Hybrid energy storage assisted frequency modulation method and system for thermal power generating unit

Abstract

The application provides a hybrid energy storage auxiliary frequency modulation method and system for thermal power generating units, obtains power grid frequency modulation demand instructions and real-time operation data of thermal power generating units, calculates a dynamic power gap; performs frequency domain decomposition on the dynamic power gap, allocates a high-frequency disturbance component to a super capacitor group for preferential response, allocates a medium-low frequency regulation component to a lithium battery group for support, and utilizes the lithium battery group to perform dynamic state of charge balancing for the super capacitor group during a frequency modulation interval; finally, through a synchronous phasor measurement device PMU, time stamp alignment and vector superposition operation are performed on bottom layer electrical data, and combined output is sent to a power grid. The application realizes accurate physical shunting of a frequency modulation task and internal energy self-balancing of energy storage, greatly prolongs the service life of the energy storage system, eliminates power synthesis errors from the bottom layer, and significantly improves the response speed and control accuracy of the combined frequency modulation of the fire and energy storage.

Description

Technical Field

[0001] This invention relates to the field of frequency regulation of thermal power units, and in particular to a method and system for frequency regulation of thermal power units assisted by hybrid energy storage. Background Technology

[0002] During the construction of the new power system, the widespread introduction of new energy power systems has led to high-frequency and bidirectional fluctuations in the power grid frequency, posing a severe challenge to traditional thermal power units when participating in frequency regulation. Frequent and large-scale load changes over a long period result in insufficient frequency regulation capability, slow response speed, and low regulation accuracy for thermal power units, making it difficult to track frequency deviations caused by new energy fluctuations in real time. Rapid load adjustments directly lead to increased coal consumption and carbon emissions, and under special operating conditions, can cause significant overheating of boiler tubes, seriously threatening the safe operation of the unit. Relying solely on the inherent conditions of the thermal power unit itself to achieve the required rapid load change rate through control optimization is extremely difficult, because the inherent characteristics of the pulverizing system cannot support rapid energy replenishment on the boiler side, and important auxiliary equipment is in a critical operating state during low-load periods, making rapid action extremely risky.

[0003] Introducing energy storage to assist in frequency regulation of thermal power units has become an inevitable path to solve the aforementioned bottlenecks. Some solutions have been proposed in the current technical field. For example, existing technology CN120109838A discloses a frequency regulation method based on hybrid energy storage units. This method involves establishing a hybrid energy storage frequency regulation platform to collect output power and simulate the power conditions to be responded to, then matching the optimal output power data and inputting it into the platform for frequency regulation.

[0004] The prior art CN120638396A discloses a method and system for joint frequency regulation control of thermal power and energy storage, which uses DCS, RTU and PMU to collect data in a coordinated manner, determines the AGC mode and generates dynamic compensation commands, adopts a hybrid architecture of supercapacitor and lithium iron phosphate battery and implements cell-level overcharge and over-discharge protection.

[0005] Existing technology CN121097844B discloses a method for optimizing frequency regulation of thermal power coupled with hybrid energy storage of lithium batteries and supercapacitors. This method decomposes the frequency regulation signal into multiple time scales, combines fuzzy power allocation based on state of charge, and performs adaptive droop control. While these existing technologies utilize hybrid energy storage to smooth power fluctuations, they generally lack a deep control strategy for autonomous energy balancing during frequency regulation intervals. They also fail to establish a vector synchronization superposition mechanism based on strict timestamp alignment at the underlying hardware level, and are limited by rigid point-to-point connections between single units, making it impossible to achieve flexible and dynamic switching of physical topologies between multiple units at the plant electrical system level. Summary of the Invention

[0006] The main objective of this invention is to provide a hybrid energy storage-assisted frequency regulation method and system for thermal power units. This application solves the technical problems of delayed response of thermal power units when participating in grid frequency regulation, lack of underlying vector synchronization superposition mechanism, flexible switching capability of multiple unit physical topology, and autonomous energy balance control during frequency regulation intervals in existing thermal power-storage joint regulation systems.

[0007] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for frequency regulation of a hybrid energy storage-assisted thermal power unit, comprising the following steps: S1. Obtain grid frequency regulation demand commands and real-time operating data of thermal power units; S2. Calculate the dynamic power gap based on the frequency regulation demand command and the real-time operating data of the thermal power unit; S3. Charge and discharge compensation is performed on a hybrid energy storage system based on dynamic power gap control. The hybrid energy storage system includes a supercapacitor pack and a lithium battery pack. S4. The output of the combined energy storage system and the output of the thermal power unit are fed into the grid.

[0008] In the preferred embodiment, step S1, which involves obtaining the grid frequency regulation demand command and the real-time operating data of the thermal power unit, includes: Establish communication connections with the power grid dispatching WAMS system and remote control units (RTUs); Receive scheduling message data packets through the communication connection; Parse the header identifier bits of the scheduling message data packet and extract the instruction type feature code; By comparing the instruction type feature code with the preset feature library, the power grid frequency regulation demand instruction can be identified as a primary frequency regulation demand instruction or an AGC frequency regulation demand instruction. When a frequency regulation demand command is identified, the grid frequency parameters and the power output of the thermal power unit are collected. When the AGC frequency regulation demand command is identified, the target command is received through the remote control unit (RTU) and the actual output of the thermal power unit is collected.

[0009] In the preferred embodiment, step S2, which involves calculating the dynamic power gap in response to the primary frequency regulation demand command, includes: Calculate the theoretical target value for primary frequency regulation based on grid frequency parameters and thermal power unit capacity parameters; The difference between the theoretical target value of the primary frequency regulation and the terminal power of the thermal power unit is calculated. The result of the difference operation is used as the dynamic power gap for primary frequency modulation.

[0010] In the preferred embodiment, step S2, which involves calculating the dynamic power gap in response to the AGC frequency modulation demand command, includes: A combined frequency regulation strategy of storage and generator is adopted; The steps of the joint frequency regulation strategy for energy storage and power generation are to keep the original DCS control logic and operation mode of the thermal power unit unchanged. The AGC command and real-time operating data of the thermal power unit are simultaneously forwarded to the energy storage control subsystem. The target power value is extracted from the AGC command through the energy storage control subsystem. ; Synchronously acquire the current power value of the actual output of thermal power units ; Calculate the target power value Compared with the current power value Real-time difference between ; Real-time difference As the dynamic power gap of AGC.

[0011] In the preferred embodiment, step S3, which involves charging and discharging compensation of the hybrid energy storage system based on dynamic power gap control, includes: A multi-objective cooperative algorithm is used to decompose the dynamic power gap in the frequency domain to obtain the high-frequency disturbance component and the mid-to-low frequency regulation component; The specific implementation steps of the multi-objective cooperative algorithm are as follows: extract the rate of change of the dynamic power gap signal and set a threshold for the rate of change; The power demand component with a rate of change greater than the rate of change threshold is classified as a high-frequency disturbance component. The power demand component with a rate of change less than or equal to the rate of change threshold is divided into low- and medium-frequency regulation components. The first control command is generated to allocate the high-frequency disturbance component to the supercapacitor bank for a rapid and priority response. The second control command is generated to allocate the low- and medium-frequency regulation components to the lithium battery pack to provide long-term power support.

[0012] In the preferred embodiment, step S3, which involves controlling the hybrid energy storage system to perform charge and discharge compensation, further includes dynamic state-of-charge (SOC) balancing algorithm control. Real-time monitoring of the state of charge of supercapacitor banks and lithium battery banks; Define the target standby state quantity of the supercapacitor bank as: ; Calculate the current state of charge and target reserve quantity of the supercapacitor bank. The difference yields the state of charge deviation. ; When the state of charge deviation is detected When the set safety range is exceeded, the dynamic grouped state of charge equalization control strategy is triggered. Monitor the frequency modulation gap interval, which is the time period when the dynamic power gap is zero; Calculate the available remaining energy of the lithium battery pack within the frequency modulation interval; The remaining energy from the lithium battery pack can be used to recharge the supercapacitor pack, restoring the supercapacitor pack's state of charge to the target reserve level. nearby.

[0013] In the preferred embodiment, step S4, which involves combining the output of the hybrid energy storage system with the output of the thermal power unit and transmitting it to the grid, includes: A synchronous phasor measurement device (PMU) is deployed on the 6 kV plant service busbar on the side of the thermal power unit. The three-phase AC voltage and current data of the hybrid energy storage system and the thermal power unit are collected in real time through the synchronous phasor measurement device (PMU). The three-phase AC voltage and current data of the hybrid energy storage system are synchronized and aligned with the three-phase AC voltage and current data of the thermal power unit. Inside the synchronous phasor measurement unit (PMU), the aligned voltage and current data are vector superimposed according to Kirchhoff's laws. The total output active power and reactive power are obtained after vector superposition. The total output active and reactive power are encapsulated into standard communication messages and uploaded to the power grid dispatch WAMS system as a joint output result for dispatch assessment.

[0014] In the preferred embodiment, the method is deployed and used through the following software and modules: The energy management system (EMS) software and energy storage coordination controller module are independently deployed on a Linux operating system edge computing server using Docker containerization technology. Write a Dockerfile to define runtime environment dependencies and package the Energy Management System (EMS) software into a Docker image; Start the Energy Management System (EMS) software container using the container orchestration tool; Establish a hardware communication connection between the Energy Management System (EMS) software and the existing Remote Terminal Units (RTUs) in the thermal power plant, and complete the configuration for forwarding control commands. A multi-objective collaborative algorithm logic is embedded in the energy storage coordinating controller module, which sends Modbus TCP control signals to the battery management system (BMS) to drive hardware actions.

[0015] In the preferred embodiment, the actual deployment and use steps of the hybrid energy storage system also include the deployment of the primary system electrical architecture: The supercapacitor bank and lithium battery bank are connected to the 6 kV plant bus of the thermal power unit through the power conversion system PCS; Multiple physical switching modules are deployed between the 6kV plant service bus and the power conversion system PCS; By switching the physical switching switch module between closed and open states using logic control commands, the hybrid energy storage system can flexibly switch between states of full-station energy storage shutdown, energy storage unidirectionally assisting the frequency regulation of Unit 1, energy storage unidirectionally assisting the frequency regulation of Unit 2, and full-station energy storage jointly assisting the frequency regulation of the target unit.

[0016] A hybrid energy storage-assisted frequency regulation system for thermal power units, the system comprising: The data acquisition module is used to acquire power grid frequency regulation demand commands and real-time operating data of thermal power units; The power gap calculation module is used to calculate the dynamic power gap based on the frequency regulation demand command and the real-time operating data of the thermal power unit. The energy storage charge and discharge control module is used to control the charge and discharge compensation of the hybrid energy storage system based on the dynamic power gap. The hybrid energy storage system includes a supercapacitor pack and a lithium battery pack. The output synthesis and transmission module is used to synthesize the output of the hybrid energy storage system and the output of the thermal power unit and transmit them to the grid side.

[0017] This invention provides a method and system for frequency regulation of thermal power units assisted by hybrid energy storage. The proposed method and system utilizes precise calculation of dynamic power gaps and a multi-objective collaborative algorithm to decompose frequency regulation requirements in the frequency domain. Supercapacitor banks are used to quickly and preferentially respond to high-frequency disturbance components, while lithium battery banks handle mid- and low-frequency regulation components. This collaborative mechanism effectively mitigates the mechanical and thermal stress impacts of high-frequency grid fluctuations on thermal power units, significantly improves the unit's frequency tracking accuracy and response speed, and reduces the additional coal consumption and physical wear and tear on equipment caused by frequent grid ramping.

[0018] This application introduces a dynamic SOC balancing algorithm into the control logic. By monitoring the SOC status of the supercapacitor bank and the lithium battery bank in real time, it cleverly utilizes the time interval between frequency regulation intervals when the dynamic power gap is zero to control the lithium battery bank to use its available remaining energy to recharge the supercapacitor bank. The internal energy self-balancing mechanism enables the supercapacitor bank to quickly and seamlessly recover to the target reserve state, greatly improving the system's full-power reserve capability in response to continuous high-frequency regulation commands. This avoids the bottleneck of individual energy depletion caused by frequent calls in traditional energy storage systems and significantly extends the cycle life of the overall hybrid energy storage system.

[0019] This application deeply optimizes the underlying data processing and communication architecture. A PMU device is deployed on the busbar of the thermal power unit to strictly synchronize and align the three-phase AC voltage and current data between the hybrid energy storage system and the thermal power unit. High-precision vector superposition calculations are performed internally in the hardware according to Kirchhoff's laws. This underlying hardware-level vector synthesis method completely eliminates the time delay and asynchronous data errors caused by traditional network transmission and software calculations, ensuring that the combined output results sent to the WAMS system have extremely high authenticity and synchronization, fundamentally improving the response performance of thermal power plants in grid dispatch.

[0020] This application achieves high flexibility and reliability in system deployment and electrical topology design. By independently deploying the EMS software on an edge computing server using Docker containerization technology, it achieves lightweight control system with high portability. Multiple physical switching modules are deployed between the plant service bus and the PCS, enabling seamless dynamic switching between various states of the hybrid energy storage system, including full-site shutdown, unidirectional auxiliary to Unit 1, unidirectional auxiliary to Unit 2, and joint auxiliary to the entire site, through logic control commands. The flexible coupled electrical architecture breaks the traditional physical limitation of strong binding between energy storage and a single unit, allowing multiple thermal power units within the plant to dynamically share energy storage resources, maximizing the overall utilization value of energy storage assets and the comprehensive frequency regulation capability of the entire site. Attached Figure Description

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the data acquisition and instruction recognition process of this invention; Figure 2 This is a flowchart of the dynamic power gap calculation of the present invention; Figure 3 This is a flowchart of the frequency domain decomposition and SOC equalization control of the present invention; Figure 4 This is a flowchart of the output vector synthesis and transmission process of the present invention; Figure 5 This is a diagram of the hybrid energy storage-assisted frequency regulation system for thermal power units according to the present invention. Detailed Implementation

[0022] Example 1 like Figure 1-5 As shown, a method for frequency regulation of a hybrid energy storage-assisted thermal power unit includes the following steps: S1. Obtain grid frequency regulation demand commands and real-time operating data of thermal power units; S2. Calculate the dynamic power gap based on the frequency regulation demand command and the real-time operating data of the thermal power unit; S3. Charge and discharge compensation is performed on a hybrid energy storage system based on dynamic power gap control. The hybrid energy storage system includes a supercapacitor pack and a lithium battery pack. S4. The output of the combined energy storage system and the output of the thermal power unit are fed into the grid.

[0023] In the preferred embodiment, step S1, which involves obtaining the grid frequency regulation demand command and the real-time operating data of the thermal power unit, includes: Establish communication connections with the power grid dispatching WAMS system and remote control units (RTUs); Receive scheduling message data packets through the communication connection; Parse the header identifier bits of the scheduling message data packet and extract the instruction type feature code; By comparing the instruction type feature code with the preset feature library, the power grid frequency regulation demand instruction can be identified as a primary frequency regulation demand instruction or an AGC frequency regulation demand instruction. When a frequency regulation demand command is identified, the grid frequency parameters and the power output of the thermal power unit are collected. When the AGC frequency regulation demand command is identified, the target command is received through the remote control unit (RTU) and the actual output of the thermal power unit is collected.

[0024] The specific implementation process of obtaining the power grid frequency regulation demand command and the real-time operation data of thermal power units in step S1 involves multiple rigorous logical layers. The system first establishes communication connections with the Wide Area Measurement System (WAMS) and the Remote Terminal Unit (RTU). The WAMS is responsible for providing high-precision global dynamic data of the power grid based on synchronous phasor measurements. The RTU acts as a data gateway between local equipment and the dispatch center, undertaking the task of converting underlying hard-wired signals to upper-layer communication protocols. The system completes a handshake protocol with the two core nodes through a dedicated physical channel to ensure the stability and real-time performance of the data transmission channel.

[0025] After the communication connection is established, the system receives dispatch message data packets through the communication connection. Dispatch message data packets are standardized data carriers that encapsulate the power grid control intent. The system then parses the header identifier bits of the dispatch message data packets to extract the instruction type feature code. The extracted instruction type feature code is defined as... The frame header identifier, located at the beginning of the data packet, contains a key sequence of numbers indicating the packet priority and instruction type. Extract the instruction type feature code. The process can quickly extract core control commands from massive background messages, significantly reducing the computational overhead of subsequent data processing.

[0026] The system then compares the instruction type signature. The system identifies either primary frequency regulation (PMR) requests or Automatic Generation Control (AGC) frequency regulation requests based on a pre-defined feature library. The pre-defined feature library is a mapping table stored in the system's local memory, containing a strict mapping relationship between different instruction types and their corresponding standard feature codes. The system uses a fast matching algorithm to look up and compare the extracted feature codes. If the matching result points to the first type of feature interval, the current PMR request is determined to be a primary PMR request. If the matching result points to the second type of feature interval, the current PMR request is determined to be an AGC frequency regulation request. This precise conditional branching logic ensures that the system calls the appropriate data acquisition channel according to different frequency regulation application scenarios.

[0027] When the system recognizes a frequency regulation request command, the action logic immediately switches to the high-frequency data acquisition channel, which has extremely high real-time requirements, to collect grid frequency parameters and thermal power unit terminal power. The real-time acquired grid frequency parameters are defined as follows: Define the rated frequency parameter of the power grid as follows: Define the terminal power of thermal power units as In a single-mode frequency regulation operation, the system needs to sense minute fluctuations in the power grid frequency in real time to calculate the frequency deviation. The frequency deviation value is defined as... The computational logic satisfies the formula: ; in the formula This represents the absolute difference between the actual frequency parameters of the power grid and their standard rated state, reflecting the degree of instantaneous power imbalance in the grid. Only by obtaining extremely accurate data... and Only then can the system determine whether the unit needs to take action within milliseconds, which provides the most basic objective input source for quickly smoothing out transient changes in grid frequency.

[0028] When the system recognizes an Automatic Generation Control (AGC) frequency regulation demand command, the action logic switches to the tracking control channel, receiving the target command and collecting the actual output of the thermal power unit via the Remote Terminal Unit (RTU). The target power value included in the target command is defined as... The actual output of a thermal power unit is defined as... At this point, the system primarily focuses on the macro-level output targets issued by the dispatch center, as well as the current physical output levels achieved by the generating units. This is achieved by acquiring... and With two key operational status data points, the system can accurately assess the steady-state deviation between the unit's current output and the scheduling expectation, providing a reliable calculation basis for performing long-term power regulation at medium and low frequencies.

[0029] The aforementioned implementation steps for acquiring grid frequency regulation demand commands and real-time operational data have significant beneficial effects. By deeply analyzing the frame header identifier bits and performing precise feature code comparison, the system can achieve physical separation of primary frequency regulation services and automatic generation control (AGC) frequency regulation services in the initial reception stage of command execution. This completely avoids command confusion and execution channel blockage under complex grid operating conditions, greatly improving the real-time performance of signal processing and the accuracy of system response. Adaptively switching the corresponding data acquisition object for different frequency regulation demands ensures that the core processor of the control system is not occupied by redundant data on the computing bus, guaranteeing the most efficient utilization of underlying hardware computing resources. This multi-level analysis and adaptive acquisition design guarantees the absolute reliability of frequency regulation judgment from the data source, effectively eliminating the risk of unit malfunctions caused by communication delays or command misreading, and laying a solid data foundation for the safe and efficient frequency domain compensation of thermal power units by hybrid energy storage systems.

[0030] In the preferred embodiment, step S2, which involves calculating the dynamic power gap in response to the primary frequency regulation demand command, includes: Calculate the theoretical target value for primary frequency regulation based on grid frequency parameters and thermal power unit capacity parameters; The difference between the theoretical target value of the primary frequency regulation and the terminal power of the thermal power unit is calculated. The result of the difference operation is used as the dynamic power gap for primary frequency modulation.

[0031] The process of calculating the dynamic power gap in response to the primary frequency regulation demand command in step S2 is the core calculation step of the entire control strategy, providing a data guideline for the precise operation of the hybrid energy storage system. The system first calculates the theoretical target value for primary frequency regulation based on the grid frequency parameters and the thermal power unit capacity parameters. The obtained grid frequency parameters are defined as follows: Define the capacity parameters of thermal power units as follows: To derive the theoretical target that conforms to power grid dispatching specifications, the system internally calls the primary frequency regulation response mathematical model for real-time analysis. The theoretical target value for primary frequency regulation is defined as... Define the rated frequency parameter of the power grid as follows: The first frequency modulation response coefficient is defined as Define the basic operating power parameters of thermal power units as follows: The specific calculation logic satisfies the formula: ; In the above calculation formula, and The difference represents the absolute degree to which the current power grid frequency deviates from the standard rated state. This frequency deviation result is then compared with the thermal power unit capacity parameters. and the set primary frequency response coefficient By multiplying these parameters consecutively, the theoretical power regulation increment that thermal power units should undertake to mitigate current grid frequency fluctuations can be calculated. (Base operating power parameters of thermal power units) This represents the unit's steady-state active power output at the moment before the frequency abrupt change. By algebraically superimposing the basic operating power parameters with the calculated power regulation increment, we can obtain the absolute total power that the unit should output under ideal, delay-free conditions under the current transient grid frequency state, which is the theoretical target value of primary frequency regulation. .

[0032] After obtaining the theoretical support standard, the system then calculates the difference between the theoretical target value of primary frequency regulation and the terminal power of the thermal power unit, and directly uses the result of the difference calculation as the dynamic power gap for primary frequency regulation. The real-time acquired terminal power of the thermal power unit is defined as... Define the dynamic power gap of primary frequency modulation as The specific difference calculation for dynamic deviation resolution satisfies the following formula: ; In this formula, the theoretical target value for primary frequency modulation is... This represents the ideal immediate support power that the power grid dispatching level expects this generation node to provide, while the power output of the thermal power unit is... This refers to the actual transient active power output of the unit, constrained by physical limitations such as the huge rotational inertia of the turbine rotor, the mechanical delay of the speed regulating valve actuator, and the lag in the pressure response of the boiler thermal system. The difference between the two is calculated in real time. The instantaneous energy deficit between the ideal demand of the power grid and the actual response of thermal power is accurately and dynamically quantified. The system strictly sets the result of this difference calculation as the dynamic power gap for primary frequency regulation and uses it as the reference input for the subsequent instruction generation module to drive the hybrid energy storage system.

[0033] The specific implementation steps for calculating the dynamic power gap in primary frequency regulation demand commands described above have significant beneficial effects. By introducing thermal power unit capacity parameters and real-time grid frequency parameters to construct a rigorous theoretical target mathematical model, the system can accurately map the absolute frequency regulation responsibility boundary suitable for units of different capacity specifications. This avoids the risk of grid frequency under-regulation or over-regulation that is easily caused by fixed power compensation strategies from the algorithmic level, providing substantial and feasible technical support for the dynamic power gap feature in the claims. By performing high-frequency real-time difference calculation between the ideal target calculation value and the actual generated power at the generator end, which includes physical lag characteristics, the high-frequency power gap that the thermal power unit itself cannot instantly compensate for can be perfectly isolated. This closed-loop differential extraction method based on physical feedback enables the control system to direct energy storage devices to precisely fill the energy gap only for the untimely response of the thermal power unit. This not only avoids wasting the valuable capacity of the hybrid energy storage system but also fully utilizes the basic adjustment capabilities of the thermal power unit in subsequent follow-up operations. It completely solves the problem of secondary frequency drops caused by slow response in primary frequency regulation of traditional single thermal power units, significantly improving the frequency regulation quality of the entire plant's grid connection point.

[0034] In the preferred embodiment, step S2, which involves calculating the dynamic power gap in response to the AGC frequency modulation demand command, includes: A combined frequency regulation strategy of storage and generator is adopted; The steps of the joint frequency regulation strategy for energy storage and power generation are to keep the original DCS control logic and operation mode of the thermal power unit unchanged. The AGC command and real-time operating data of the thermal power unit are simultaneously forwarded to the energy storage control subsystem. The target power value is extracted from the AGC command through the energy storage control subsystem. ; Synchronously acquire the current power value of the actual output of thermal power units ; Calculate the target power value Compared with the current power value Real-time difference between ; Real-time difference As the dynamic power gap of AGC.

[0035] In step S2, during the calculation of the dynamic power gap in response to the frequency regulation demand command of the Automatic Generation Control (AGC), the system adopts a combined energy storage and turbine frequency regulation strategy. The core implementation step of this strategy is to maintain the original distributed control system (DCS) control logic and operating mode of the thermal power unit completely unchanged. The DCS is the underlying core hub for automating the production process in a thermal power plant, directly controlling the coordinated operation of heavy physical equipment such as boilers, turbines, and generators. By overlaying bypass control nodes into the external communication network, the system avoids any intrusive modifications to the original DCS underlying control code and boiler / turbine safety interlock logic of the thermal power unit. This non-intrusive, decoupled underlying design concept ensures the absolute safety and stability of the core basic power generation operations of the thermal power unit to the greatest extent possible.

[0036] Based on the aforementioned energy storage and generator joint frequency regulation control architecture, the system simultaneously forwards the Automatic Generation Control (AGC) commands issued by the dispatch center and the real-time operating data of the thermal power units to the independent energy storage control subsystem. The energy storage control subsystem, as an independent edge computing unit specifically designed to handle high-frequency response requirements and coordinate the energy storage resources across the entire station, undertakes the core task of accurately identifying transient power gaps. Subsequently, the system extracts the target power value contained in the AGC commands through the parsing module within the energy storage control subsystem. The extracted target power value is defined as... Target power value This represents the absolute active power output benchmark level that the grid-connected thermal power unit should strictly achieve within the current extremely short adjustment cycle, based on the real-time frequency fluctuations of the power grid and the active power load balance requirements of the entire grid. Simultaneously, the system synchronously acquires the current power value of the actual output of the thermal power unit through high-precision measurement sensors. The acquired current power value is defined as... Current power value It accurately and objectively reflects the total amount of transient active power actually transmitted from the generator end to the main grid after being constrained by multiple physical factors such as the huge rotor inertia of the steam turbine, the mechanical action delay of the speed regulating valve, and the pressure lag of the boiler thermal system.

[0037] After simultaneously acquiring the two key power parameters mentioned above, the microprocessor inside the energy storage control subsystem executes high-frequency difference calculation logic to calculate the target power value. Compared with the current power value The real-time difference between them. The calculated real-time difference is defined as... The calculation logic for this power difference satisfies the following mathematical formula: ; In the above mathematical formula, This represents the ideal active power target benchmark that the dispatcher expects to issue. This represents the actual physical response state of the thermal power unit's current physical system. By performing high-frequency, real-time algebraic subtraction between the target baseline and the actual physical state, the system can accurately isolate the transient active power portion that the thermal power unit fails to provide in time due to its own physical ramp-up rate limit. The calculated real-time difference is... It possesses a very clear physical indication meaning, essentially directly quantifying the instantaneous energy lag deficit of thermal power units during the process of tracking the dispatch command curve. The system will calculate the real-time difference. Strictly defined as the final dynamic power gap for the automatic generation control (AGC) frequency regulation operation, and used as the sole quantitative input command to drive the downstream hybrid energy storage system to perform charge and discharge compensation actions.

[0038] The specific implementation steps described above for calculating the dynamic power deficit in response to AGC (Automatic Generation Control) frequency regulation demand commands have significant and irreplaceable benefits. First, the energy storage-following-machine strategy, which maintains the completely independent operation of the distributed control system (DCS) of thermal power units, greatly reduces the engineering implementation difficulty and underlying equipment safety risks of traditional thermal power plants participating in energy storage-based frequency regulation retrofits, avoiding the extremely complex resetting of boiler-turbine coordination parameters and system shutdown testing. Second, by completely delegating and isolating the core tasks of command parsing and power difference calculation to a dedicated energy storage control subsystem, an independent data closed loop with extremely high response speed is constructed, completely eliminating data calculation delays caused by communication congestion in the control network within traditional thermal power plants. Finally, by acquiring the ideal target power and actual physical output in real time and performing high-frequency mathematical subtraction, the control system can continuously and extremely accurately pinpoint the dynamic energy deficit region of the thermal power unit during the tracking of AGC scheduling commands. This dynamic gap calculation mechanism, which perfectly converts the mechanical hysteresis of thermal power generation into the electrochemical energy storage compensation target, enables the hybrid energy storage system to have millisecond-level precise compensation capabilities. It accurately fills the tiny energy gap between the output curve of the thermal power unit and the ideal scheduling target curve, thereby multiplying the overall regulation rate and response regulation accuracy of the entire plant's grid-connected control point without changing the hardware performance of the thermal power unit itself.

[0039] In the preferred embodiment, step S3, which involves charging and discharging compensation of the hybrid energy storage system based on dynamic power gap control, includes: A multi-objective cooperative algorithm is used to decompose the dynamic power gap in the frequency domain to obtain the high-frequency disturbance component and the mid-to-low frequency regulation component; The specific implementation steps of the multi-objective cooperative algorithm are as follows: extract the rate of change of the dynamic power gap signal and set a threshold for the rate of change; The power demand component with a rate of change greater than the rate of change threshold is classified as a high-frequency disturbance component. The power demand component with a rate of change less than or equal to the rate of change threshold is divided into low- and medium-frequency regulation components. The first control command is generated to allocate the high-frequency disturbance component to the supercapacitor bank for a rapid and priority response. The second control command is generated to allocate the low- and medium-frequency regulation components to the lithium battery pack to provide long-term power support.

[0040] Step S3, which involves controlling the charging and discharging compensation of the hybrid energy storage system based on the dynamic power gap, is a core control step that enables complementary advantages and precise power delivery from different types of energy storage media. The system first employs a multi-objective cooperative algorithm to perform frequency domain decomposition on the dynamically calculated power gap, thereby obtaining high-frequency disturbance components and mid-to-low-frequency regulation components. The dynamically calculated power gap signal sequence is defined as follows: The high-frequency disturbance component decomposed is defined as The decomposed low- and mid-frequency adjustment components are defined as follows: The mathematical and physical significance of frequency domain decomposition lies in the fact that the actual power grid gap is often a composite signal consisting of rapidly fluctuating transient spikes and slowly changing steady-state trends. Directly inputting this undivided composite signal into a single energy storage device would cause the device to frequently switch between charging and discharging states in a very short time, resulting in extremely serious electrochemical damage. Therefore, the primary goal of multi-objective collaborative algorithms is to accurately match the adjustment requirements of different frequency characteristics to energy storage hardware with corresponding physical properties through signal decomposition.

[0041] The specific implementation steps of the multi-objective cooperative algorithm are as follows: extract the dynamic power gap signal sequence. The signal change rate is defined, and a change rate threshold is set. The signal change rate is defined as... Define the set rate of change threshold as Signal rate of change This represents the magnitude of the power deficit change within a very short time interval, and is a real-time discretized solution for the slope of the power deficit curve. The specific calculation logic satisfies the formula: ; In the above formula, Represents the current moment The dynamic power deficit value, Represents the previous sampling time The dynamic power deficit value, This represents an extremely short discrete sampling time interval. The formula calculates the absolute rate of change, representing the intensity of the signal, by dividing the absolute value of the power gap difference between two adjacent sampling points by the sampling time interval. Rate of change threshold This is an empirical constant pre-calibrated through offline simulation and engineering experience before the system is put into operation, based on the transient power limit of the selected supercapacitor group and the cycle life loss tolerance of the lithium battery group, and serves as a hard decision boundary for dividing high and low frequency components.

[0042] The rate of change of the signal is calculated continuously in real time. Subsequently, the system classifies the power demand component with a rate of change greater than the rate of change threshold as high-frequency disturbance components. The power demand portion with a rate of change less than or equal to the rate of change threshold is divided into low- and medium-frequency regulation components. The specific partitioning logic satisfies the following piecewise function: when At that time, it is determined that the adjustment demand at that moment falls into the category of high-frequency, rapid changes. ; ; when At that time, it is determined that the adjustment demand at that moment belongs to the range of slow changes in the low to mid frequency range. ; ; Through the above rigorous threshold comparison and piecewise function mapping, the multi-objective cooperative algorithm cleanly and efficiently separates the originally chaotic composite gap signal into two sets of independent control command sequences with completely different frequency characteristics.

[0043] After completing the frequency domain decomposition, the system generates a first control command based on the electrochemical intrinsic characteristics of different energy storage hardware to remove the high-frequency disturbance components. The system prioritizes and allocates resources to the supercapacitor bank for rapid response. The supercapacitor bank stores energy using an electrostatic field and does not undergo complex internal electrochemical phase transitions, thus possessing microsecond-level extremely fast charging and discharging capabilities and a virtually unlimited cycle life, perfectly suited for absorbing high-frequency, intense, but relatively small-energy disturbance spikes. Simultaneously, the system generates a second control command to adjust the mid-to-low frequency components. The lithium battery pack is allocated to provide long-term power support. The lithium battery pack achieves energy conversion by the insertion and extraction of lithium ions between the positive and negative electrodes. Although its transient power response speed is not as fast as that of supercapacitors, it has extremely high energy density and can provide stable and long-lasting energy output for steady-state power gaps that change gradually, last for a long time, and require large capacity support.

[0044] The specific implementation steps for charge and discharge compensation of the hybrid energy storage system based on dynamic power gap control described above have extremely significant beneficial effects. Employing a frequency domain decomposition algorithm based on signal change rate, it completely breaks away from the crude scheduling mode of "simultaneous input and output" for hybrid energy storage devices in traditional control strategies, achieving precise "targeted treatment" of complex grid frequency regulation needs to different underlying physical media. By setting a scientific change rate threshold, it forcibly cuts off the impact of high-frequency, high-rate charge and discharge commands on the lithium battery pack, transferring all the most severe operating conditions that easily lead to battery life degradation to the long-life supercapacitor bank for absorption. This highly targeted collaborative current diversion mechanism not only fully leverages the extremely agile advantage of the supercapacitor bank, minimizing the most dangerous frequency glitches in the grid, but also fundamentally protects the valuable lithium battery pack from deep and high-frequency cycling damage, significantly extending the overall life cycle of the hybrid energy storage system, demonstrating significant engineering economic benefits and grid security and stability value.

[0045] In the preferred embodiment, step S3, which involves controlling the hybrid energy storage system to perform charge and discharge compensation, further includes dynamic state-of-charge (SOC) balancing algorithm control. Real-time monitoring of the state of charge of supercapacitor banks and lithium battery banks; Define the target standby state quantity of the supercapacitor bank as: ; Calculate the current state of charge and target reserve quantity of the supercapacitor bank. The difference yields the state of charge deviation. ; When the state of charge deviation is detected When the set safety range is exceeded, the dynamic grouped state of charge equalization control strategy is triggered. Monitor the frequency modulation gap interval, which is the time period when the dynamic power gap is zero; Calculate the available remaining energy of the lithium battery pack within the frequency modulation interval; The remaining energy from the lithium battery pack can be used to recharge the supercapacitor pack, restoring the supercapacitor pack's state of charge to the target reserve level. nearby.

[0046] Step S3, which controls the charge and discharge compensation of the hybrid energy storage system, further introduces a dynamic state-of-charge (SOC) balancing algorithm. This control logic is the internal energy self-sufficiency mechanism that ensures the long-term, high-frequency stable operation of the hybrid energy storage system. The system first continuously and synchronously collects internal state data of the energy storage devices through the underlying battery management system, thereby monitoring the SOC of the supercapacitor bank and the lithium battery bank in real time. The current SOC of the supercapacitor bank is defined as... Define the current state of charge of the lithium battery pack as Real-time acquisition of these two key energy indicators forms the data prerequisite for the entire system to perform internal energy scheduling and closed-loop state management.

[0047] After acquiring the real-time state of charge, the system explicitly defines the target standby state quantity of the supercapacitor bank within the control logic as follows: Target standby state quantity This represents the optimal energy reference level that the supercapacitor bank must maintain to respond to sudden high-frequency power disturbances in the power grid. Subsequently, the system calculates the current state of charge of the supercapacitor bank. With the target reserve state quantity The difference between them yields the state of charge deviation. The logic for calculating this deviation satisfies the mathematical formula: ; In the above formula, This reflects the actual surplus capacity of the supercapacitor bank at the current moment. It is the system's preset ideal energy anchor point, obtained by subtracting the absolute values ​​of the two. The absolute distance between the current state of the supercapacitor bank and the safe reserve range is rigorously quantified. By continuously calculating this deviation, the system can accurately grasp the potential risk of the supercapacitor bank facing energy depletion or overcharging.

[0048] The system control unit continuously transmits the calculated state of charge deviation. The system performs real-time comparisons with preset thresholds. The preset safety range threshold is defined as follows: When a deviation in state of charge is detected Strictly exceeding the set safety range Upon this, the system immediately triggers a dynamic grouped state-of-charge (SOC) equalization control strategy. A deviation exceeding the safe range indicates that the supercapacitor bank has undergone deep unidirectional charging and discharging during the preceding intensive high-frequency modulation operations, and its remaining usable capacity is insufficient to guarantee the complete execution of the next maximum-amplitude high-frequency disturbance command. Triggering the equalization control strategy at this point is a proactive protective and restorative action taken by the system to prevent failure of core high-frequency response devices.

[0049] After triggering the equalization control strategy, the system does not immediately and forcibly take over the power output, but closely monitors the frequency modulation gap interval. The system strictly defines the frequency modulation gap interval as the time period during which the dynamic power gap calculated in the preceding steps is strictly zero. The dynamic power gap is defined as... Then the criterion for determining the frequency modulation gap interval satisfies the mathematical formula: ; In this formula, when A value of zero indicates that the actual active power output of the thermal power unit has fully met the demands of the grid dispatch command, and the hybrid energy storage system is in a transient idle phase where it does not need to output or absorb active power from the grid. Within the accurately captured frequency regulation gap, the system quickly calculates the available remaining energy of the lithium battery pack. The available remaining energy of the lithium battery pack is defined as... This variable represents the ample amount of electricity that the lithium battery pack can freely allocate after meeting its own discharge lower limit protection and reserving basic energy for medium and low frequency regulation.

[0050] After obtaining the available remaining energy parameters, the system controls the bidirectional power conversion device to utilize the available remaining energy of the lithium battery pack. As an energy source, it recharges the supercapacitor bank internally. This reverse energy transfer process can forcibly intervene in the physical charge of the supercapacitor bank, enabling the supercapacitor bank to smoothly and quickly restore its state of charge to the target reserve level. nearby.

[0051] The specific implementation steps of the aforementioned dynamic state-of-charge (SOC) balancing algorithm control have extremely significant and irreplaceable beneficial effects. By introducing a frequency regulation gap interval with zero dynamic power deficit as the time window for triggering balancing actions, the system ingeniously achieves energy self-balancing within the energy storage system without occupying any grid frequency regulation resources or interfering with the active power output response accuracy of thermal power units. This strategy fully utilizes the high energy density of lithium battery packs as a solid backing, continuously "replenishing" the supercapacitor bank during grid transient stability periods. This mechanism completely solves the fatal flaw in traditional hybrid energy storage systems where supercapacitor banks are prone to hitting the upper and lower limits of power supply and shutting down due to frequent high-frequency disturbances. It ensures that the supercapacitor bank is always in the best prepared state to meet the next sudden frequency flicker, thereby greatly improving the continuous guarantee time of full-power backup of the hybrid energy storage system and the long-term reliability of the station's frequency regulation actions.

[0052] In the preferred embodiment, step S4, which involves combining the output of the hybrid energy storage system with the output of the thermal power unit and transmitting it to the grid, includes: A synchronous phasor measurement device (PMU) is deployed on the 6 kV plant service busbar on the side of the thermal power unit. The three-phase AC voltage and current data of the hybrid energy storage system and the thermal power unit are collected in real time through the synchronous phasor measurement device (PMU). The three-phase AC voltage and current data of the hybrid energy storage system are synchronized and aligned with the three-phase AC voltage and current data of the thermal power unit. Inside the synchronous phasor measurement unit (PMU), the aligned voltage and current data are vector superimposed according to Kirchhoff's laws. The total output active power and reactive power are obtained after vector superposition. The total output active and reactive power are encapsulated into standard communication messages and uploaded to the power grid dispatch WAMS system as a joint output result for dispatch assessment.

[0053] Step S4, which involves combining the output of the hybrid energy storage system and the thermal power unit and transmitting it to the grid, is a crucial physical and data exchange link at the end of the power grid to achieve joint thermal-storage response and receive precise grid dispatch. The system first deploys a synchronous phasor measurement unit (PMU) on the 6kV plant auxiliary bus on the thermal power unit side. The PMU possesses extremely high time reference accuracy and transient waveform capture capabilities. Through this device, the system collects real-time three-phase AC voltage and current data from both the hybrid energy storage system and the thermal power unit side. The real-time collected three-phase AC voltage from the hybrid energy storage system side is defined as U_{storage}, and the three-phase AC current as I_{storage}. The real-time collected three-phase AC voltage from the thermal power unit side is defined as U_{unit}, and the three-phase AC current as I_{unit}. Obtaining the underlying raw AC analog waveform data is the physical basis for ensuring that subsequent power synthesis is distortion-free and highly accurate.

[0054] After acquiring the aforementioned underlying analog data, the system timestamps and aligns the three-phase AC voltage and current data from the hybrid energy storage system side with those from the thermal power unit side. Because alternating current has a phase characteristic that changes periodically with time, directly adding data collected at different physical moments algebraically would introduce a significant phase error, causing the synthesized power result to become completely invalid. The system utilizes the high-precision clock source built into the synchronization phasor measurement unit (PMU) to assign an absolute time stamp to each voltage and current sampling point. The time stamp for the hybrid energy storage system side data is defined as follows: Define the time stamp of the thermal power unit side data as The process of timestamp synchronization and alignment involves finding and locking the timestamp with microsecond-level precision. and Completely identical instantaneous sampling points. The synchronization moment for this strict alignment is defined as... After synchronization and alignment, the system extracts two sets of phasor data that are on the same absolute time profile, thus eliminating time-dimension errors for subsequent vector operations.

[0055] After timestamp synchronization is completed, the system performs vector superposition of the aligned voltage and current data within the synchronization phasor measurement unit (PMU) according to Kirchhoff's laws. Kirchhoff's current law states that at any physical node in a circuit, the sum of the current vectors flowing into that node must equal the sum of the current vectors flowing out of that node. The current phasor of the hybrid energy storage system after timestamp alignment is defined as follows: The aligned thermal power unit side current phasor is defined as follows: The total output current phasor after superposition is defined as follows: The logic of vector superposition operations strictly satisfies the following mathematical formula: ; In the above formula, the arrow above the current variable indicates that the variable is a complex vector containing amplitude and phase angle, rather than a simple scalar value. By performing this complex addition operation, the system can accurately reproduce the actual physical current aggregation state of the thermal power unit and the hybrid energy storage system at the intersection node of the 6kV station service bus within the underlying hardware. Simultaneously, the system extracts the reference voltage phasor of the 6kV station service bus at this synchronization moment, defined as... .

[0056] Based on the above vector superposition operation results, the system further calculates the total output active power and reactive power after the vector superposition operation. The calculated total output active power is defined as... The calculated total output reactive power is defined as... Define the reference voltage phasor. With total output current phasor The included angle between them is the power factor angle. The computational logic satisfies the following physical formula: ; ; In the above calculation formula, This indicates the magnitude of the effective value of the reference voltage phasor. This indicates the magnitude of the effective value of the total output current phasor. The active power factor represents the efficiency of active power conversion. Represents the reactive power factor. Constant term. These are the inherent physical constants for calculating three-phase AC power. Through these two rigorous formula models, the system perfectly transforms the underlying voltage and current vector characteristics into the macroscopic active and reactive power indicators that are of utmost concern to the power grid dispatching level.

[0057] After completing the power parameter calculation, the system will output the total active power. and reactive power The data is encapsulated into a standard communication message and uploaded to the power grid dispatching WAMS system as a joint output result for dispatching evaluation. The standard communication message is generated according to the data transmission protocol required by the power grid dispatching center and includes data type identifiers, valid values, checksums, and high-precision timestamps, ensuring data integrity and tamper-proofing during transmission. After being uploaded to the power grid dispatching WAMS system, the dispatching end can use it as the sole and legitimate evaluation criterion for assessing the response speed and regulation accuracy of the thermal power-storage joint regulation point, as well as calculating the compensation benefits for frequency regulation ancillary services across the entire station.

[0058] The specific implementation steps for transmitting the combined output of the hybrid energy storage system and the thermal power unit to the grid have extremely significant beneficial effects. This technical solution completely abandons the outdated synthesis mechanism of reading the data from the thermal power unit system meters and the energy storage system meters separately and then performing simple scalar mathematical addition by the remote terminal. By directly deploying a high-frequency synchronous phasor measurement device (PMU) on the 6kV plant service bus, the system directly sinks the power synthesis link to the underlying electrical hardware layer and introduces strict timestamp alignment and AC phasor vector superposition calculation, eliminating the huge algebraic calculation errors caused by asynchronous sampling and differences in power factor between devices from a physical source. This hardware-level real-time hard synthesis mechanism based on phasors enables the combined output results reported to the grid to have extremely high time resolution and extremely low measurement error rate. This not only fundamentally ensures the absolute fairness and accuracy of the grid dispatching end's evaluation of the comprehensive frequency regulation performance of thermal power units, but also greatly shortens the communication layer latency of data integration and processing within the plant, and greatly enhances the closed-loop response quality of the combined system receiving automatic generation control frequency regulation commands.

[0059] In the preferred embodiment, the method is deployed and used through the following software and modules: The energy management system (EMS) software and energy storage coordination controller module are independently deployed on a Linux operating system edge computing server using Docker containerization technology. Write a Dockerfile to define runtime environment dependencies and package the Energy Management System (EMS) software into a Docker image; Start the Energy Management System (EMS) software container using the container orchestration tool; Establish a hardware communication connection between the Energy Management System (EMS) software and the existing Remote Terminal Units (RTUs) in the thermal power plant, and complete the configuration for forwarding control commands. A multi-objective collaborative algorithm logic is embedded in the energy storage coordinating controller module, which sends Modbus TCP control signals to the battery management system (BMS) to drive hardware actions.

[0060] This method involves practical deployment and use through a specific software architecture and hardware modules. The system first independently deploys the Energy Management System (EMS) software and the Energy Storage Coordination Controller (ESC) module on a Linux-based edge computing server using Docker containerization technology. The edge computing server is deployed close to the data source at the thermal power unit site; this physical proximity provides extremely low network latency and high-bandwidth local data computing capabilities. The EMS serves as the data monitoring and scheduling hub for the entire hybrid energy storage station, responsible for distributing macro-control strategies, recording data, and monitoring the overall station's operational status. Docker containerization technology is a lightweight operating system-level virtualization technology that completely isolates and encapsulates the core code, runtime environment, system tools, and system dependencies required for the EMS software and ESC module to run. Through this independent deployment and environment isolation, the energy storage control system achieves physical and logical decoupling from the original control system of the thermal power plant, avoiding mutual contention for software resources.

[0061] In the specific implementation of containerization, developers write Dockerfiles to strictly define runtime environment dependencies and package the Energy Management System (EMS) software into a Docker image. A Dockerfile is essentially a plain text script containing configuration instructions for the entire software build lifecycle. The system automatically pulls the underlying operating system image from the internal image repository by sequentially parsing and executing the instructions in this script, and then installs the network communication protocol stack, database driver components, and the executable program of the EMS software layer by layer, ultimately generating a standardized Docker image. A Docker image is a highly integrated, statically packaged package that can run anytime, anywhere, containing the complete minimal environment set required for the software to run. Subsequently, the system uses a container orchestration tool to start the EMS software container. The container orchestration tool can automatically deploy, map network ports, mount storage, and monitor the health status of multiple underlying software container instances. When the container orchestration tool detects abnormal blocking or crashes in the EMS software container process, it automatically triggers a process restart mechanism, thereby ensuring the high availability and uninterrupted continuous operation of the entire energy storage monitoring service.

[0062] After completing the underlying containerized deployment of the software system, the system establishes a hardware communication connection between the Energy Management System (EMS) software and the existing Remote Terminal Units (RTUs) in the power plant, and completes the configuration for control command forwarding. The RTU is a standardized physical communication gateway device for bidirectional data interaction between the power plant and the higher-level power grid dispatch center. The system uses industrial Ethernet cables or shielded twisted-pair cables as physical media to directly connect the network port of the edge computing server running the EMS software to the communication expansion board of the RTU. At the network protocol layer, maintenance personnel configure the specific listening port of the EMS software and the data forwarding routing table within the RTU. After this configuration, upon receiving a frequency regulation command data frame from the power grid dispatch center, the RTU can perform lossless and low-latency replication of the command data packet and synchronously mirror it to the local EMS software, thus completely establishing the underlying physical channel for the energy storage system to obtain the original power grid dispatch commands.

[0063] For the underlying device-driven and hardware-coordinated execution phase, the system embeds multi-objective collaborative algorithm logic into the energy storage coordination controller module. This multi-objective collaborative algorithm logic, acting as the controller's brain, is responsible for real-time reading of the dynamic power gap, accurately decomposing it into high-frequency disturbance components and mid-to-low-frequency regulation components in the frequency domain, and calculating in real-time the absolute values ​​of the specific charging and discharging power that different types of energy storage devices should undertake. After the algorithm calculation is completed, the energy storage coordination controller module sends Modbus TCP control signals to the lower-level communication node to the battery management system (BMS) to drive hardware actions. Modbus TCP is a standard Ethernet communication protocol widely used in industrial automation control, characterized by its simple transmission mechanism, strong anti-interference capability, and extremely high real-time data read / write performance. The battery management system (BMS) is the core hardware control hub that directly manages the charging and discharging, thermal status monitoring, and safety protection of the underlying supercapacitor cells and lithium battery cells. The energy storage coordination controller module, through a standard industrial Ethernet interface and according to the register address mapping rules of the Modbus TCP protocol, encapsulates the real-time calculated active power target value into standard write-to-single-register or write-to-multiple-register instruction data frames, and sends them to the lower-level battery management system (BMS) via a switch. After parsing and verifying the received control signals, the Battery Management System (BMS) directly drives the inverter hardware under its jurisdiction to change the switching frequency and duty cycle of the Insulated Gate Bipolar Transistor, thereby precisely changing the output direction and amplitude of the current, and finally completing the physical exchange of energy between the energy storage cell and the grid.

[0064] The aforementioned technical solution for system deployment and implementation using edge computing servers, Docker containerization technology, and standard industrial communication protocols has extremely significant benefits. The novel software architecture combining Linux edge computing servers and Docker containerization technology completely eliminates the drawbacks of traditional power control software's heavy reliance on specific hardware servers and operating system versions, greatly improving the efficiency of seamless cross-platform portability of the control system and the agility of on-site software deployment and upgrades. The independently packaged Docker image mechanism ensures absolute consistency between the Energy Management System (EMS) software in the development and testing environment and the actual production and operation environment. The introduction of container orchestration tools provides the software business layer with automatic disaster recovery and dynamic load balancing capabilities, completely eliminating the risk of a single point of failure causing a station-wide shutdown from the bottom layer of the software architecture. Through direct hardware connection and routing configuration with the Remote Terminal Unit (RTU), high-speed bypass sniffing of grid dispatch commands is achieved without altering the original dispatch communication topology of the thermal power plant, absolutely guaranteeing the information security of the unit's native control network. The energy storage coordination controller module utilizes the standard Modbus TCP protocol to directly and rapidly handshake and issue commands to the underlying Battery Management System (BMS), constructing a millisecond-level hard real-time control closed loop from upper-level complex algorithm decision-making to lower-level high-power physical transformation. This overall deployment scheme enables the core multi-objective cooperative algorithm of this application to directly apply to the physical battery hardware with extremely high logic operation efficiency and extremely low network communication latency, perfectly bridging the gap between software algorithm theory and industrial field hardware control, and fully meeting the stringent engineering requirements of millisecond-level response speed for primary frequency regulation and automatic generation control frequency regulation in new power systems.

[0065] In the preferred embodiment, the actual deployment and use steps of the hybrid energy storage system also include the deployment of the primary system electrical architecture: The supercapacitor bank and lithium battery bank are connected to the 6 kV plant bus of the thermal power unit through the power conversion system PCS; Multiple physical switching modules are deployed between the 6kV plant service bus and the power conversion system PCS; By switching the physical switching switch module between closed and open states using logic control commands, the hybrid energy storage system can flexibly switch between states of full-station energy storage shutdown, energy storage unidirectionally assisting the frequency regulation of Unit 1, energy storage unidirectionally assisting the frequency regulation of Unit 2, and full-station energy storage jointly assisting the frequency regulation of the target unit.

[0066] The actual deployment steps of the hybrid energy storage system further encompass the deep integration and deployment of the primary system's electrical architecture. The system connects the supercapacitor bank and lithium battery bank to the 6kV plant service bus of the thermal power unit via a power conversion system (PCS). The PCS is defined as the core hardware hub connecting the AC and DC power grids. It possesses a bidirectional full-bridge converter topology, capable of inverting the DC power generated within the supercapacitor bank and lithium battery bank into 6kV AC power conforming to grid phase and frequency standards, or rectifying the 6kV AC power into DC power to charge the energy storage medium. The 6kV plant service bus of the thermal power unit is a core medium-voltage distribution node within the thermal power plant. Selecting this 6kV voltage level node for grid connection minimizes Joule heat loss during high-power charging and discharging processes and ensures the hybrid energy storage system and the generator unit are in the closest physical coupling state in terms of electrical distance. This provides a solid and low-impedance physical hardware foundation for achieving millisecond-level frequency regulation power support.

[0067] After establishing the basic grid connection nodes, the system deploys multiple physical switching modules between the 6kV plant service bus and the power conversion system (PCS). The set of physical switching modules is defined as K_{matrix}, which contains several independent high-voltage vacuum circuit breakers or isolating switches. Each physical switching module has two defined electrical states: closed and open. The closed state represents complete conduction of the primary electrical circuit, allowing large currents to flow bidirectionally without obstruction; the open state represents physical disconnection of the primary electrical circuit, forming a clear safety insulation break in space. This is achieved by deploying the physical switching modules in a matrix configuration within the primary electrical circuit. The system breaks through the static physical limitations of traditional energy storage power stations and single grid connection points, and gives the entire hybrid energy storage system the flexible adjustment capability of dynamically reconfiguring the physical connection architecture without changing the basic routing of hardware cables.

[0068] The system issues logic control commands through a central controller to precisely switch the closed and open states of the aforementioned multiple physical switching modules. The logic control command set is defined as follows: The system executes a set of logical control instructions based on the overall command requirements of the power grid dispatch and the real-time operating conditions of all generating units in the plant. The system enables flexible switching between four core operating states for the hybrid energy storage system. The first state is the full-site energy storage shutdown state, where the system controls all physical switching modules to enter the detached state, completely severing all electrical connections between the power conversion system (PCS) and the 6kV plant bus. This state is specifically suitable for offline conditions such as deep maintenance or full-site shutdown for maintenance. The second state is the energy storage unidirectionally assisting Unit 1 in frequency regulation. In this state, the system commands only to close the physical switching modules corresponding to the branch of Unit 1, ensuring that the hybrid energy storage resources of the entire station uniquely inject or absorb active power into or from the bus where Unit 1 is located. The third state is the energy storage unidirectionally assisting Unit 2 in frequency regulation, with its hardware execution logic being completely symmetrical with that of Unit 1, meaning all energy storage frequency regulation resources are tilted and locked to Unit 2. The fourth state is the full-site energy storage jointly assisting the target unit in frequency regulation. The real-time target frequency regulation power requirement of Unit 1 is defined as... Define the real-time target frequency regulation power requirement of Unit 2 as follows: Define the current maximum adjustable total power of the hybrid energy storage system as: In this combined assisted state, the system must satisfy the dynamic power distribution balance equation: ; Under the strict constraint of the aforementioned mathematical balance, the system controls multiple physical switching modules to be simultaneously in specific closed states, enabling the power conversion system (PCS) to have an electrical path for simultaneous energy throughput to the buses of both generator units. Based on the real-time energy shortage ratio and urgency of response of the two units, the system achieves joint dynamic support of overall hybrid energy storage resources across units.

[0069] The aforementioned deployment of the primary system electrical architecture and the implementation steps for flexible multi-state switching have yielded extremely significant benefits. This flexible topology electrical architecture based on a physical switching module matrix completely solves the problems of resource idleness and initial investment waste caused by the traditional single energy storage system being passively and permanently bound to a single generator unit. Dynamic reconfiguration of the underlying primary electrical architecture through logic control commands enables multiple thermal power units throughout the plant to dynamically share the frequency regulation capabilities of this expensive yet efficient hybrid energy storage system, greatly improving the overall utilization rate of the core energy storage asset throughout its entire lifecycle. Simultaneously, this physical deployment method endows the thermal power plant with extremely high operational fault tolerance and on-site maintenance flexibility. When any generator unit is shut down for maintenance, the system can instantly and seamlessly switch idle energy storage resources to an adjacent, operating generator unit, ensuring that the plant's acquisition of grid frequency regulation ancillary service benefits is not physically interrupted. The dual physical isolation design for the entire station's energy storage withdrawal state fundamentally ensures the safety of on-site equipment maintenance personnel during live maintenance, providing a perfect engineering solution for the safe, efficient, and intelligent implementation of large-scale hybrid energy storage systems in traditional thermal power plants, and fully meeting the stringent requirements of modern new power systems for high reliability and high flexibility of frequency regulation resources.

[0070] Example 2 Further explanation in conjunction with Example 1, such as Figure 1-5 As shown, a hybrid energy storage-assisted frequency regulation system for thermal power units includes: The data acquisition module is used to acquire power grid frequency regulation demand commands and real-time operating data of thermal power units; The power gap calculation module is used to calculate the dynamic power gap based on the frequency regulation demand command and the real-time operating data of the thermal power unit. The energy storage charge and discharge control module is used to control the charge and discharge compensation of the hybrid energy storage system based on the dynamic power gap. The hybrid energy storage system includes a supercapacitor pack and a lithium battery pack. The output synthesis and transmission module is used to synthesize the output of the hybrid energy storage system and the output of the thermal power unit and transmit them to the grid side.

[0071] A hybrid energy storage-assisted frequency regulation system for thermal power units serves as the hardware and logic carrier for executing the aforementioned core frequency regulation control strategy. Its system architecture is rigorously divided into four highly collaborative functional modules to ensure the closed-loop, stable, and efficient operation of the entire frequency regulation process. The system first includes a data acquisition module, whose core function is to acquire grid frequency regulation demand commands and real-time operating data of the thermal power units. As the sensory organ of the entire frequency regulation system, the data acquisition module integrates the underlying communication driver interface for interfacing with remote terminal units and wide-area measurement systems. This module is responsible not only for lossless reception of raw data streams from massive network packets but also for parsing and cleaning key operating status parameters. In actual operation, the data acquisition module can capture and output grid frequency parameters in real time. Thermal power unit terminal power Automatic power generation control target power value and the current power output of the thermal power unit Core benchmark data. The specific implementation of the data acquisition module not only breaks down the data barriers between the power grid dispatching layer and the plant execution layer, but also, through hardware-level high-speed sampling and multi-channel isolation design, greatly filters out noise interference caused by external communication networks, providing downstream modules with an absolutely pure underlying data source with extremely high time resolution, thereby effectively avoiding erroneous frequency regulation actions caused by initial data distortion.

[0072] Closely cascaded with the data acquisition module within the system is the power gap calculation module. This module calculates the dynamic power gap based on frequency regulation demand commands and real-time operating data of thermal power units. The power gap calculation module is the mathematical computation center of the entire system, internally containing dual-channel difference calculation logic for primary frequency regulation and automatic generation control frequency regulation. When facing primary frequency regulation conditions, the module derives the theoretical target value of primary frequency regulation based on fundamental variables such as thermal power unit capacity parameters, and performs a high-frequency difference calculation with the unit's terminal power to output the dynamic power gap for primary frequency regulation. When facing automatic generation control frequency regulation conditions, the module directly extracts the real-time difference between the target power value and the current power value. Regardless of the command condition, the power gap calculation module ultimately outputs a unified and precisely quantified dynamic power gap variable. This highly integrated computing architecture has extremely outstanding benefits. It perfectly and dynamically translates the inherent regulation lag phenomenon caused by the mechanical and thermodynamic inertia of thermal power units into an electrical power compensation target that the hybrid energy storage system can directly identify. It completely decouples the complex logical entanglement between the unit's native control system and the external auxiliary system, laying a solid algorithmic foundation for achieving high-precision new joint frequency regulation.

[0073] After determining the dynamic compensation target, the system transfers instructions to the energy storage charge-discharge control module. This module is used to control the charge-discharge compensation of the hybrid energy storage system based on the dynamic power gap. The hybrid energy storage system is physically defined as including a supercapacitor bank and a lithium battery bank. The energy storage charge-discharge control module is the system's brain and execution decision-making center, internally running highly sophisticated multi-objective collaborative algorithms and dynamic state-of-charge balancing algorithms. This module calculates the dynamic power gap variable in real time. The signal change rate forcibly tears apart the complex compensation needs of the power grid in the frequency domain, precisely distributing the high-frequency disturbance components caused by high-frequency abrupt changes to the supercapacitor bank with microsecond-level response physical characteristics, while simultaneously distributing the medium- and low-frequency regulation components that require massive energy to support them to the high-energy-density lithium battery bank. During the intervals between frequency regulation operations, the module also actively monitors the state-of-charge deviation. Dispatching the remaining energy of lithium battery packs It enables internal reverse energy recovery for supercapacitor banks. The energy storage charge and discharge control module deeply integrates algorithm scheduling with the physical intrinsic characteristics of the underlying electrochemical medium. Its beneficial effect is that it fundamentally eliminates the permanent lifespan damage to the internal lattice structure of lithium battery banks caused by harsh high-frequency and high-current charge and discharge conditions. At the same time, it gives supercapacitor banks the ability to cope with grid frequency spikes an unlimited number of times, maximizing the comprehensive return on investment of tens of millions of energy storage assets throughout their entire life cycle.

[0074] The final closed-loop component of the system is the output synthesis and transmission module, which synthesizes the output of the hybrid energy storage system and the thermal power unit and transmits it to the grid. Deployed at the very end of the plant's electrical data aggregation, this module highly integrates underlying logic based on synchronous phasor measurement. It enforces extremely strict timestamp synchronization and alignment of the three-phase AC voltage and current data from the hybrid energy storage side and the thermal power unit side within a very short time slice. Furthermore, it performs transient vector superposition calculations according to Kirchhoff's laws in the underlying computational core, ultimately deriving the total active power output representing the true physical state of the entire plant. and total output reactive power The establishment of this module completely overturns the traditional remote control system's crude synthesis method, which relies on asynchronous polling of multiple meters and simple scalar summation. Its greatest benefit lies in completely eliminating, from the physical level, the huge algebraic synthesis errors caused by spatial sampling asynchrony and differences in the power factor between generators and converters. The highly accurate and low-latency joint output messages provide the power grid dispatch center with a panoramic and realistic picture of the plant and substation response, ensuring that thermal power plants can obtain the highest level of regulation performance index scores and full economic compensation when participating in the power ancillary services market settlement, and further enhancing the wide-area stability control capability of the entire regional power grid in an environment of high interaction between power generation, grid, load, and storage.

[0075] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A method for frequency regulation of a hybrid energy storage-assisted thermal power unit, characterized by the following steps: include: S1. Obtain grid frequency regulation demand commands and real-time operating data of thermal power units; S2. Calculate the dynamic power gap based on the frequency regulation demand command and the real-time operating data of the thermal power unit; S3. Charge and discharge compensation is performed on a hybrid energy storage system based on dynamic power gap control. The hybrid energy storage system includes a supercapacitor pack and a lithium battery pack. S4. The output of the combined energy storage system and the output of the thermal power unit are fed into the grid.

2. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 1, characterized in that: Step S1, which involves obtaining the grid frequency regulation demand command and the real-time operating data of the thermal power unit, includes: Establish communication connections with the power grid dispatching WAMS system and remote control units (RTUs); Receive scheduling message data packets through the communication connection; Parse the header identifier bits of the scheduling message data packet and extract the instruction type feature code; By comparing the instruction type feature code with the preset feature library, the power grid frequency regulation demand instruction can be identified as a primary frequency regulation demand instruction or an AGC frequency regulation demand instruction. When a frequency regulation demand command is identified, the grid frequency parameters and the power output of the thermal power unit are collected. When the AGC frequency regulation demand command is identified, the target command is received through the remote control unit (RTU) and the actual output of the thermal power unit is collected.

3. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 2, characterized in that: Step S2 The steps for calculating the dynamic power deficit in response to a primary frequency regulation demand command include: Calculate the theoretical target value for primary frequency regulation based on grid frequency parameters and thermal power unit capacity parameters; The difference between the theoretical target value of the primary frequency regulation and the terminal power of the thermal power unit is calculated. The result of the difference operation is used as the dynamic power gap for primary frequency modulation.

4. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 2, characterized in that: Step S2, which calculates the dynamic power gap for the AGC frequency modulation demand command, includes: A combined frequency regulation strategy of storage and generator is adopted; The steps of the joint frequency regulation strategy for energy storage and power generation are to keep the original DCS control logic and operation mode of the thermal power unit unchanged. The AGC command and real-time operating data of the thermal power unit are simultaneously forwarded to the energy storage control subsystem. The target power value is extracted from the AGC command through the energy storage control subsystem. ; Synchronously acquire the current power value of the actual output of thermal power units ; Calculate the target power value Compared with the current power value Real-time difference between ; Real-time difference As the dynamic power gap of AGC.

5. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 1, characterized in that: Step S3, which involves charging and discharging compensation of the hybrid energy storage system based on dynamic power gap control, includes: A multi-objective cooperative algorithm is used to decompose the dynamic power gap in the frequency domain to obtain the high-frequency disturbance component and the mid-to-low frequency regulation component; The specific implementation steps of the multi-objective cooperative algorithm are as follows: extract the rate of change of the dynamic power gap signal and set a threshold for the rate of change; The power demand component with a rate of change greater than the rate of change threshold is classified as a high-frequency disturbance component. The power demand component with a rate of change less than or equal to the rate of change threshold is divided into low- and medium-frequency regulation components. The first control command is generated to allocate the high-frequency disturbance component to the supercapacitor bank for a rapid and priority response. The second control command is generated to allocate the low- and medium-frequency regulation components to the lithium battery pack to provide long-term power support.

6. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 5, characterized in that: Step S3, which controls the hybrid energy storage system to perform charge and discharge compensation, also includes dynamic state-of-charge (SOC) balancing algorithm control: Real-time monitoring of the state of charge of supercapacitor banks and lithium battery banks; Define the target standby state quantity of the supercapacitor bank as: ; Calculate the current state of charge and target reserve quantity of the supercapacitor bank. The difference yields the state of charge deviation. ; When the state of charge deviation is detected When the set safety range is exceeded, the dynamic grouped state of charge equalization control strategy is triggered. Monitor the frequency modulation gap interval, which is the time period when the dynamic power gap is zero; Calculate the available remaining energy of the lithium battery pack within the frequency modulation interval; The remaining energy from the lithium battery pack can be used to recharge the supercapacitor pack, restoring the supercapacitor pack's state of charge to the target reserve level. nearby.

7. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 1, characterized in that: Step S4, which involves combining the output of the hybrid energy storage system with the output of the thermal power unit and transmitting it to the grid, includes: A synchronous phasor measurement device (PMU) is deployed on the 6 kV plant service busbar on the side of the thermal power unit. The three-phase AC voltage and current data of the hybrid energy storage system and the thermal power unit are collected in real time through the synchronous phasor measurement device (PMU). The three-phase AC voltage and current data of the hybrid energy storage system are synchronized and aligned with the three-phase AC voltage and current data of the thermal power unit. Inside the synchronous phasor measurement unit (PMU), the aligned voltage and current data are vector superimposed according to Kirchhoff's laws. The total output active power and reactive power are obtained after vector superposition. The total output active and reactive power are encapsulated into standard communication messages and uploaded to the power grid dispatch WAMS system as a joint output result for dispatch assessment.

8. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 1, characterized in that: The method is deployed and used in practice through the following software and modules: The energy management system (EMS) software and energy storage coordination controller module are independently deployed on a Linux operating system edge computing server using Docker containerization technology. Write a Dockerfile to define runtime environment dependencies and package the Energy Management System (EMS) software into a Docker image; Start the Energy Management System (EMS) software container using the container orchestration tool; Establish a hardware communication connection between the Energy Management System (EMS) software and the existing Remote Terminal Units (RTUs) in the thermal power plant, and complete the configuration for forwarding control commands. A multi-objective collaborative algorithm logic is embedded in the energy storage coordinating controller module, which sends Modbus TCP control signals to the battery management system (BMS) to drive hardware actions.

9. The frequency regulation method for hybrid energy storage-assisted thermal power units according to claim 8, characterized in that: The actual deployment and use of hybrid energy storage systems also includes the deployment of the primary system electrical architecture: The supercapacitor bank and lithium battery bank are connected to the 6 kV plant bus of the thermal power unit through the power conversion system PCS; Multiple physical switching modules are deployed between the 6kV plant service bus and the power conversion system PCS; By switching the physical switching switch module between closed and open states using logic control commands, the hybrid energy storage system can flexibly switch between states of full-station energy storage shutdown, energy storage unidirectionally assisting the frequency regulation of Unit 1, energy storage unidirectionally assisting the frequency regulation of Unit 2, and full-station energy storage jointly assisting the frequency regulation of the target unit.

10. A hybrid energy storage-assisted frequency regulation system for thermal power units, characterized in that, The system is used to execute the hybrid energy storage-assisted frequency regulation method for thermal power units according to any one of claims 1 to 9, the system comprising: The data acquisition module is used to acquire power grid frequency regulation demand commands and real-time operating data of thermal power units; The power gap calculation module is used to calculate the dynamic power gap based on the frequency regulation demand command and the real-time operating data of the thermal power unit. The energy storage charge and discharge control module is used to control the charge and discharge compensation of the hybrid energy storage system based on the dynamic power gap. The hybrid energy storage system includes a supercapacitor pack and a lithium battery pack. The output synthesis and transmission module is used to synthesize the output of the hybrid energy storage system and the output of the thermal power unit and transmit them to the grid side.

Images