With the proposal of carbon peaking and carbon neutrality goals in China, the installed capacity of new energy power generation dominated by wind and photovoltaic power has been increasing year by year. However, due to local climatic influences, the output power of new energy power generation is prone to rapid ramps or steep drops, which poses challenges to the frequency regulation margin of the power system. Because of the output power characteristics of new energy generation and the complex grid-connected impedance characteristics, the power system is prone to frequency oscillations under large-scale centralized or distributed grid connections, leading to stability issues and affecting load safety. By integrating battery energy storage systems, it is possible to achieve peak shaving and valley filling for power system loads, alleviate the peak regulation pressure of thermal power units, and act as primary frequency regulation to balance load fluctuations and keep the system frequency within allowable ranges.
Extensive research has been conducted both domestically and internationally on battery energy storage systems. For instance, some studies have introduced the prospects and advantages of battery energy storage systems; others have analyzed the costs and benefits of generation-side and user-side battery energy storage systems; some have investigated the control and management systems to improve stability; and several works have designed supporting battery energy storage system schemes for photovoltaic-storage integrated projects and offshore wind projects, alleviating peak regulation pressure and improving energy utilization. Additionally, research on hybrid battery energy storage system optimization and configuration models has explored complementary advantages of various flexible regulation resources.
To better improve the impact of new energy grid integration on power system stability, this paper analyzes the application and role of battery energy storage systems in grid-connected new energy power generation systems. A specific photovoltaic-storage integrated microgrid project is used as a case study to examine the economic viability of battery energy storage systems.
A typical battery energy storage system installation is illustrated in the figure below, which shows the physical integration of battery modules and power conversion equipment.
Roles of Battery Energy Storage Systems in Grid-Connected New Energy Power Systems
In this section, we analyze the roles of battery energy storage systems in grid-connected new energy power systems that include wind and solar generation.
Peak Shaving and Valley Filling
New energy generation exhibits long-term fluctuations over a day and mismatch with load patterns, i.e., anti-peak characteristics. This increases the required reserve capacity for both upward and downward regulation in the power system. During evening peak hours (typically 19:00–22:00), photovoltaic generation has zero output, while wind generation may achieve full power output during the lowest load point of the day (around 24:00). Consequently, a certain proportion of curtailment occurs due to insufficient transmission capacity. Battery energy storage systems can store the energy from wind generation during low-load periods and release it during evening peaks, effectively time-shifting the electricity. This maximizes the utilization of transmission lines to match load trends and reduces the need for thermal power unit ramping, thereby achieving peak shaving and valley filling.
By equipping new energy generation with battery energy storage systems, the equivalent load (sum of daily load and renewable output) can be constrained within the maximum and minimum effective power limits for grid connection. This avoids curtailment and load shedding, improves the system’s ability to absorb renewable energy, and reduces reserve capacity requirements, enhancing overall system efficiency.
The economic benefit of peak shaving and valley filling depends on the peak-valley electricity price difference and the storage system’s operational strategy. A typical daily charge–discycle schedule is presented in Table 1.
| Time Period | Action | State of Charge (SOC) Target | Power (kW) |
|---|---|---|---|
| 23:00–07:00 (Off-peak) | Charge from grid | 10% to 90% | 250 |
| 07:00–09:00 (Mid-peak) | Idle/self-consumption | 90% to 80% | 0 |
| 09:00–12:00 (Peak) | Discharge to grid | 80% to 20% | 250 |
| 12:00–14:00 (Mid-peak) | Idle | 20% to 15% | 0 |
| 14:00–17:00 (Peak) | Discharge to grid | 15% to 30% (partial) | 150 |
| 17:00–19:00 (Mid-peak) | Charge from PV surplus | 30% to 50% | Variable |
| 19:00–22:00 (Peak) | Discharge to grid | 50% to 10% | 250 |
Stabilizing the Power System
The short-term rate of change of output power from new energy generation must meet system stability requirements. The current grid code mandates limits on active power variations for grid-connected renewable systems, as summarized in Table 2.
| Installed Capacity of Renewable System (MW) | Maximum 10-min Active Power Change (MW) | Maximum 1-min Active Power Change (MW) |
|---|---|---|
| < 30 | 10 | 3 |
| 30 – 150 | 10 – 50 | 3 – 15 |
| > 150 | 50 | 15 |
Smoothing the fluctuation of renewable energy output is achieved by controlling the battery energy storage systems to store or release power, so that the total combined output \(P = P_{BES} + P_{NE}\) satisfies the limits in Table 2, where \(P_{BES}\) is the battery output and \(P_{NE}\) is the renewable output. Two main control algorithms are used: pointwise limitation and low-pass filtering.
Pointwise Limitation Method
In this method, the allowed output of battery energy storage systems at time \(j\) must satisfy:
\[
\max(\Delta P_{10}(j) – P_{y,10},\; \Delta P_{1}(j) – P_{y,1}) < P_{BES}(j) < \min(\Delta P_{10}(j) – P_{y,10},\; \Delta P_{1}(j) – P_{y,1})
\]
where \(\Delta P_{10}(j)\) is the change in combined output over the past 10 minutes relative to the current moment, \(P_{y,10}\) is the maximum allowed 10-minute power deviation, \(\Delta P_{1}(j)\) is the 1-minute change, and \(P_{y,1}\) is the maximum allowed 1-minute deviation.
Low-Pass Filtering Method
The low-pass filter smooths the signal by attenuating high-frequency components. The battery output at time \(j\) is given by:
\[
P_{BES}(j) = \frac{\tau}{t} \left[ \sum P(j) – \sum P(j-1) \right]
\]
where \(\tau\) is the time constant, \(t\) is the control period, and \(\sum P(j)\) is the cumulative sum of battery and renewable outputs up to time \(j\). The time constant relates to the cutoff frequency \(f_c\) of the low-pass filter:
\[
\tau = \frac{1}{2\pi f_c}
\]
Table 3 compares typical performance metrics of these two methods under a simulated 10-minute window with a 30 MW wind farm.
| Control Algorithm | Max 10-min Ramp (MW) | Max 1-min Ramp (MW) | Battery Energy Throughput (kWh) |
|---|---|---|---|
| Pointwise limitation | 8.2 | 2.4 | 95 |
| Low-pass filtering (\(\tau=300s\)) | 7.9 | 2.1 | 87 |
| Low-pass filtering (\(\tau=600s\)) | 6.5 | 1.8 | 112 |
The choice of algorithm depends on the specific operational goals and the state of charge (SOC) management requirements of the battery energy storage systems.
Primary Frequency Regulation
Primary frequency regulation responds to rapid short-term load changes. When the system frequency exceeds a deadband, the battery energy storage systems autonomously inject or absorb active power to support frequency. The deadbands for different generation types vary: thermal units have 50±0.033 Hz, hydro units 50±0.05 Hz, photovoltaic 50±0.06 Hz, and wind 50±0.10 Hz. Battery energy storage systems can respond faster than conventional generators, offering high precision and fast response times.
For a 300 MW thermal unit, the primary frequency regulation capacity is typically 8% of rated capacity (24 MW), with a load adjustment factor of 160 MW/Hz and frequency deviation range of 0.033–0.183 Hz, corresponding to regulation power from 0 to 24 MW. Each over-limit event requires a step change of \(\pm 0.2\%\) of rated power (\(\pm 600\) kW). If battery energy storage systems independently take over this duty, a single charge/discharge event lasts only about 10 seconds. With roughly equal probability of upward and downward events, a 600 kW / 0.5 h battery energy storage system is suitable. Proper SOC management (e.g., maintaining near 50% SOC) ensures shallow cycling and long battery life.
To further reduce required capacity, a dual-boundary improved smoothing control algorithm can be employed, which frequently adjusts the SOC to keep it within a narrow optimal window. Table 4 summarizes the primary frequency regulation performance of battery energy storage systems versus a conventional thermal unit.
| Parameter | Thermal Unit (300 MW) | Battery Energy Storage System (600 kW / 0.5 h) |
|---|---|---|
| Response time to 90% setpoint | ~3–5 s | < 200 ms |
| Deadband | ±0.033 Hz | ±0.02 Hz (adjustable) |
| Maximum power deviation | 24 MW (8% of rated) | 0.6 MW (100% of rating) |
| Sustained duration | Continuously | ~10 s per event |
| Cycle life impact | Negligible | Shallow cycles (≤10% DoD) extend life |
Case Study: A Photovoltaic-Storage Integrated Microgrid Project
We analyzed a specific photovoltaic-storage integrated microgrid project to evaluate the economics of battery energy storage systems. The project consists of an 800 kW photovoltaic system, a 250 kW / 500 kWh lithium iron phosphate (LFP) battery energy storage system, and local loads. The battery storage system operates at a maximum voltage level of 10 kV. When photovoltaic generation exceeds self-consumption, the surplus energy charges the battery; during peak load periods, the battery discharges to the grid.
The main equipment list is given in Table 5.
| Equipment | Specification | Quantity |
|---|---|---|
| PV modules | 550 W monocrystalline silicon | 1455 |
| Inverters | 33 kW rated power | 22 |
| LFP battery cells | 3.2 V / 130 Ah | 1224 |
| Power conversion system (PCS) | 250 kW | 1 |
| 10 kV step-up transformer | 10 kV / 0.4 kV, 800 kVA | 1 |
Key Operational Functions of the Microgrid
Black Start
The microgrid stores surplus photovoltaic energy in the LFP battery energy storage system. In the event of a utility grid outage, the point of common coupling disconnects, and the battery energy storage system initiates a black start, supplying power to loads together with the photovoltaic system without relying on the utility grid.
Voltage-Current Dual Closed-Loop Control
The battery energy storage system output uses two bus sections. When multiple battery strings under one bus operate in voltage-current dual closed-loop mode, the other bus operates under the grid-connected control strategy. This dual-loop mode ensures stable charging/discharging control, maintains DC bus voltage balance, and enhances system stability and efficiency.
Energy Management System (EMS)
The EMS ensures efficient, stable, safe, and reliable operation of the microgrid and optimal utilization of the photovoltaic system. Its functions include data and status monitoring, equipment management, system fault protection, information storage and recording, distribution automation, smart metering, smart consumption, video and environmental monitoring, and comprehensive energy management.
Economic Benefit Calculation of Battery Energy Storage Systems
In this study, the battery energy storage systems operated in a peak-shaving and valley-filling economic mode: charging during off-peak hours and discharging during peak hours. The local peak-valley electricity price difference is 0.7 CNY/kWh. With a depth of discharge (DoD) of 90%, the annual revenue from energy arbitrage is calculated as:
\[
\text{Annual revenue} = 0.7\;(\text{CNY/kWh}) \times 500\;(\text{kWh}) \times 0.9 \times 365 \approx 115,\!000\; \text{CNY}
\]
Additionally, by integrating battery energy storage systems, the microgrid reduces the required transformer capacity, saving the purchase cost of a larger box-type transformer. Table 6 shows the sensitivity of annual revenue to different peak-valley price differences and depths of discharge.
| Peak-Valley Price Difference (CNY/kWh) | Depth of Discharge | Annual Revenue (CNY) |
|---|---|---|
| 0.5 | 90% | 82,125 |
| 0.6 | 90% | 98,550 |
| 0.7 | 90% | 115,000 |
| 0.7 | 80% | 102,200 |
| 0.7 | 100% | 127,750 |
These figures indicate that battery energy storage systems can generate non-negligible revenue while also reducing equipment capital expenditure. The total economic benefit, including both avoided costs and arbitrage income, makes the investment attractive under favorable electricity tariff structures.
Discussion
Battery energy storage systems play a crucial role in assisting grid-connected new energy systems. Their peak shaving, valley filling, and primary frequency regulation capabilities enable the existing power grid to better accommodate renewable generation, reducing the need for thermal reserve capacity and mitigating curtailment. As minute-level forecasting of renewable output improves, the schedulability and predictability of grid-connected renewables increase, enhancing grid friendliness and reducing the demand for fast-ramping resources.
However, under current market mechanisms, solely relying on these functions may not always be economically viable. The benefits can vary significantly with seasonal electricity price patterns. Therefore, further exploring additional roles of battery energy storage systems – such as providing reactive power support, black start, or participating in demand response – or improving battery performance across all seasons, will be important for future research and deployment.
Conclusion
This paper analyzed the application and role of battery energy storage systems in grid-connected new energy power systems. Through a case study of a photovoltaic-storage integrated microgrid project, we evaluated the economic performance of battery energy storage systems. The analysis showed that battery energy storage systems primarily provide peak shaving and valley filling, power system stabilization, and primary frequency regulation. Based on the local peak-valley electricity price difference of 0.7 CNY/kWh, the battery energy storage system operating in the peak-shaving economic mode can achieve an annual revenue of approximately 115,000 CNY. In addition, the system saves electricity costs and reduces the required transformer investment. Thus, the application of battery energy storage systems not only improves the reliability and stability of the power system but also yields tangible economic benefits.