A double-layer control system for energy storage and thermal power combined primary frequency modulation
By optimizing the power adjustment and coefficient allocation layers of the dual-layer control system, combined with the improved whale algorithm, the joint frequency regulation of energy storage and thermal power is optimized, solving the problems of energy storage battery life loss and poor frequency regulation effect, and realizing the reduction of energy storage battery life and the improvement of frequency regulation effect under different operating conditions.
Patent Information
- Application Number
- CN202211465197.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-11-22
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2042-11-22
Smart Images

Figure CN115776123B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a primary frequency regulation control method, and more particularly to a two-layer control system for primary frequency regulation combining energy storage and thermal power. Background Technology
[0002] In recent years, the grid frequency fluctuations caused by the integration of a large number of renewable energy sources have made it impossible for thermal power units to meet the frequency regulation requirements when participating in primary frequency regulation alone. At the same time, the heavy frequency regulation requirements have also increased the losses of thermal power units. In order to improve the frequency regulation effect of thermal power units and reduce their losses, some scholars have proposed using a frequency regulation method that combines energy storage with thermal power units.
[0003] Currently, there are numerous studies on frequency regulation using energy storage to assist thermal power units. Two models are used for primary frequency regulation: one does not restrict the state of charge, while the other considers state of charge (SOC) recovery, improving the effectiveness of primary frequency regulation. A profit-oriented primary frequency regulation control method is employed, aiming for optimal regulation performance and coordinating the participation of multiple energy storage systems. Setting the SOC of the energy storage system within an optimal range minimizes the cost of its participation in frequency regulation. Developing energy storage charging and discharging strategies improves the frequency regulation performance of conventional thermal power units. A comprehensive control method is used, employing different control strategies based on system frequency regulation requirements to enhance the regulation effect. A proportional gain coefficient is used to control the allocation of two different control strategies, avoiding disturbances caused by switching between the two strategies. Predictive methods are used to evaluate the frequency regulation benefits of energy storage batteries participating in primary frequency regulation. Some scholars have constrained the SOC of energy storage participating in frequency regulation, but have not considered energy storage lifespan losses.
[0004] In summary, most current research employs single-layer control, which neglects energy storage lifetime when considering frequency regulation performance. Therefore, how to combine energy storage power and allocation coefficient to balance frequency regulation performance with energy storage lifetime has become a technical problem that needs to be solved. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of the prior art by providing a dual-layer control system that combines energy storage and thermal power for primary frequency regulation, thereby improving the primary frequency regulation effect and reducing energy storage lifespan loss.
[0006] The objective of this invention can be achieved through the following technical solutions:
[0007] According to one aspect of the present invention, a two-layer control system for primary frequency regulation of energy storage and thermal power is provided. The system includes a power adjustment layer and a coefficient allocation layer, wherein the power adjustment layer adjusts the unit regulation power of energy storage based on ultra-short-term load forecast data using an improved whale algorithm with the goal of minimizing energy storage lifetime loss, and transmits the unit regulation power to the coefficient allocation layer.
[0008] The coefficient allocation layer uses an improved whale algorithm with the goal of minimizing frequency deviation to dynamically adjust the allocation coefficients of virtual inertial control, virtual negative inertial control, and virtual droop control, and finally determine the energy storage power.
[0009] As a preferred technical solution, the power adjustment layer constructs a relationship between the energy storage unit's regulating power and the energy storage battery's lifespan.
[0010] As a preferred technical solution, the relationship between the energy storage unit's regulating power and the energy storage battery's lifespan includes:
[0011] The relationship between depth of discharge and unit regulated power is as follows:
[0012]
[0013] In equation (1), D is the depth of discharge of the energy storage battery, and M E (t) represents the unit regulation power of the virtual inertial control, K E (t) represents the unit regulation power of the virtual droop control, Q represents the energy storage battery capacity, and t represents the current iteration number;
[0014] Formula for energy storage battery life degradation:
[0015]
[0016] In equation (2), B(t) represents the energy storage lifetime loss, a1 and a2 are battery coefficients, and N... r The rated number of battery cycles is denoted by , and D is the depth of discharge of the energy storage battery. The greater the depth of discharge, the greater the lifespan loss of the energy storage battery.
[0017] As a preferred technical solution, the power adjustment layer determines the correction coefficient of the unit adjustment power based on the real-time load disturbance and the ultra-short-term predicted load disturbance, and corrects the unit adjustment power.
[0018] As a preferred technical solution, the correction coefficient for the unit regulating power is calculated as follows:
[0019]
[0020] In equation (3), ΔPL(t) is the actual load disturbance, and ΔPL'(t) is the load disturbance corresponding to the ultra-short-term forecast.
[0021] As a preferred technical solution, the correction of the unit adjustment power specifically involves:
[0022] Corrected unit virtual droop control power K E1 for:
[0023] K E1 =K E ·a (4)
[0024] Corrected unit virtual inertial control power M E1 for:
[0025] M E1 =M E ·a (5)
[0026] K E M is the unit regulating power for virtual droop control. E This refers to the unit adjustment power for virtual inertial control.
[0027] As a preferred technical solution, the coefficient allocation layer uses an improved whale algorithm to solve the problem that the basic whale algorithm is prone to getting trapped in local optima, while having lower complexity and faster search capability.
[0028] As a preferred technical solution, the convergence coefficient of the basic whale algorithm is b:
[0029] b = 2 - 2t / T max (6)
[0030] In equation (6), Tmax is the maximum number of iterations set, and t is the current number of iterations.
[0031] As a preferred technical solution, the improved whale algorithm replaces the linear convergence factor represented by equation (6) in the basic whale algorithm with the nonlinear convergence factor q(t) of equation (7);
[0032]
[0033] As a preferred technical solution, the coefficient allocation layer dynamically adjusts the allocation coefficients for virtual inertial control, virtual negative inertial control, and virtual droop control specifically as follows:
[0034] In virtual droop control, the system frequency deviation Δf is:
[0035]
[0036] In Equation (8), ΔPL(s), ΔPG(s) and ΔPB(s) represent the real-time load disturbance, the output of thermal power units participating in primary frequency regulation and the output of energy storage participating in primary frequency regulation, respectively, and M, D and s represent the grid inertia time constant, load damping coefficient and Laplace operator, respectively.
[0037] Active power increment ΔP under virtual droop control E1 for:
[0038] ΔP E1 =-K E Δf (9)
[0039] In equation (9), Δf is the frequency deviation, and K E The unit regulating power for virtual droop control;
[0040] Active power increment ΔP of virtual inertial control E2 for:
[0041] ΔP E2 =-M E1 d(Δf) / dt (10)
[0042] In equation (10), Δf is the frequency deviation, and M E1 The unit adjustment power for virtual inertial control;
[0043] Active power increment ΔP of virtual negative inertia control E3 for:
[0044] ΔP E3 =M E2 d(Δf) / dt (11)
[0045] In equation (11), Δf is the frequency deviation, and M E2 The unit control power for virtual negative inertia;
[0046] The output power of the energy storage system is:
[0047]
[0048] In equation (12), P is the output power of the energy storage system, α is the correction coefficient for unit adjustment power, A1 is the virtual droop control allocation coefficient, A2 is the virtual positive inertia control allocation coefficient, A3 is the virtual negative inertia control allocation coefficient, and K... E Virtual droop unit control power, M E1 M is the virtual inertial unit control power. E2 The virtual negative inertia unit control power is Δf, where Δf is the frequency deviation.
[0049] Compared with the prior art, the present invention has the following advantages:
[0050] 1) This invention solves the problems of battery life loss and poor frequency regulation effect caused by the participation of energy storage systems in primary frequency regulation;
[0051] 2) Under different operating conditions, the present invention was compared with fixed droop control, variable coefficient droop control and no energy storage control. The results show that the present invention is superior to the other three methods in reducing the life loss of energy storage batteries and improving the primary frequency regulation effect under different operating conditions. Attached Figure Description
[0052] Figure 1 This is a schematic diagram of the primary frequency regulation model combining energy storage and thermal power in this invention;
[0053] Figure 2 This is a comparison chart of the whale algorithm of this invention;
[0054] Figure 3 This is a flowchart of the primary frequency modulation process of the present invention. Detailed Implementation
[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0056] The energy storage-thermal power combined primary frequency regulation model provided by this invention equates the various parts of the thermal power unit and the energy storage system, resulting in the energy storage-thermal power combined primary frequency regulation model as follows: Figure 1 As shown. Figure 1 In the middle: Δf, ΔP L (s), ΔP G (s) and ΔP B (s) represent the grid frequency deviation, real-time load disturbance, output of thermal power units participating in primary frequency regulation, and output of energy storage participating in primary frequency regulation, respectively; K G The droop control coefficient for thermal power units is given by M, D, and s, respectively, representing the grid inertia time constant, load damping coefficient, and Laplace operator. The power adjustment layer of the integrated control strategy dynamically adjusts the unit regulating power, and the parameters of the power adjustment layer are as follows: M E1 For virtual inertial control, unit regulation power, M E2 For virtual negative inertia control, the unit regulation power, K ELet 'a' be the unit regulating power for virtual droop control and 'a' be the correction coefficient for unit regulating power. The coefficient allocation layer of the integrated control strategy dynamically adjusts the allocation coefficients. The coefficient allocation layer has the following parameters: A1 is the allocation coefficient for virtual droop control; A2 is the allocation coefficient for virtual inertia control; and A3 is the allocation coefficient for virtual negative inertia control. The governor and turbine of the thermal power unit determine the frequency characteristics; therefore, governor and turbine models are constructed. G It is the time constant of the governor of the thermal power unit. F HP F IP and F LP These are the power coefficients of the high-pressure cylinder, intermediate-pressure cylinder, and low-pressure cylinder of the steam turbine in a thermal power unit, respectively. RH T CO and T CH These represent the time constants of the reheater in a thermal power unit, the steam volume in the low- and medium-pressure cylinders of a steam turbine, and the high-pressure steam volume in a steam turbine, respectively. The energy storage system consists of two important parts: energy storage batteries and a power generation system (PCS). N represents the number of individual battery energy storage units, and V... batt and V ov These represent the terminal voltage and open-circuit voltage, respectively. R1 is the internal resistance in ohms, and R2, R3, C2, and C3 are the resistance and capacitance values of the parallel branch. rate This represents the rated capacity of the energy storage system. (T) pcs and T d T represents the time constant of the PCS (Power Conversion System) and the delay element. s is the conversion factor.
[0057] The control strategy proposed in this invention consists of two layers: a power adjustment layer and a coefficient allocation layer. Since ultra-short-term forecasting is relatively mature in power grid dispatching, this invention assumes that ultra-short-term load forecast curves have already been obtained. The power adjustment layer uses an improved whale algorithm to adjust the unit adjustment power based on the ultra-short-term load forecast data, thereby reducing battery life loss. The coefficient allocation layer obtains the unit adjustment power from the power adjustment layer and uses an improved whale algorithm to dynamically adjust the allocation coefficients with the goal of minimizing frequency deviation, ultimately determining the energy storage output power.
[0058] The power adjustment layer provided by this invention reduces the lifespan loss of energy storage when participating in frequency regulation by adjusting the unit adjustment power. The unit adjustment power is solved by an improved whale algorithm based on ultra-short-term load forecast data, and then the unit adjustment power is corrected based on real-time load disturbances.
[0059] An energy storage lifetime model is established in the power regulation layer provided by this invention. The lifetime of the lithium battery used for energy storage is mainly determined by two factors: temperature and depth of discharge. Since the lithium battery is installed in a container with a constant temperature, temperature is not considered as the main factor affecting the life of the energy storage battery. When the temperature is constant, the deeper the depth of discharge of the energy storage battery, the fewer the number of cycles the energy storage battery has. When the energy storage battery participates in primary frequency regulation, it performs different actions according to different load disturbances, resulting in changes in the depth of discharge. Therefore, the depth of discharge is considered as the main factor affecting the life of the energy storage battery. The depth of discharge of the energy storage battery is closely related to the battery power, thus constructing a relationship between the unit regulation power of the energy storage battery and the life of the energy storage battery. Relationship between depth of discharge and unit regulation power:
[0060]
[0061] In equation (1), D is the depth of discharge of the energy storage battery, and M E K represents the unit regulation power for virtual inertial control. E Let Q be the unit regulation power of virtual droop control and Q be the energy storage battery capacity. Within a certain time period, the greater the unit regulation power of virtual inertial control and virtual droop control, the greater the depth of discharge of the energy storage battery.
[0062] Formula for energy storage battery life degradation:
[0063]
[0064] In equation (2), B(t) represents the energy storage lifetime loss, a1 and a2 are battery coefficients, and N... r Where is the number of battery cycles and D is the depth of discharge. The greater the depth of discharge, the greater the loss of battery life.
[0065] The unit regulation power correction for ultra-short-term load forecasting will have some error. When the ultra-short-term load forecast and the actual load disturbance are inconsistent, the solved virtual droop unit regulation power and virtual inertia unit regulation power will affect the frequency regulation results. Therefore, it is necessary to determine the correction factor for unit regulation power based on the real-time load disturbance and the load disturbance predicted in the ultra-short-term forecast, and then correct the unit regulation power accordingly. α is the correction factor for unit regulation power:
[0066]
[0067] In equation (3): ΔP L (t) represents the actual load disturbance, ΔP L ′(t) represents the load disturbance corresponding to the ultra-short-term forecast.
[0068] The corrected unit virtual droop control power is:
[0069] K E1 =KE ·a (4)
[0070] The corrected unit virtual inertial control power is:
[0071] M E1 =M E ·a (5)
[0072] By setting a correction factor for unit regulation power, the impact of short-term load forecast accuracy on unit regulation power is resolved.
[0073] The improved whale optimization algorithm (WOA) provided by this invention mainly includes surrounding the prey, capturing the prey, and searching for the prey. WOA has low complexity, fast convergence speed, and strong search capability; however, because the convergence coefficient b of the whale algorithm changes linearly, it is prone to getting trapped in local optima. The convergence coefficient b of the whale algorithm is:
[0074] b = 2 - 2t / T max (6)
[0075] In equation (6), T max The maximum number of iterations is set here, and t is the current number of iterations.
[0076] The improved whale algorithm solves the problem of the basic whale algorithm easily getting trapped in local optima, while possessing lower complexity and faster search capability. The linear change of the convergence factor in the basic whale algorithm leads to the problem of getting trapped in local optima in the later stages of the search. Therefore, the improved whale algorithm replaces the linear convergence factor represented by equation (6) in the basic whale algorithm with a nonlinear convergence factor, such as equation (7). The nonlinear convergence factor solves the problem of the basic whale algorithm getting trapped in local optima due to the linear change of the linear convergence factor b. The convergence factor equation is shown in equation (7):
[0077]
[0078] Because load disturbances change rapidly and the unit regulation power varies over a wide range, the improved whale algorithm has low complexity and fast search capability, enabling it to quickly solve for the unit regulation power. The constraints are set as follows: maximum number of iterations is 400, and the number of search agents is 30. The basic process of using the improved whale algorithm to solve for battery lifetime loss is as follows: using energy storage lifetime loss B(t) as the fitness function, considering load disturbances, and determining K based on minimizing energy storage lifetime loss. E and M E .
[0079] The improved whale algorithm and the comparison chart of the whale algorithm are shown below. Figure 2As shown, the improved whale algorithm results in less energy storage lifetime loss per unit of regulation power, and the improved whale algorithm requires fewer iterations than the original whale algorithm.
[0080] In the coefficient allocation layer provided by this invention, the three control methods for primary frequency regulation involving energy storage—virtual droop control, virtual inertia control, and virtual negative inertia control—each have their own drawbacks. Virtual inertia control cannot improve steady-state deviation. Virtual droop control, when used alone, cannot improve the rate of frequency change or the maximum frequency deviation. The coefficient allocation layer provided by this invention combines these three methods for control, which can compensate for the shortcomings of a single control method, thereby improving the frequency regulation effect.
[0081] In the virtual droop control provided by this invention, the system frequency deviation is:
[0082]
[0083] In equation (8), ΔP L (s), ΔP G (s) and ΔP B (s) Real-time load disturbance, output of thermal power units participating in primary frequency regulation, and output of energy storage participating in primary frequency regulation, respectively.
[0084] The active power increment for virtual droop control is:
[0085] ΔP E1 =-K E Δf (9)
[0086] In equation (9), Δf is the frequency deviation, and K E The unit adjustment power for virtual droop control.
[0087] In the virtual inertial control provided by this invention, the active power increment of the virtual inertial control is:
[0088] ΔP E2 =-M E1 d(Δf) / dt (10)
[0089] In equation (10), Δf is the frequency deviation, and M E1 This refers to the unit adjustment power for virtual inertial control.
[0090] When the frequency deviation is large, virtual inertial control can quickly reduce the frequency deviation, give full play to the speed of the energy storage system to participate in frequency regulation, and prevent further deterioration of the frequency.
[0091] In the virtual inertial control provided by this invention, the active power increment of virtual negative inertial control is:
[0092] ΔP E3 =M E2d(Δf) / dt (11)
[0093] In equation (11), Δf is the frequency deviation, and M E2 The unit control power for virtual negative inertia.
[0094] During the frequency recovery phase, virtual negative inertia control can accelerate frequency recovery.
[0095] In the unit power coefficient adjustment method provided by this invention, after the unit regulation power is determined in the power adjustment layer, the coefficient allocation layer optimizes the frequency regulation effect of energy storage participating in primary frequency regulation by dynamically adjusting the allocation coefficients. Using the frequency deviation of equation (8) as the fitness function, the constraint adjustment is set as follows: the maximum number of iterations is 400, the number of search agents is 30, and the allocation coefficients A1, A2, and A3 are dynamically adjusted. Finally, the output power of energy storage participating in primary frequency regulation is determined.
[0096] The output power of the energy storage system is:
[0097]
[0098] In equation (12), P is the output power of the energy storage system, α is the correction coefficient for unit adjustment power, A1 is the virtual droop control allocation coefficient, A2 is the virtual positive inertia control allocation coefficient, A3 is the virtual negative inertia control allocation coefficient, and K... E Virtual droop unit control power, M E1 M is the virtual inertial unit control power. E2 The virtual negative inertia unit control power is Δf, where Δf is the frequency deviation.
[0099] In the flowchart of the two-layer control strategy provided in this invention, as follows: Figure 3 As shown, in the power adjustment layer, the unit adjustment power is adjusted with the goal of minimizing energy storage lifetime loss. First, an energy storage lifetime loss model is established, and ultra-short-term load forecast data is obtained. The constraints of the improved whale algorithm are determined, and the population is initialized. Using the energy storage lifetime loss B(t) as the fitness function, the corresponding fitness is calculated and the parameters are updated to solve for the unit adjustment power. Based on real-time load disturbances, correction coefficients are calculated to correct the unit adjustment power, and the unit adjustment power is transmitted to the coefficient allocation layer. In the coefficient allocation layer, the constraints of the improved whale algorithm are determined. Using Δf as the fitness function, the corresponding fitness value is calculated, the allocation coefficients are determined, and finally, the energy storage battery output is obtained.
[0100] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A two-layer control system for combined energy storage and thermal power primary frequency regulation, characterized in that, The system includes a power adjustment layer and a coefficient allocation layer. The power adjustment layer uses an improved whale algorithm based on ultra-short-term load forecast data to adjust the unit regulation power of the energy storage with the goal of minimizing energy storage lifetime loss, and then transmits the unit regulation power to the coefficient allocation layer. The improved whale algorithm uses a nonlinear convergence factor. To replace the linear convergence factor in the basic whale algorithm; The coefficient allocation layer uses an improved whale algorithm with the goal of minimizing frequency deviation to dynamically adjust the allocation coefficients of virtual inertial control, virtual negative inertial control, and virtual droop control, and finally determine the energy storage power. The coefficient allocation layer dynamically adjusts the allocation coefficients for virtual inertial control, virtual negative inertial control, and virtual droop control as follows: In virtual droop control, the system's frequency deviation for: (8) In equation (8), ΔPL(s), ΔPG(s) and ΔPB(s) represent the real-time load disturbance, the output of thermal power units participating in primary frequency regulation and the output of energy storage participating in primary frequency regulation, respectively, and M, D and s represent the grid inertia time constant, load damping coefficient and Laplace operator, respectively. Active power increment of virtual droop control for: (9) In equation (9), Δf is the frequency deviation. The unit regulating power for virtual droop control; Active power increment of virtual inertial control for: (10) In equation (10), Δf is the frequency deviation. The unit adjustment power for virtual inertial control; Active power increment of virtual negative inertia control for: (11) In equation (11), Δf is the frequency deviation. The unit control power for virtual negative inertia; The output power of the energy storage system is: (12) In equation (12), P is the output power of the energy storage system, α is the correction coefficient for unit adjustment power, A1 is the virtual droop control allocation coefficient, A2 is the virtual positive inertia control allocation coefficient, A3 is the virtual negative inertia control allocation coefficient, and K... E Virtual droop unit control power, M E1 M is the virtual inertial unit control power. E2 The virtual negative inertia unit control power is Δf, where Δf is the frequency deviation.
2. The dual-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 1, characterized in that, The power adjustment layer is constructed to establish the relationship between the energy storage unit's regulating power and the energy storage battery's lifespan.
3. The dual-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 2, characterized in that, The relationship between the energy storage unit's regulation power and the energy storage battery's lifespan includes: The relationship between depth of discharge and unit regulated power is as follows: (1) In equation (1), D represents the depth of discharge of the energy storage battery. For virtual inertial control, the unit adjustment power, Here, Q represents the unit regulating power of the virtual droop control, Q is the energy storage battery capacity, and t is the current iteration number. Formula for energy storage battery life degradation: (2) In equation (2), B(t) represents the energy storage lifetime loss, and a1 and a2 are battery coefficients. The rated number of battery cycles is denoted by , and D is the depth of discharge of the energy storage battery. The greater the depth of discharge, the greater the lifespan loss of the energy storage battery.
4. The dual-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 1, characterized in that, The power adjustment layer determines the correction coefficient of unit adjustment power based on real-time load disturbance and ultra-short-term predicted load disturbance, and corrects the unit adjustment power.
5. A two-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 4, characterized in that, The correction factor for the unit regulating power is calculated as follows: (3) In equation (3), ΔPL(t) is the actual load disturbance, and ΔPL'(t) is the load disturbance corresponding to the ultra-short-term forecast.
6. A two-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 5, characterized in that, The specific steps for correcting the unit adjustment power are as follows: Corrected unit virtual droop control power for: (4) Corrected unit virtual inertial control power for: (5) For virtual droop control, the unit adjustment power. This refers to the unit adjustment power for virtual inertial control.
7. A two-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 1, characterized in that, The coefficient allocation layer uses an improved whale algorithm to solve the problem that the basic whale algorithm is prone to getting trapped in local optima, while having lower complexity and faster search capability.
8. A two-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 7, characterized in that, The convergence coefficient of the basic whale algorithm is b: (6) In equation (6), Tmax is the maximum number of iterations set, and t is the current number of iterations.
9. A two-layer control system for combined primary frequency regulation of energy storage and thermal power as described in claim 8, characterized in that, The improved whale algorithm uses Equation (7) as the nonlinear convergence factor. Replace the linear convergence factor represented by equation (6) in the basic whale algorithm; (7)。
Citation Information
Patent Citations
System oscillation suppression method and device based on energy storage type virtual synchronous power generation technology
CN109980686A
Energy storage auxiliary wind power primary frequency modulation control method based on dynamic task coefficient
CN113452036A
