Recently, many regions have experienced consecutive days of scorching heat, pushing grid loads to record highs. At a 200 MW/400 MWh energy storage power station in Xuzhou, Jiangsu Province, the Energy Management System (EMS) is quietly on duty. The real-time data flashing on the large screen captures the countless decisions it makes within milliseconds—adjusting charge and discharge power, smoothing out fluctuations from new energy sources, and responding to grid frequency regulation demands.

As the largest energy storage power station in Jiangsu Province at the time, this project is just one of many undertaken by Nanjing Zhonghui Electric. To date, Nanjing Zhonghui Electric’s energy storage EMS system has managed over 30 GWh of commissioned energy storage facilities, providing comprehensive solutions across various sectors including user-side energy storage, photovoltaic/wind power with integrated storage, and grid-side energy storage.

In recent years, as China’s power grid operations have faced numerous challenges—including a continuous increase in electricity demand, a growing share of intermittent energy sources, and limited peak-shaving capabilities—energy storage technologies, especially peak-shaving power sources, have experienced unprecedented development.

In 2024, “developing new energy storage technologies” was included in the Government Work Report for the first time. As early as 2018, Nanjing Zhonghui Electric had already initiated a project to develop software that would provide centralized monitoring of energy storage power station BMS and PCS systems, enabling unified operation, maintenance, inspection, and management. This software would facilitate rapid fault isolation, relieve grid pressure during peak load periods, enable fast and smooth active power control, and effectively suppress large voltage fluctuations.

This software suite is applied in the fields of energy management for new-energy storage power stations and power control technologies for energy storage systems. Through advanced energy management and control strategies, it rationally schedules and allocates resources among PCS units, battery packs, and other controllable devices, thereby controlling, improving, and enhancing the overall performance and transient characteristics of the energy storage power station. At the same time, it ensures that the charging and discharging characteristics of each converter and battery pack are balanced and optimized.

Since this summer, sustained high temperatures have repeatedly pushed the power grid’s load to its capacity limit. At 12:35 on August 20, 2025, Jiangsu’s power grid reached a peak electricity load of 155 million kilowatts—the fourth time this summer that the load has broken the historical record—representing a year-on-year increase of 5.99% over last year’s peak. Meanwhile, the instability of new energy generation has been further exacerbated under extreme weather conditions.

Based on more than a decade of experience in the development of secondary equipment for new-energy power systems, Nanjing Zhonghui Electric has developed an energy storage management system—a next-generation monitoring and management platform specifically designed for energy storage power stations. This system adopts a hierarchical architecture featuring a “platform-plus-application” design and boasts powerful cross-platform capabilities, enabling it to run on Windows, Linux, Unix, and domestically produced operating systems. Furthermore, Nanjing Zhonghui Electric’s energy management system is centered around EMS software for energy storage and is complemented by a range of hardware support devices, including direct-supply primary frequency regulation units, power control units, protocol conversion units, microgrid coordination units, coordinated controllers, emergency frequency and voltage protection devices, and anti-islanding protection devices, thereby providing precise and tailored solutions.

Graph-based Multi-level Fault Diagnosis and Tracking Technology

Using graph modeling technology, we construct a topological relationship map within the energy storage system (cell ↔ battery cluster ↔ battery pack ↔ PCS ↔ EMS ↔ grid connection point). In the event of a fault, the system can automatically trace the source of the fault, pinpoint the faulty link, and analyze the scope of the fault’s impact.

Key advantages:

● Implement fault path inversion “from symptoms to root causes”;

● Can quickly determine whether a PCS fault is associated with a specific battery cluster or a communication link anomaly.

● Supports graph visualization for easy presentation and delivery.

Multidimensional Loss Calculation and Dynamic Efficiency Analysis

By collecting in real time the charge and discharge data from various PCSs, transformer operating parameters, and power data at grid connection points, the system can perform dynamic statistical analysis and trend analysis of line losses, transformer losses, and inverter conversion efficiency.

Key advantages:

● Supports time-of-use loss assessment (e.g., whether line losses during peak periods are higher than those during off-peak periods);

● Can convert daily or annual energy consumption into economic losses, helping enterprises optimize energy efficiency.

● Able to identify abnormal energy paths (such as abnormal discharge or reduced conversion efficiency in a particular branch circuit).

Battery Health Assessment and Dynamic Lifetime Prediction

Real-time tracking of operational data such as battery SOC curves, charge-discharge frequency, and temperature deviations, combined with cell models and degradation trend prediction, to generate SOH (State of Health) curves and RUL (Remaining Useful Life) prediction reports.

Key advantages:

● Warning of single-cell aging or capacity degradation;

●Can be used for maintenance plan development and asset valuation;

● Dynamically adjust the maximum charge and discharge depth in coordination with the control strategy.

Intelligent Demand Control and Peak-Shaving Scheduling Module

An integrated load forecasting model (such as LSTM and XGBoost) combined with a sliding-window average power detection logic enables dynamic scheduling of strategies to achieve maximum load reduction within 15 minutes. Additionally, it provides peak-shaving benefit assessments and plots of reduction curves.

Key advantages:

● Provide estimates of demand-based electricity tariff savings and “heatmap” for peak-shaving;

● Supports self-calibration of prediction errors and multi-strategy switching;

● Can be linked to the steel plant’s monthly energy consumption targets and automatically adjust peak-shaving intensity.

Energy Storage Mirror Compensation Control and Stable Regulation of Electricity Purchased Online

Real-time power consumption from online shopping is collected, and the energy storage system adjusts its charging and discharging in a “mirror-image” manner to automatically keep the online shopping power consumption close to the set target (e.g., 10 MW), thereby buffering against impulsive loads.

Key advantages:

● Features millisecond-level response capability (can be directly controlled by the GOOSE protocol via PCS);

● Reduce gas flaring and grid reverse power flow issues, and improve the utilization rate of self-generated power.

● It can be linked with a slow controller to achieve integrated power generation and supply balancing through “fast adjustment + slow adjustment.”

Quality creates the future, and innovation transforms the world. As the construction of a new power system continues to deepen, energy storage EMS systems will become critical infrastructure in the energy transition. The corporate philosophy of Nanjing Zhonghui Electric is now being translated into concrete actions that ensure grid security, providing vital support for building a clean, low-carbon, safe, and efficient energy system.