A hybrid energy storage frequency-voltage cooperative adaptive control method

Abstract

The application discloses a kind of hybrid energy storage frequency-voltage collaborative adaptive control methods, including collecting real-time operating state parameters;Establish multi-dimensional safety constraint model, quantitatively solve to obtain the active callable margin of each energy storage unit, reactive power support margin and maximum sustained support duration;Build multi-objective optimization model;Respectively calculate flywheel, electrochemical energy storage, molten salt energy storage each unit safety margin, further obtain hybrid energy storage system overall safety margin index;Solve optimization model output each energy storage unit active, reactive power reference instruction, issue hybrid energy storage converter executes, realize hybrid energy storage system to the collaborative adaptive support control of power grid frequency-voltage.It fully considers the differentiated safety boundary of heterogeneous energy storage, dynamically adjusts control priority according to power grid disturbance level and energy storage itself remaining safety margin, considers power grid stability support demand and energy storage equipment operation safety, and is suitable for high proportion new energy access power grid scene.

Description

Technical Field

[0001] This invention relates to the field of power system energy storage control technology, and in particular to a hybrid energy storage frequency-voltage coordinated adaptive control method. Background Technology

[0002] With the large-scale grid connection of high-proportion renewable energy sources, the grid inertia level continues to decrease, making it prone to problems such as frequency drops and voltage fluctuations under fault disturbances. Hybrid energy storage can provide frequency and voltage support for grids with a high proportion of renewable energy. Flywheel energy storage has a fast response speed and is suitable for short-term high-power support, but its speed window is limited, requiring restrictions on the rate of speed change to avoid mechanical shock; electrochemical energy storage has a relatively fast power response, but is constrained by multiple conditions such as SOC, cell temperature, and aging status; molten salt energy storage has long-term output capability, but it has thermal inertia, and its output cannot change rapidly and drastically.

[0003] Most existing hybrid energy storage collaborative control systems adopt a fixed weight allocation mode, which does not quantify the real-time safety margin of each energy storage unit, nor does it have an adaptive weight adjustment mechanism that combines the grid disturbance level with the energy storage's own safety boundary. Some control strategies only consider the hard power limit and ignore internal state constraints such as flywheel mechanical impact constraints, molten salt thermal inertia delay constraints, and electrochemical energy storage temperature constraints. This can easily lead to control commands exceeding the equipment's allowable operating range, causing equipment protection lockout, premature withdrawal of support, and deterioration of the stability control effect under grid disturbances. Summary of the Invention

[0004] The purpose of this invention is to provide a hybrid energy storage frequency-voltage coordinated adaptive control method based on multi-type energy storage state constraints, which solves the technical problems of fixed weights in traditional strategies and insufficient consideration of the differentiated internal safety constraints of multi-type energy storage, while taking into account both the grid disturbance support requirements and the operational safety of multiple types of energy storage devices.

[0005] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0006] S1. Real-time acquisition of power grid operation data and operation status data of multiple types of hybrid energy storage systems; the multiple types of hybrid energy storage systems include at least flywheel energy storage units, electrochemical energy storage units, and molten salt energy storage units.

[0007] The system initializes by reading inherent parameters of the equipment, including the PCS rated apparent power and safety thresholds of each energy storage unit; during the control cycle, it collects grid frequency, voltage, and frequency change rate; it collects flywheel real-time speed and output power; electrochemical energy storage SOC, output power, power change rate, and cell temperature; and molten salt energy storage remaining thermal capacity and actual output power.

[0008] S2. Based on the collected operating status data, establish safety constraint models for flywheel energy storage, electrochemical energy storage, and molten salt energy storage respectively. Based on each safety constraint model, quantify the active power availability margin of each energy storage unit. No margin of power available and maximum duration of continuous support .

[0009] For flywheel energy storage, speed constraints, instantaneous power constraints, and mechanical shock constraints are established:

[0010] Steady-state constraints on rotational speed: In the formula, This refers to the real-time rotational speed of the flywheel. , These are the minimum and maximum safe speed settings for the flywheel, respectively.

[0011] Instantaneous power constraints: In the formula, This refers to the real-time active power of the flywheel. , The upper and lower limits of the flywheel's charging and discharging power.

[0012] Mechanical impact constraint: In the formula, The rotational speed of the previous control cycle; This represents the maximum permissible speed change per single cycle.

[0013] The safety constraint model for electrochemical energy storage includes SOC range constraints, power ramping constraints, and temperature constraints.

[0014] SOC interval constraints: In the formula, This refers to the real-time state of charge of the battery. , This is the threshold value for the battery's safe charging range.

[0015] Power ramping constraints: In the formula, The rate of change of battery power; This is the battery's inherent maximum ramp rate.

[0016] Cell temperature constraints: In the formula, Real-time temperature of the battery cell; , This refers to the safe operating temperature range for the battery.

[0017] The safety constraint model for molten salt energy storage includes constraints on remaining thermal storage capacity and long-term continuous power output.

[0018] Remaining thermal storage capacity constraints: In the formula, This represents the real-time remaining thermal storage capacity. , These represent the upper and lower limits of thermal storage capacity.

[0019] Steady-state and dynamic ramping constraints: In the formula, To provide real-time power output for molten salt; To achieve maximum steady-state output; This refers to the real-time output rate of the molten salt. This is the maximum permissible rate of molten salt output.

[0020] All grid-connected energy storage power is limited by the apparent power of the PCS and must meet the following constraints: In the formula, , The real-time active and reactive power of the i-th type of energy storage; The rated apparent power of the PCS.

[0021] Based on the quantification of each safety constraint model, the active power availability margin, reactive power support margin, and maximum continuous support duration of each energy storage unit are obtained, specifically as follows:

[0022] Available margin for use: ;

[0023] No reactive power margin:

[0024] ;

[0025] Maximum duration of continuous support: In the formula, This represents the remaining usable energy that can be released to the outside after deducting the safety lower limit from the i-th type of energy storage.

[0026] S3. Establish a target optimization function F that integrates grid frequency and voltage recovery factors, energy storage loss suppression factors, and energy storage safety margin factors. The target optimization function includes frequency and voltage recovery factor F1, loss suppression factor F2, and safety margin factor F3. The safety constraints, active power availability margin, reactive power support margin, and maximum continuous support duration of each energy storage unit obtained in step S2 are used as constraints of the optimization model.

[0027] Frequency and voltage recovery factor F1: With the goal of suppressing grid frequency and voltage deviations, it enables the grid to recover quickly under disturbances; In the formula: This refers to the system frequency deviation. The rate of change of frequency; This refers to the voltage deviation at the grid connection point. The frequency change rate weighting coefficient; The start and end times of the disturbance support.

[0028] Loss suppression factor F2: Combines flywheel mechanical loss, electrochemical energy storage aging loss, and molten salt energy storage heat loss to reduce the overall operating loss of hybrid energy storage; In the formula: These are the weighting coefficients for flywheel, electrochemical, and molten salt energy storage losses, respectively. For the real-time mechanical wear of the flywheel; This is due to electrochemical aging loss; This is for the real-time heat dissipation loss of molten salt.

[0029] Safety margin factor F3: ensures that the operating point of each energy storage unit is far from the constraint boundary, and retains sufficient safety control margin. In the formula: Reference commands for active and reactive power in energy storage; This represents the minimum support time required for grid disturbances; the smaller this sub-target value, the higher the safety margin for energy storage operation.

[0030] Construct the objective function with an out-of-bounds penalty function: ,satisfy ; As an out-of-bounds penalty term, the penalty function increases in value when the solution exceeds the energy storage safety constraint boundary, forcing the optimization result to fall within the allowed operating domain of the equipment.

[0031] S4. Calculate the severity index of power grid disturbances based on power grid operation data. Calculate the safety margin of the flywheel energy storage unit respectively. Safety margin of electrochemical energy storage units Safety margin of molten salt energy storage unit Then, the overall safety margin index of hybrid energy storage is obtained by solving the problem. ;

[0032] The severity index of power grid disturbances is calculated using a normalized weighted average:

[0033] , , , For power grid frequency deviation, The rate of change of frequency, For voltage deviation, This is the maximum allowable offset threshold for the power grid.

[0034] Safety margin of flywheel energy storage unit The flywheel speed constraint, power constraint, and mechanical shock constraint are combined, normalized, and the minimum value is taken.

[0035]

[0036] Safety margin of electrochemical energy storage units Integrating SOC constraints, power ramp-up constraints, and temperature constraints

[0037] Perform normalization and take the minimum value;

[0038]

[0039] Safety margin of molten salt energy storage unit The remaining thermal storage capacity constraint and the output ramp-up constraint are combined and normalized, and the minimum value is taken.

[0040]

[0041] The overall safety margin of hybrid energy storage is taken as the minimum value of the safety margins of the three types of energy storage units:

[0042]

[0043] The safety margin of each unit ranges from [0,1]; the closer the value is to 1, the more sufficient the energy storage safety margin is.

[0044] A value of 0 indicates that the energy storage unit is approaching the safety constraint boundary.

[0045] S5. Based on the aforementioned power grid disturbance severity index and overall safety margin index of hybrid energy storage The weight coefficients corresponding to frequency voltage recovery, loss suppression, and safety margin factors in the objective optimization function are adaptively adjusted. , , Furthermore, a smoothing filter is applied to the adjusted weight coefficients.

[0046] Weighted adaptive adjustment logic: The higher the severity of power grid disturbance, the greater the weight of frequency and voltage recovery factors. Prioritize ensuring grid stability; the lower the overall safety margin of hybrid energy storage, the greater the weight of safety margin factors. Prioritize the protection of energy storage equipment; increase the weight of loss suppression factors under steady-state grid operating conditions. This reduces energy storage operation losses. Specifically:

[0047]

[0048]

[0049]

[0050] In the formula These are the upper and lower thresholds for each weight. Weight normalization is then performed.

[0051]

[0052]

[0053]

[0054] To avoid power command oscillations caused by weight jumps, a first-order low-pass filter is used to smooth the weights:

[0055] In the formula: The original weights for the current period; These are the effective weights after filtering; These are the filter weights from the previous cycle; This is the filtering smoothing coefficient, ranging from 0 to 1, which effectively suppresses weight jumps and ensures smooth and stable control throughout the entire process.

[0056] S6. Filter the weight coefficients. Substituting the objective optimization function F, the active power reference commands corresponding to the flywheel energy storage unit, electrochemical energy storage unit, and molten salt energy storage unit are obtained by solving the problem. Reactive power reference command The power reference command is sent to the hybrid energy storage converter for execution, completing the grid frequency-voltage coordinated adaptive support control; return to step S1 to enter the next control cycle and realize closed-loop control.

[0057] The beneficial effects of this invention are:

[0058] 1. This invention establishes multi-dimensional safety constraint models that include internal states for flywheel, electrochemical, and molten salt energy storage respectively. It not only considers the power limitation of PCS hardware, but also introduces flywheel mechanical impact constraints, electrochemical energy storage temperature constraints, and molten salt thermal inertia constraints, which are more in line with the actual operating boundaries of the equipment.

[0059] 2. The safety margin of each energy storage unit is obtained by normalization and quantification. The overall system margin is determined by the unit with the most stressful state, realizing a unified quantitative evaluation of the safety status of multiple types of energy storage and providing calculable indicators for weighted adaptive operation.

[0060] 3. Simultaneously, a dual-dimensional driving weight adaptive approach is introduced, using indicators of grid disturbance severity and overall safety margin of hybrid energy storage. Priority is given to ensuring grid stability in the event of large disturbances; priority is given to ensuring equipment safety when energy storage capacity is insufficient; and priority is given to reducing losses under steady-state conditions, thus solving the problem of poor adaptability of traditional fixed weights.

[0061] 4. Implement low-pass smoothing filtering on the weights to avoid power command oscillations caused by sudden weight changes and improve the stability of hybrid energy storage control. Attached Figure Description

[0062] Figure 1This is a flowchart of a hybrid energy storage frequency-voltage coordinated adaptive control method according to the present invention;

[0063] Figure 2 This is a simulation diagram of a hybrid energy storage frequency-voltage coordinated adaptive control method according to the present invention. Detailed Implementation

[0064] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0065] like Figure 1 As shown: This embodiment constructs a hybrid energy storage power supply system that includes flywheel energy storage, electrochemical energy storage and molten salt energy storage. The system has a rated frequency of 50Hz, a rated voltage of 1.0pu, and an allowable voltage operating range of 0.9 to 1.1pu. The flywheel energy storage system is configured with upper and lower limits for operating speed, instantaneous power constraints, and mechanical shock constraints. The minimum safe speed is 94.2 rad / s, the maximum safe speed is 314.2 rad / s, the initial speed is 219.9 rad / s, the maximum active power is 0.8 pu, the PCS apparent power is 1.2 pu, and the moment of inertia is 100 kg·m². The electrochemical energy storage system is configured with SOC operating range constraints (0.1 to 0.9), power ramp constraints of 0.2 pu / s, and operating temperature constraints of -10℃ to 55℃. The initial SOC is 0.8, the maximum active power is 0.8 pu, and the PCS apparent power is 1.2 pu. The molten salt energy storage system is configured with remaining thermal storage capacity constraints (0.05 to 1.0 pu), long-term continuous output constraints, and thermal inertia delay constraints (time constant 1.0 s). The initial thermal storage capacity is 0.9 pu, the maximum active power is 0.5 pu, and the PCS apparent power is 0.8 pu. The power grid inertia time constant H is set to 10s, the damping coefficient D is set to 3.0, and the control period is 0.01s.

[0066] To verify the effectiveness of the proposed method, the following typical load disturbance condition was set: the system operates in steady state from 0 to 10 seconds, with an active power of 1.0 pu and a reactive power of 0.3 pu; at 10 seconds, a load surge occurs, with the active power jumping to 1.3 pu (an increase of 30%) and the reactive power jumping to 0.4 pu. This disturbance lasts until 80 seconds; after 80 seconds, the load recovers to 1.0 pu and the reactive power recovers to 0.3 pu. This condition simulates a common load surge scenario in a real power grid, used to test the dynamic response characteristics and adaptive adjustment capabilities of hybrid energy storage in terms of frequency-voltage coordinated support.

[0067] During each control cycle, the system collects real-time data on grid operation and the operating status parameters of the three types of energy storage. On the grid side, it collects the voltage amplitude at the grid connection point, system frequency, and rate of frequency change; on the flywheel side, it collects the rotor speed and output power; on the electrochemical side, it collects the state of charge (SOC), charge / discharge power, and cell temperature; and on the molten salt side, it collects the remaining thermal storage capacity and actual output. Based on the collected data, the available capacity margin and maximum continuous support duration of each unit are quantified. Initially, the flywheel's active power available capacity margin is 0.8 pu, its reactive power available capacity margin is approximately 0.9 pu, and its maximum continuous support duration is approximately 180 s; the electrochemical energy storage's active power available capacity margin is 0.8 pu, its reactive power available capacity margin is approximately 0.9 pu, and its maximum continuous support duration is approximately 2880 s; the molten salt energy storage's active power available capacity margin is 0.5 pu, its reactive power available capacity margin is approximately 0.62 pu, and its maximum continuous support duration is approximately 1800 s. These margin values ​​provide a quantitative basis for subsequent multi-objective optimization.

[0068] This embodiment constructs an objective optimization function that includes three factors: frequency and voltage recovery, loss suppression, and safety margin. The frequency and voltage recovery factor aims to suppress grid frequency and voltage deviations, achieved by integrating the squares of the frequency deviation, the rate of change of frequency, and the voltage deviation over the disturbance period. The loss suppression factor integrates flywheel mechanical losses, electrochemical energy storage aging losses, and molten salt energy storage heat losses to reduce the overall operating losses of the hybrid energy storage system. The safety margin factor ensures that the operating points of each energy storage unit are far from the constraint boundaries, maintaining sufficient safety control margin. The objective function is a weighted sum of the three factors plus an out-of-bounds penalty term. The weight coefficients satisfy the normalization condition, and the out-of-bounds penalty term increases sharply when the solution exceeds the energy storage safety constraint boundary, forcing the optimization result to fall within the allowable operating domain of the equipment.

[0069] At the moment of the 10-second load surge, the system frequency began to drop, with a frequency change rate of approximately -0.8 Hz / s, and the voltage dropped to approximately 0.98 pu. The severity index S of the power grid disturbance was calculated using the method proposed in this invention. dis With a frequency deviation weight of 0.4, a frequency change rate weight of 0.3, and a voltage deviation weight of 0.3, and with maximum frequency deviation thresholds of 0.2Hz, maximum frequency change rate thresholds of 2.0Hz / s, and maximum voltage deviation thresholds of 0.05pu, S is calculated. dis The value was 1.24, and after being truncated to the upper limit, it was taken as 1.0, indicating that the power grid was under severe disturbance at this time. Meanwhile, the safety margins of each energy storage unit were calculated: the flywheel safety margin, taking into account both speed and power constraints, was taken as the minimum value of 0.8; the electrochemical safety margin, taking into account both SOC and temperature constraints, was taken as the minimum value of 0.85; the molten salt safety margin, calculated based on the remaining thermal storage capacity, was 0.89; and the overall safety margin of the hybrid energy storage was taken as the minimum of the three values, 0.8, indicating that the energy storage system has a relatively sufficient safety margin.

[0070] Based on the severity of the disturbance Sdis =1.0 and overall safety margin S mar =0.8. The original weights are calculated according to the adaptive adjustment formula proposed in this invention. After normalization, w1, w2, and w3 are 0.80, 0.08, and 0.12, respectively. After first-order low-pass filtering and smoothing, the final effective weights are w1=0.42, w2=0.19, and w3=0.39. From Figure 2 As can be seen, after a sudden load surge of 10 seconds, the weight w1 of the frequency-voltage recovery sub-target rapidly increases, reflecting the control logic of "prioritizing grid stability support when the disturbance is greater." As the electrochemical energy storage continues to discharge, the SOC gradually decreases, and the overall safety margin gradually declines. The weight adaptive mechanism then gradually increases the safety margin sub-target weight w3, reflecting the adaptive adjustment function of "prioritizing equipment safety when energy storage capacity is insufficient." After the load recovers at 80 seconds, w2 rebounds, and the control strategy shifts to reducing energy storage operating losses.

[0071] The filtered weighting coefficients are substituted into the objective optimization function for solution. In this embodiment, the core function `fmincon` from the MATLAB optimization toolbox, used for finding the minimum value of constrained nonlinear multivariate functions, is used to solve for various active and reactive power reference commands for energy storage. In the initial 10s disturbance phase, the optimization yields approximately 0.35 pu for flywheel active power, approximately 0.45 pu for electrochemical active power, approximately 0.29 pu for molten salt active power, and 0.01 pu, 0.02 pu, and 0.01 pu for reactive power, respectively. After the solution is completed, the active and reactive power reference commands are sent to each energy storage converter for execution, achieving frequency-voltage coordinated support control. During the simulation operation of the hybrid energy storage system, the active power output of the flywheel, electrochemical active power, and molten salt is as follows: Figure 2 As shown.

[0072] like Figure 2 As shown, in terms of power balance, the total active power output of energy storage quickly tracks load changes after a disturbance. After about 30 seconds, the energy storage output stabilizes at about 1.29 pu, and the difference between it and the load (1.3 pu) is reduced to less than 0.01 pu, achieving rapid power balance. After the load recovers after 80 seconds, the total energy storage output smoothly decreases and tracks load changes.

[0073] This embodiment fully verifies the effectiveness of the proposed control method under a typical load surge condition. The multidimensional safety constraint model accurately characterizes the differentiated physical characteristics such as flywheel speed limits, electrochemical SOC and temperature limits, molten salt thermal inertia, and thermal storage limits, and all energy storage state parameters do not exceed the limits. The dual-dimensional driven weighted adaptive design of disturbance severity index and safety margin index realizes the intelligent decision of "prioritizing grid protection for large disturbances, prioritizing equipment protection for low margins, and ensuring stable operation to reduce losses." Weighted smoothing filtering effectively suppresses sudden power command changes, and the output of each energy storage unit is smooth and oscillating. The method proposed in this invention can effectively solve the frequency-voltage coordination support problem of hybrid energy storage in scenarios with a high proportion of new energy access, and has engineering application value.

[0074] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A hybrid energy storage frequency-voltage cooperative adaptive control method, characterized in that, Includes the following steps: S1: Real-time acquisition of power grid operation data and operation status data of multiple types of hybrid energy storage systems; S2: Based on the collected operating status data, establish safety constraint models for the flywheel energy storage unit, electrochemical energy storage unit, and molten salt energy storage unit respectively. Based on each safety constraint model, quantify the active power availability margin, reactive power support margin, and maximum continuous support duration of each energy storage unit. S3: Establish an objective optimization function that integrates grid frequency and voltage recovery factors, energy storage loss suppression factors, and energy storage safety margin factors, and use the safety constraint model, active power availability margin, reactive power support margin, and maximum continuous support duration of each energy storage unit as constraints of the optimization function. S4: Calculate the severity index of power grid disturbances based on power grid operation data. Based on various energy storage safety constraints, the safety margins of flywheel energy storage units, electrochemical energy storage units, and molten salt energy storage units are calculated respectively. Then, the overall safety margin index of hybrid energy storage is obtained by solving. ; S5: Based on the severity index of power grid disturbance and overall safety margin index of hybrid energy storage The weight coefficients of the grid frequency and voltage recovery factor, energy storage loss suppression factor, and energy storage safety margin factor in the objective optimization function are adaptively adjusted, and the adjusted weight coefficients are subjected to smoothing filtering. S6: Substitute the filtered weighting coefficients into the target optimization function to solve for the active power reference command and reactive power reference command of each energy storage unit. Send the power reference command to the hybrid energy storage converter (PCS) for execution to complete the grid frequency-voltage coordinated adaptive support control.

2. The hybrid energy storage frequency-voltage coordinated adaptive control method according to claim 1, characterized in that, The multi-type hybrid energy storage system mentioned in step S1 includes at least a flywheel energy storage unit, an electrochemical energy storage unit, and a molten salt energy storage unit; the grid operation data and the multi-type hybrid energy storage system operation status data include: voltage amplitude at the grid connection point, system frequency, and frequency change rate RoCoF; rotor speed, instantaneous output power, and mechanical operating status of the flywheel energy storage; SOC, charge and discharge power, and cell temperature of the electrochemical energy storage; remaining thermal storage capacity, actual output, and thermal state of the molten salt energy storage; factory hardware parameters of each energy storage unit, inherent maximum charge and discharge ramp rate, and PCS apparent power limit.

3. The hybrid energy storage frequency-voltage coordinated adaptive control method according to claim 1, characterized in that, The safety constraint model for flywheel energy storage in step S2 includes speed constraints, instantaneous power constraints, and mechanical shock constraints. Steady-state constraints on rotational speed: In the formula, This refers to the real-time rotational speed of the flywheel. , These are the minimum and maximum safe speed settings for the flywheel, respectively. Instantaneous power constraints: In the formula, This refers to the real-time active power of the flywheel. , The upper and lower limits of flywheel charging and discharging power; Mechanical impact constraint: In the formula, The rotational speed of the previous control cycle; This represents the maximum permissible speed change per single cycle. The safety constraint model for electrochemical energy storage includes SOC range constraints, power ramping constraints, and temperature constraints. SOC interval constraints: In the formula, This refers to the real-time state of charge of the battery. , This refers to the threshold value for the battery's safe charge range. Power ramping constraints: In the formula, The rate of change of battery power; This is the battery's inherent maximum ramp rate; Cell temperature constraints: In the formula, Real-time temperature of the battery cell; , This refers to the safe operating temperature range for the battery. The safety constraint model for molten salt energy storage includes constraints on remaining thermal storage capacity and long-term continuous power output. Remaining thermal storage capacity constraints: In the formula, This represents the real-time remaining thermal storage capacity. , These are the upper and lower limits of thermal storage capacity; Steady-state and dynamic ramping constraints: In the formula, To provide real-time power output for molten salt; To achieve maximum steady-state output; This refers to the real-time output rate of the molten salt; This is the maximum permissible rate of molten salt output; All grid-connected energy storage power is limited by the apparent power of the PCS and must meet the following constraints: In the formula, , The real-time active and reactive power of the i-th type of energy storage; The rated apparent power of the PCS.

4. The hybrid energy storage frequency-voltage coordinated adaptive control method according to claim 3, characterized in that, In step S2, the active power availability margin, reactive power support margin, and maximum continuous support duration of each energy storage unit are quantified based on each security constraint model, specifically as follows: Available margin for use: ; No reactive power margin: ; Maximum duration of continuous support: In the formula, This represents the remaining usable energy that can be released to the outside after deducting the safety lower limit from the i-th type of energy storage.

5. The hybrid energy storage frequency-voltage coordinated adaptive control method according to claim 1, characterized in that, The frequency-voltage recovery factor F1, loss suppression factor F2, and safety margin factor F3 fused in the objective optimization function in step S3 are specifically as follows: In the formula: This refers to the system frequency deviation. The rate of change of frequency; This refers to the voltage deviation at the grid connection point. The frequency change rate weighting coefficient; The start and end times of the disturbance support; In the formula: These are the weighting coefficients for flywheel, electrochemical, and molten salt energy storage losses, respectively. For the real-time mechanical wear of the flywheel; This is due to electrochemical aging loss; This is to account for the real-time heat dissipation loss of molten salt; In the formula: Reference commands for active and reactive power in energy storage; This represents the minimum support time required for grid disturbances; the smaller this sub-target value, the higher the safety margin for energy storage operation.

6. The hybrid energy storage frequency-voltage coordinated adaptive control method according to claim 5, characterized in that, The objective optimization function in step S3, which integrates grid frequency and voltage recovery factors, energy storage loss suppression factors, and energy storage safety margin factors, is as follows: satisfy ; This is an out-of-bounds penalty term; the penalty function increases in value when the solution exceeds the energy storage safety constraint boundary.

7. The hybrid energy storage frequency-voltage cooperative adaptive control method according to claim 1, characterized in that, Safety margin of flywheel energy storage unit in step S4 The flywheel speed constraint, power constraint, and mechanical shock constraint are combined, normalized, and the minimum value is taken. Safety margin of electrochemical energy storage units The system combines SOC constraints, power ramp-up constraints, and temperature constraints, performs normalization, and takes the minimum value. Safety margin of molten salt energy storage units The remaining thermal storage capacity constraint and the output ramp-up constraint are combined and normalized, and the minimum value is taken. The overall safety margin of hybrid energy storage is taken as the minimum value of the safety margins of the three types of energy storage units: The safety margin of each unit is in the range of [0,1]; the closer the value is to 1, the more sufficient the safety margin of energy storage is; the value equal to 0 means that the energy storage unit has approached the safety constraint boundary.

8. The hybrid energy storage frequency-voltage cooperative adaptive control method according to claim 7, characterized in that, Indicators of power grid disturbance severity in step S4: , , , For power grid frequency deviation, The rate of change of frequency, For voltage deviation, This is the maximum allowable offset threshold for the power grid.

9. The hybrid energy storage frequency-voltage cooperative adaptive control method according to claim 1, characterized in that, The adaptive adjustment logic in step S5 is as follows: the higher the severity of the power grid disturbance, the greater the weight of the frequency and voltage recovery factor. The lower the overall safety margin of hybrid energy storage, the higher the weight of factors that increase the safety margin. ; Increasing the weight of loss suppression factors under steady-state power grid operating conditions Specifically: In the formula These are the upper and lower thresholds for each weight; Weight normalization processing: The smoothing filtering process uses a first-order low-pass filter: In the formula: The original weights for the current period; These are the effective weights after filtering; These are the filter weights from the previous cycle; This is the filtering smoothing coefficient, with a value ranging from 0 to 1. It effectively suppresses weight jumps and ensures smooth and stable control throughout the entire process.

Images