearly stating that the cost of various separately approved grid electricity prices, including electric vehicles, pumped storage power stations, and electrochemical energy storage stations, should not be included in the pricing cost of transmission and distribution. Therefore, it is difficult to recover the cost of energy storage system facilities from the cost of transmission and distribution, and therefore it is difficult to participate in electricity market transactions like traditional energy sources such as thermal power and hydropower. In addition, in recent years, power grid companies have further strictly controlled the investment in grid side battery energy storage power stations. In December 2019, State Grid Corporation of China issued a notice clearly stipulating that construction of grid side energy storage power stations should not be carried out arbitrarily, which will greatly limit the large-scale application of battery energy storage.
(3) Lack of a comprehensive electricity pricing system and compensation mechanism. In terms of traditional power generation methods such as thermal power generation, hydroelectric power generation, nuclear power generation, and renewable energy power generation, which has become increasingly popular this year, the government regulatory department has established a comprehensive online pricing policy. However, due to the wide variety, small scale, and relatively high unit capacity cost of battery energy storage equipment, the government regulatory authorities have not yet established a suitable grid electricity price policy for this. In October 2017, the government issued guidance on the profitability and development direction of battery energy storage projects in China, clearly encouraging investors and developers of battery energy storage projects to find their own business models, such as utilizing existing electricity market mechanisms and providing auxiliary peak shaving and frequency regulation services. At present, the electricity pricing policies and government compensation mechanisms related to the development of large-scale battery energy storage systems mainly include peak valley electricity prices, “two detailed rules” and auxiliary service markets, as well as subsidy policies related to energy storage. However, these policies are not specifically designed for large-scale battery energy storage, making it difficult to obtain stable sources of income to achieve cost recovery of large-scale battery energy storage. For example, although the sales side peak valley electricity price policy is implemented in most provinces across the country, the peak valley price difference in most provinces and regions of China is below 0.6 yuan/kWh, far lower than the level of peak valley electricity price difference in developed countries and regions such as the United States and Europe. The small executable range of peak valley electricity prices on the user side in China and the small difference in peak valley prices will result in insufficient profits for investors in battery energy storage projects, making it impossible to recover investment within the lifespan of electrochemical energy storage stations.
2.2 Current Status of Research on Capacity Optimization and Configuration of Battery Energy Storage Systems
Due to its randomness and unpredictability, the integration of new energy generation into the grid exhibits opposite characteristics to peak shaving measures. There is data indicating that the peak valley difference of China’s daily load curve is constantly increasing, which will be detrimental to the stable operation of the power system. The rapid charging and discharging characteristics of large-scale energy storage systems can alleviate system pressure during peak load periods and reduce energy waste during low power consumption periods, thus “peak shaving and valley filling”. On the basis of large-scale energy storage system “peak shaving and valley filling”, reasonable allocation of energy storage capacity can reduce costs, reduce resource waste, increase the economy of energy storage power stations, and improve the stable operation ability of the system.
The optimization configuration methods for energy storage system capacity are mainly divided into two categories: analytical methods and optimization models. Among them, mathematical analysis methods mainly include three mathematical methods: low-pass filtering, Fourier transform, and enumeration, which are suitable for determining the number of pre-selection schemes for energy storage projects.
The optimization model method divides capacity allocation into single or multiple objectives, and establishes a mathematical model based on this. Combining the objective function of the model, appropriate algorithms are selected to solve or optimize the function.
The optimization configuration of energy storage in the DC power grid was considered in the scenario of wind power generation and photovoltaic array complementarity. This article uses predictive model control algorithm to analyze the impact of centralized and decentralized energy storage methods on energy storage configuration when connected to the grid in energy storage. The variables used are different types of new energy generation, different number of distribution station nodes, and different sampling intervals from 0 to 30 seconds for comparative analysis. The conclusion is that wind solar complementary energy can effectively reduce energy storage costs.
An improved whale algorithm was used to optimize the configuration of energy storage capacity, and a multi-objective configuration model was established with the goals of minimizing network loss, minimizing investment, and minimizing voltage fluctuations. Simulation was conducted on IEEE33 nodes, and better configuration results were obtained for both safety and economy, demonstrating the effectiveness of the improved whale algorithm. However, this paper only considers the impact of distributed energy wind power on the system, without considering photovoltaic factors, and does not compare the effectiveness of algorithms in different weather environments.
The optimization configuration of energy storage capacity was carried out using the scenario of optical storage joint system participating in energy storage frequency modulation response. This literature considered the effects of weather and frequency modulation auxiliary electricity prices, and obtained the results that the participation of energy storage equipment would improve the frequency response ability of photovoltaic systems.
From an economic perspective, the photovoltaic array and energy storage system units are combined as a whole for capacity optimization configuration. An independent photovoltaic storage model is constructed with the maximum output power of the photovoltaic system as the goal. Sensitivity analysis is conducted on the charging and discharging time, cycle life, usage efficiency, and other factors of five different types of energy storage batteries. The conclusion is that a hybrid energy storage system constructed with power type and energy type lead-acid batteries has the best economic performance. However, the load pattern in the literature model is fixed and there is no load prediction.
Establish a three-layer grid structure consisting of distribution network, cluster, and node for the division of distribution network side clusters, and establish a two-layer site selection and capacity planning model for distributed photovoltaic and energy storage systems. Apply a two-layer optimization particle swarm optimization algorithm, with the total annual cost as the upper level model objective, and the capacity and power of distributed photovoltaic and energy storage systems at each node as decision variables; The minimum total system loss of the distribution network is the lower level objective function, and the energy storage location and photovoltaic system capacity of each node in the cluster are used as decision variables. The optimal configuration considering comprehensive energy permeability, power permeability, and capacity permeability indicators was obtained.
2.3 Current research status of control methods for energy storage systems
The control method of energy storage system, as the core technology for stable operation of energy storage power plants, has been optimized and studied by a considerable number of literature, and there are also many entry points. From a functional perspective, there are peak shaving and valley filling, wave suppression, auxiliary frequency regulation, etc; From the perspective of scenarios, it can be divided into power generation side, grid side, and user side. The energy storage systems in different scenarios have different functions, so their control methods will also be different.
Starting from multiple dimensions such as power generation side, grid side, energy storage operators, and social value, a benefit model for energy storage participating in grid frequency regulation was established. The least squares method was used to determine various income indicators, and finally, the cost was quantified to obtain the transient investment payback period of the energy storage system.
A comparative analysis was conducted on the performance values of energy storage before and after frequency regulation and the compensation effect of automatic power generation control, verifying the effectiveness of energy storage in improving system operation safety, reducing the risk of exceeding standards, and reducing coal consumption.
A comprehensive multi-dimensional evaluation method was proposed for both single and multiple subjects, using expert scoring as the evaluation indicator. This evaluation method has strong subjectivity and is prone to significant deviation from objective results.
Based on the analysis of the frequency deviation and rate of change requirements of the system after primary frequency regulation, combined with the response time and conversion efficiency of different types of energy storage, the energy storage capacity was determined, and a benefit model for energy storage assisted primary frequency regulation was established. By quantitatively comparing and analyzing the economic benefits of different energy storage under different control methods, the most suitable application scenarios for different energy storage were determined.
Summarized the application prospects of superconducting energy storage technology systems in the field of new energy. Superconducting energy storage technology can directly store electromagnetic energy in magnets, and evaluated the development direction of superconducting energy storage in the current energy storage environment.
An optimized control strategy for large-scale battery energy storage systems to suppress large disturbances in the event of power grid faults has been proposed. A modular battery energy storage frequency regulation and voltage regulation control model adapted to the regulation of the large power grid has been derived based on the relationship between the frequency, voltage, and output active and reactive power of the power grid. The model parameters were optimized for different scale energy storage systems and power grid scheduling needs, improving the flexibility of the control system. The effectiveness of this control strategy in suppressing high-power disturbances under power grid faults was verified using PSASP simulation software, and the final results were as follows:
1) When the power grid malfunctions and the frequency becomes unstable, the battery energy storage model can effectively control the charging and discharging of the energy storage system, maintaining the system frequency around (50 ± 0.2) Hz;
2) When the power grid malfunctions and the voltage fluctuation exceeds the standard operating range, the improved control method of the large-scale battery energy storage system will regulate the bus voltage within the 10% fluctuation range of the standard voltage.
The scenario adopts a hybrid energy storage system composed of supercapacitors and lithium-ion batteries, which aims to balance power and energy application scenarios. The system power is decomposed and redistributed through mathematical methods such as Fourier transform. The allocation method is to distribute high-frequency signals to supercapacitors and low-frequency signals to lithium-ion batteries. Adopting fuzzy control method to regulate the SOC of the energy storage system, thereby achieving power distribution and suppressing fluctuations in the output power of wind power generation.
Consider the strategy research of deep reinforcement learning (DRL) for wind farm power generation. Based on the optimization algorithm of DRL for big data mining, the irregularity of wind power and electricity prices was excavated through data, fully utilizing the advantages of quantity and effective information in big data to avoid the limitations of mathematical optimization methods. However, due to the common end of the DRL method for value functions, their action space still needs to be discretized.
3. Main research content
This article focuses on the research of large-scale battery energy storage systems consisting of photovoltaic power stations and energy storage power stations.
(1) Studied the characteristics of large-scale battery energy storage systems, analyzed centralized and decentralized energy storage structures; Established mathematical models and economic models for photovoltaic power generation and large-scale energy storage systems.
(2) The full life cycle cost and system kilowatt hour cost of electrochemical energy storage stations were calculated. By comparing different energy storage technologies, it was found that lithium iron phosphate battery energy storage stations have the highest comprehensive economic efficiency. Using the improved particle swarm optimization algorithm to obtain the optimal power/energy ratio with the optimization objectives of maximizing the annual net income of the energy storage system and minimizing the amount of solar energy discarded by the photovoltaic power plant. By analyzing the relationship between the cost of electricity purchase and the annual benefits of each power station, taking a project in the Haixi region of Qinghai Province as an example, the electricity purchase price for achieving an annual return rate of over 8% within the lifespan of the energy storage power station is given.
(3) Optimize the active power control strategy for peak shaving and valley filling. Aiming at the effect of energy storage system on load peak shaving and valley filling, adopt a variable parameter power difference control strategy, establish three sub objective functions: peak shaving and valley filling evaluation index, SOC evaluation index, and power supply dissatisfaction. Introduce 2 SOC state control parameters and 4 charging and discharging control parameters, and use adaptive algorithm to scroll optimize these 6 parameters, On the basis of meeting the peak shaving and valley filling effect, the SOC is not exceeded, thus ensuring the healthy state of the energy storage battery.