Coordinated operation optimization method and system of coal-fired unit and hybrid energy storage system

Abstract

This application provides a method and system for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system. The method includes: generating an optimal parameter combination for performing variational mode decomposition (MODE) using a particle swarm optimization algorithm under preset variational mode decomposition conditions; receiving grid control commands, performing MODE based on the optimal parameter combination to obtain initial command components, and allocating them to the generator unit and the hybrid energy storage system respectively; simulating the generator unit and the hybrid energy storage system using a dynamic simulation model and calculating the dynamic tracking error of the generator unit; generating compensation commands based on the dynamic tracking error and allocating them to the hybrid energy storage system; if the simulation is considered complete, using the simulation results as the fitness value of the particle swarm optimization algorithm to generate a new optimal parameter combination for the next iteration, repeating the iteration until the results converge to obtain the final optimal operating parameters for controlling the generator unit and the hybrid energy storage system. Finally, reliable operating parameters are output.

Description

Technical Field

[0001] This invention relates to the field of energy system operation control technology, and in particular to a method and system for optimizing the coordinated operation of coal-fired power units and hybrid energy storage systems. Background Technology

[0002] In related technologies, to improve the self-regulation capability of coal-fired power units, they are coupled with energy storage devices with different response characteristics, such as supercapacitors, lithium batteries, and vanadium redox flow batteries, to form a hybrid energy storage system, thereby enhancing the overall system flexibility. However, this multi-device coupling introduces new control challenges: on the one hand, it is difficult to efficiently distribute grid frequency regulation commands between coal-fired power units and various energy storage systems; on the other hand, the actual operating conditions of coal-fired power units are complex and variable, making it difficult to achieve effective coordination with various energy storage systems. Summary of the Invention

[0003] The technical problem to be solved by this invention is: how to provide an effective control strategy to achieve adaptation to complex working conditions.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system includes: Under the preset variational mode decomposition conditions, the particle swarm optimization algorithm generates a set of optimal parameter combinations for performing the variational mode decomposition. Receive grid control commands, perform variational mode decomposition according to the optimal parameter combination to obtain initial command components, and allocate them to generator sets and hybrid energy storage systems respectively; The generator set and hybrid energy storage system were simulated using a dynamic simulation model, and the dynamic tracking error of the generator set was calculated. Compensation commands are generated based on the dynamic tracking error and distributed to the hybrid energy storage system; If the simulation is deemed complete, the simulation result is used as the fitness value of the particle swarm optimization algorithm to generate a new optimal parameter combination for the next iteration. This process is repeated until the result converges to obtain the final optimal combination of operating parameters for controlling the generator set and the hybrid energy storage system.

[0005] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows: A collaborative operation optimization system for a coal-fired power unit and a hybrid energy storage system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the collaborative operation optimization method for a coal-fired power unit and a hybrid energy storage system as described above.

[0006] The beneficial effects of this invention are as follows: by optimizing the key parameters of variational mode decomposition using the particle swarm optimization algorithm, an optimal initial power allocation strategy is formulated; then, the initial command components generated by the grid control command and the optimal parameter combination are simulated using a dynamic simulation model. That is, the dynamic simulation model serves as a test environment for evaluating the performance of the output strategy of the particle swarm optimization algorithm, and can accurately simulate complex physical processes such as generator delay and energy storage dynamics; at the same time, the dynamic tracking error of the generator set is calculated during the simulation process, and compensation commands are generated and allocated to the hybrid energy storage system. The simulation results are used as the fitness value of the particle swarm optimization algorithm to further iterate in the next round, repeating the iteration until the results converge, thereby finally outputting a reliable combination of operating parameters that meets the optimization objective and has been verified by a high-precision environment. Attached Figure Description

[0007] Figure 1 This is a flowchart illustrating an optimization method for the coordinated operation of a coal-fired power unit and a hybrid energy storage system, according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the optimized method for the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a collaborative operation optimization system for a coal-fired power unit and a hybrid energy storage system according to an embodiment of the present invention. Detailed Implementation

[0008] Definitions:

[0009] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0010] In existing technologies, coal-fired power units need to undertake faster and more frequent frequency regulation tasks to smooth out random fluctuations in wind and solar power output. To improve the regulation capability of coal-fired power units themselves, a hybrid energy storage system is usually constructed to enhance the overall system flexibility. However, this multi-device coupling introduces new control challenges: on the one hand, there is a lack of efficient and reliable strategy generation methods, making it difficult to achieve dynamic and adaptive optimal allocation of grid frequency regulation commands between coal-fired power units and various energy storage systems; on the other hand, the actual operating conditions of coal-fired power units are complex and variable, while existing methods mostly rely on fixed rules or offline optimization based on highly simplified coal-fired power unit models, making it difficult to cope with the complex and variable actual operating conditions and unable to fully utilize the real-time status of energy storage devices.

[0011] To address at least some of the aforementioned issues, this invention constructs a two-layer closed-loop optimization architecture of "decision-simulation-evaluation." The upper-layer particle swarm optimization algorithm optimizes strategy parameters in real time, generating initial command components based on grid control commands. This drives the lower-layer dynamic simulation model for verification and evaluation, calculating dynamic tracking errors and generating compensation commands. This approach establishes the optimization process of the operating strategy on a simulation environment close to engineering applications. In this way, adaptive and precise tuning of strategy parameters can be achieved, effectively improving the overall frequency regulation performance and economy of the system.

[0012] The following details a method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system, using a two-layer operation optimization strategy architecture applied to a coal-fired power unit and a hybrid energy storage system as an example. This architecture includes: The coal-fired power unit includes a boiler, a steam turbine, a control subsystem, and a generator; a hybrid energy storage unit is coupled to the coal-fired power unit and includes a supercapacitor with a response speed on the order of seconds, a lithium battery suitable for adjustment on the order of minutes, and a vanadium redox flow battery suitable for scheduling on the order of hours or more; a cooperative control system is connected to the control subsystem, the generator, and the hybrid energy storage unit of the coal-fired power unit; the cooperative control system is configured to execute the method of this embodiment, please refer to... Figure 1 as well as Figure 2 The method 100 includes steps 101 to 105: Step 101: Using a particle swarm optimization algorithm, under a preset variational mode decomposition condition, generate a set of optimal parameter combinations for performing the variational mode decomposition. For example, given a preset total number of VMD modes... K Under the condition that, a random initialization is performed within its feasible solution space by... N A population of particles; the position vector of each particle. X i This represents a set of candidate VMD runtime strategy parameter combinations, specifically defined as a quadratic penalty factor. α With three frequency division orders k c1 , k c2 , k c3 ,Right now X i =[ α i , k c1 , k c2 , k c3 Simultaneously, a flight velocity vector is randomly initialized for each particle. V i .

[0013] Step 102: Receive grid control commands, perform variational mode decomposition based on the optimal parameter combination to obtain initial command components, and allocate them to the generator set and hybrid energy storage system respectively; for example, receive a real-time AGC command sequence from the grid. P AGC ( t and parameter combinations X i Using the parameters issued α The instruction is adaptively decomposed using the VMD algorithm based on the frequency division order. After decomposition, four initial instruction components are obtained and then allocated to the supercapacitor, lithium battery, vanadium redox flow battery, and coal-fired unit, respectively.

[0014] Step 103: Simulate the generator set and hybrid energy storage system using a dynamic simulation model, and calculate the dynamic tracking error of the generator set. The dynamic simulation model performs a closed-loop system simulation including the coal-fired generator set and the hybrid energy storage unit to obtain the actual dynamic response of the system and calculate the dynamic tracking error of the coal-fired generator set; for example, calculating the dynamic delay error generated when the coal-fired generator set responds to its fundamental frequency and low-frequency initial command components yields the dynamic tracking error. The model receives the aforementioned initial allocation command and performs a closed-loop dynamic simulation for a preset duration. In the industrial-grade high-precision simulation platform, the coal-fired generator set is constructed as a complex multi-physics dynamic system including a boiler, turbine, generator, and control loop. This surpasses the simple mathematical approximation of a first-order inertial transfer function with pure time delay, accurately simulating the complete physical process from command issuance to actual mechanical power output in the time domain, and calculating the dynamic tracking error. e ( t Meanwhile, the lithium batteries, supercapacitors, and vanadium redox flow batteries in the hybrid energy storage unit are all modeled using equivalent circuit models or electrochemical models that consider internal resistance, charge / discharge efficiency, and capacity decay characteristics, in order to accurately simulate their power response limits, real-time evolution of state of charge, and aging effects. During simulation, the industrial-grade high-precision simulation platform model solves the problem in fixed millisecond steps to ensure that high-frequency dynamic components in the grid AGC commands and unit responses are captured.

[0015] Step 104: Generate compensation instructions based on the dynamic tracking error and allocate them to the hybrid energy storage system; for example, based on the frequency characteristics of the dynamic tracking error, allocate them to the supercapacitor and lithium battery to share the burden.

[0016] Step 105: If the simulation is deemed complete, the simulation results are used as the fitness value of the particle swarm optimization algorithm to generate a new optimal parameter combination for the next iteration. This iteration is repeated until the results converge to obtain the final optimal operating parameter combination for controlling the generator set and hybrid energy storage system. For example, after the simulation ends, relevant data such as the system's frequency regulation performance indicators, energy storage equipment state of charge (SOC) exceedance data, and economic calculation results are fed back to the upper-level structure. The frequency regulation performance indicators include frequency regulation rate, frequency regulation accuracy, and response time. The economic calculation is based on the levelized energy storage cost model. Subsequently, the particle swarm optimization algorithm is driven to perform the next iteration. That is, the simulation constitutes a strict closed loop: the allocation strategy issued by the upper level determines the reference power command for each device; this command is affected by the dynamic characteristics of the device's physical constraints such as the unit ramp rate and energy storage power limit, as well as the unit's inertia, in the simulation model, resulting in actual output and SOC changes; the "actual" unit output and SOC state output by the simulation model are fed back to the controller inside the model in real time to calculate the tracking error. e ( t It then executes subsequent compensation and state management logic to form a high-precision operating environment.

[0017] As described above, the beneficial effects of this invention are as follows: by optimizing the key parameters of variational mode decomposition using the particle swarm optimization algorithm, an optimal initial power allocation strategy is formulated; then, the initial command components generated by the grid control command and the optimal parameter combination are simulated using a dynamic simulation model. That is, the dynamic simulation model serves as a test environment for evaluating the performance of the output strategy of the particle swarm optimization algorithm, and can accurately simulate complex physical processes such as generator delay and energy storage dynamics; at the same time, the dynamic tracking error of the generator set is calculated during the simulation process, and compensation commands are generated and allocated to the hybrid energy storage system. The simulation results are used as the fitness value of the particle swarm optimization algorithm to further iterate in the next round, repeating the iteration until the results converge, thereby finally outputting reliable operating parameters that meet the optimization objectives and have been verified by a high-precision environment.

[0018] In one embodiment of the present invention, step 102 includes: allocating the base frequency component to the generator set, and allocating the high frequency component, medium frequency component and low frequency component to the target energy storage device in the hybrid energy storage system, respectively.

[0019] As can be seen from the above description, by generating components of different frequencies, it is possible to adapt to different devices and achieve effective control of different devices.

[0020] In one embodiment of the present invention, step 102, which involves performing variational mode decomposition based on the optimal parameter combination to obtain initial command components, includes: The variational mode decomposition is performed according to the quadratic penalty factor and the three frequency division orders to obtain a preset number of modal components; the preset number of modal components are arranged in order of high to low frequency and divided according to the three frequency division orders to obtain the high frequency component, the mid frequency component, the low frequency component and the fundamental frequency component.

[0021] As described above, after obtaining a preset number of modal components, sorting the modal components can quickly distinguish the frequencies of different modal components, thereby quickly generating high-frequency components, mid-frequency components, low-frequency components, and fundamental frequency components.

[0022] In one embodiment of the present invention, variational mode decomposition is performed according to the optimal parameter combination to obtain initial instruction components; wherein, the core of VMD is to solve a constrained variational problem, the goal of which is to find K eigenmode functions u k ( t The goal is to find a solution that sums all modes to the original signal and minimizes the estimated bandwidth of each mode. This problem is solved by introducing a quadratic penalty factor. α and Lagrange multipliers λ ( t The augmented Lagrangian function is constructed as follows for solution:

[0023] in, u k ( t ) represents the intrinsic mode functions to be solved; K is the total modulus; P AGC ( t ) represents the power grid control command; α is the quadratic penalty factor; λ(t) is the Lagrange multiplier; after solving, K modal components u are obtained arranged from low frequency to high frequency. 1(t) ,…,u k(t) Based on parameter k c1 k c2 k c3 , will the first k c1 The superposition of multiple modes serves as the base command P for coal-fired power units. G,base(t) , kth c1+1 To k c2 The superposition of multiple modes serves as the lithium battery reference instruction P. Li,base(t) , kth c2+1 To k c3 The superposition of multiple modes serves as the reference command P for the all-vanadium redox flow battery. VFB,base(t) The superposition of remaining modes serves as the supercapacitor reference command P. SC,base(t) j is the imaginary unit, w k Let be the center frequency of the k-th modal component. Let be the partial derivative operator with respect to time t.

[0024] As described above, in this embodiment, the constraints that must be precisely satisfied in the original equation are removed by introducing a quadratic penalty factor, turning them into soft constraints that allow deviations but are penalized. The Lagrange multipliers continuously accumulate information about the reconstruction error and use the accumulated amount to modify the objective function of the next optimization, forcing the sum of modes to approach the original signal. That is, the soft constraints are calibrated back to hard constraints using the dual rising rule. The accuracy of the overall calculation is guaranteed while reducing the computation time.

[0025] In one embodiment of the present invention, the calculation of the dynamic tracking error of the generator set in step 103 includes: acquiring the generator output data and charge change status data output by the simulation model; and calculating the dynamic tracking error based on the generator output data and charge change status data.

[0026] As described above, by calculating the dynamic tracking error using unit output data and charge change status data, the actual error of unit output can be accurately reflected.

[0027] In one embodiment of the present invention, step 104, generating a compensation instruction based on the dynamic tracking error and allocating it to the hybrid energy storage system, includes: performing a secondary decomposition on the dynamic tracking error to obtain a secondary frequency component; obtaining a compensation instruction based on the secondary frequency component; and sending the compensation instruction to the target energy storage device in the hybrid energy storage system; the compensation instruction includes a charge compensation amount. For example, the unit dynamic tracking error... e ( t According to preset fixed rules, such as a low-pass filter with a fixed cutoff frequency, the mixture is decomposed into two parts and then distributed to the supercapacitor and lithium battery to share the compensation.

[0028] As described above, by performing a second decomposition of the dynamic tracking error to obtain the second frequency component, a compensation command is generated based on the second frequency component to instruct the supercapacitor and lithium battery to bear the corresponding amount of charge, thereby achieving effective allocation of excess charge.

[0029] In one embodiment of the present invention, the energy storage system includes a capacitor device and a battery device; wherein the capacitor device includes a supercapacitor, and the battery device includes a lithium battery and a vanadium redox flow battery; sending the compensation command to the target energy storage device in the hybrid energy storage system includes: first sending the compensation command to the capacitor device; if the capacity of the capacitor device is insufficient, the remaining charge compensation is carried out by the battery device. That is, the capacitor device first undertakes the task, and then the battery device undertakes the task.

[0030] As described above, in this embodiment, the SOC state of the supercapacitor, which responds faster, is prioritized. For any remaining issues, the SOC state of the lithium battery is assessed, and so on. The supercapacitor, lithium battery, and vanadium redox flow battery respond sequentially, thereby enabling the rapid handling of any excess charge.

[0031] In one embodiment of the present invention, the method further includes: if the state of charge (SOC) of any target energy storage device in the hybrid energy storage system is detected to be lower than its corresponding target safety threshold, then the amount of charge exceeding the safety threshold in the target energy storage device is allocated to the remaining energy storage devices. That is, while continuously monitoring the state of charge of each energy storage device, if the SOC of any energy storage device is lower than the safety threshold, then the portion of the target command it is responsible for is dynamically transferred to the remaining energy storage devices; during the transfer, the transfer is preferentially made to the energy storage devices with faster response times based on the response speed of the energy storage devices.

[0032] As described above, when any energy storage device falls below the safety threshold, the charge is carried by other energy storage devices in turn, which can quickly carry the excess charge.

[0033] In one embodiment of the present invention, step 105, using the simulation results as the fitness value of the particle swarm optimization algorithm, includes: obtaining the average value of the frequency modulation comprehensive performance index and the state-of-charge (SOC) exceedance statistics of the hybrid energy storage system; and calculating the fitness value by weighting the average value and the SOC exceedance statistics, specifically: After completing a full simulation cycle, the lower-level module analyzes the simulation results and calculates a set of quantitative evaluation indicators, mainly including: the overall frequency modulation performance index K of the system. p The average value, statistical data on the state of charge exceeding the limit for each energy storage, and preliminary economic calculation results based on the levelized cost model, where K p The calculation formula is as follows: ; K1, K2, and K3 represent the frequency modulation rate, frequency modulation accuracy, and response time, respectively.

[0034] The upper-level module calculates a fitness value (Xi) by weighting the above indicators, which is used to evaluate the set of parameters X. i The fitness value indicates the overall performance of the operating strategy corresponding to that set of parameters.

[0035] After receiving the fitness values ​​of all particles, the upper-level policy optimization module drives the particle swarm optimization algorithm to update and iterate. First, it updates the individual historical best position P of each particle. best,i and the global optimal position G of the entire population bestSubsequently, based on the velocity and position update formulas for the particle swarm, the position of each particle in the next generation is calculated. The velocity and position update formulas for particle i in the d-th dimension are as follows: ; ; Where w is the inertia weight, c1 and c2 are learning factors, and r1 and r2 are random numbers within [0,1]. Using this formula, the particle adjusts its flight direction and step size based on individual and global optimalities, exploring towards a better solution space.

[0036] The updated particle swarm optimization (PSO) parameters, i.e., the new sets of VMD parameters, will be sent down to the next layer again, repeating the above steps to form a closed loop of "parameter decision-simulation evaluation-function feedback-re-decision". This process iterates until the preset conditions of maximum iteration count or fitness value convergence are met. Finally, the algorithm outputs the globally optimal position, which is a set of VMD operation strategy parameters that have been repeatedly verified by high-precision simulations and have the best overall performance. α i,opt , k c1,opt , k c2,opt , k c3,opt ].

[0037] Please refer to Figure 3 A collaborative operation optimization system for a coal-fired power unit and a hybrid energy storage system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the collaborative operation optimization method for a coal-fired power unit and a hybrid energy storage system as described above.

[0038] In summary, this invention provides a method and system for optimizing the coordinated operation of coal-fired power units and hybrid energy storage systems. The method employs a two-layer optimization architecture to address the problem that fixed operating strategies are ill-suited to complex operating conditions. The upper layer uses a particle swarm optimization algorithm to optimize the quadratic penalty factor and frequency segmentation order of variational mode decomposition. The lower layer uses a high-precision co-simulation model to replace the traditional simplified model for closed-loop dynamic simulation, serving as a test environment to evaluate the performance of the upper-layer strategy and accurately simulating complex physical processes such as unit delays and energy storage dynamics. Based on grid AGC commands, the method decomposes and distributes the commands to supercapacitors, lithium batteries, vanadium redox flow batteries, and coal-fired power units using optimized parameters. The simulation accurately calculates the actual system response and unit tracking error, and distributes these errors to the hybrid energy storage units based on frequency characteristics. During operation, the load is dynamically transferred according to the energy storage status. Finally, the simulation performance indicators are fed back to the upper-layer driving parameter update, forming a "decision-simulation-evaluation" closed loop, ultimately outputting reliable operating parameters that meet the optimization objectives and have been verified by a high-precision environment. This approach fundamentally overcomes the problem of optimization results deviating from engineering practice due to insufficient model accuracy in traditional methods. It achieves significant improvement in the overall frequency modulation performance, operational economy, and engineering reliability of the strategy through precise optimization of software strategies without changing the hardware configuration.

[0039] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system, characterized in that, include: Under the preset variational mode decomposition conditions, the particle swarm optimization algorithm generates a set of optimal parameter combinations for performing the variational mode decomposition. Receive grid control commands, perform variational mode decomposition according to the optimal parameter combination to obtain initial command components, and allocate them to generator sets and hybrid energy storage systems respectively; The generator set and hybrid energy storage system were simulated using a dynamic simulation model, and the dynamic tracking error of the generator set was calculated. Compensation commands are generated based on the dynamic tracking error and distributed to the hybrid energy storage system; If the simulation is deemed complete, the simulation result is used as the fitness value of the particle swarm optimization algorithm to generate a new optimal parameter combination for the next iteration. This process is repeated until the result converges to obtain the final optimal combination of operating parameters for controlling the generator set and the hybrid energy storage system.

2. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 1, characterized in that, The initial command components include high-frequency components, intermediate-frequency components, low-frequency components, and fundamental-frequency components; The base frequency component is allocated to the generator set, and the high frequency component, medium frequency component, and low frequency component are respectively allocated to the target energy storage device in the hybrid energy storage system.

3. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 2, characterized in that, The optimal parameter combination includes a quadratic penalty factor and three frequency segmentation orders; Performing the variational mode decomposition based on the optimal parameter combination yields the following initial command components: The variational mode decomposition is performed based on the quadratic penalty factor and the three frequency division orders to obtain a preset number of mode components. The preset number of modal components are arranged in descending order of frequency, and then divided according to the three frequency division orders to obtain the high-frequency component, the intermediate-frequency component, the low-frequency component, and the fundamental frequency component.

4. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 3, characterized in that, Performing the variational mode decomposition based on the optimal parameter combination yields the following initial command components: in, u k ( t ) represents the intrinsic mode functions to be solved; K is the total modulus; P AGC ( t ) represents the power grid control command; α represents the quadratic penalty factor; λ(t) represents the Lagrange multiplier; j represents the imaginary unit; w k Let be the center frequency of the k-th modal component. Let be the partial derivative operator with respect to time t.

5. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 1, characterized in that, The calculation of the dynamic tracking error of the generator set includes: Obtain the unit output data and charge change status data output by the simulation model; The dynamic tracking error is calculated based on the unit output data and charge change status data.

6. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 1, characterized in that, The step of generating compensation commands based on the dynamic tracking error and allocating them to the hybrid energy storage system includes: The dynamic tracking error is decomposed into secondary frequency components. The compensation command is obtained based on the secondary frequency component and sent to the target energy storage device in the hybrid energy storage system; the compensation command includes a charge compensation amount.

7. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 6, characterized in that, The energy storage system includes capacitor devices and battery devices; The step of sending the compensation command to the target energy storage device in the hybrid energy storage system includes: The compensation command is sent to the capacitor device first. If the capacity of the capacitor device is insufficient, the remaining charge compensation is carried by the battery device.

8. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 6, characterized in that, Also includes: If the charge state of any of the target energy storage devices in the hybrid energy storage system is detected to be lower than its corresponding target safety threshold, the amount of charge exceeding the safety threshold in the target energy storage device will be allocated to the remaining energy storage devices.

9. The method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system according to claim 1, characterized in that, The step of using the simulation results as the fitness value of the particle swarm optimization algorithm includes: Obtain the average value of the frequency regulation comprehensive performance index and the state of charge exceeding the limit statistics of the hybrid energy storage system; The fitness value is calculated by weighting the average value and the out-of-limit statistics of the state of charge.

10. A collaborative operation optimization system for a coal-fired power unit and a hybrid energy storage system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements each step of the method for optimizing the coordinated operation of a coal-fired power unit and a hybrid energy storage system as described in any one of claims 1 to 9.

Images