Network construction type optimization control method of battery energy storage system

Abstract

The invention relates to a network construction type optimization control method for a battery energy storage system, and the method comprises the steps: building a day-ahead stage optimization model, and calculating a power offset and a frequency adjustment droop control coefficient; calculating a power set point of the battery energy storage system by using a model predictive control algorithm; and converting the power set point obtained in the intra-day stage into a feasible frequency set point of the grid-forming inverter, establishing an inverter capacity curve, performing real-time problem expression, solving an optimization problem, and obtaining an optimal reference frequency and an optimal reference voltage of the inverter compatible with the inverter capacity curve. The battery energy storage system is utilized to realize the schedulability of the power distribution feeder line of the network construction type control unit; the function of simultaneously providing voltage regulation and frequency regulation by the network-forming type control unit is realized; the battery energy storage system is utilized to the maximum extent under the conditions of high randomness of power demand, distributed power generation grid connection, power grid frequency and other uncertainties.

Description

Technical Field

[0001] The present invention relates to the technical field of battery energy storage system control, and in particular to a grid-based optimization control method for a battery energy storage system. Background Art

[0002] Grid control was first proposed in the context of microgrids and is expected to be widely used in power transmission and distribution networks with a wide range of application scenarios. For a "double-high" power system with a high proportion of new energy and a high proportion of power electronic equipment, grid control technology can ensure stable operation of the system under normal, disturbed and fault conditions without relying on the assistance of synchronous generators. The role of grid control technology in future power systems includes improving the system short-circuit current level and improving system strength; providing damping and inertia for the system and improving system frequency stability; responding quickly when the system loses step and is disconnected, improving the stability of the system's first swing cycle, and actively supporting system recovery; and weakening the impact of inter-harmonics and unbalanced voltage in the power system.

[0003] With the rapid development of the "double-high" power system, grid-type converters will gradually replace synchronous generators, and the problems that come with it are gradually manifested. The grid-type converter adopts a power synchronization strategy. In a weak power grid, its small-disturbance synchronization stability is better than that of the grid-following converter. However, if the grid strength is large, its small-disturbance stability margin is reduced, and it may lose synchronization with the grid. The grid-type converter has similar synchronization characteristics to synchronous generators, and may also experience transient instability similar to synchronous generators under large interference. When a short-circuit fault occurs in the power grid, the grid-type converter may cause a large short-circuit current to maintain the internal potential unchanged, which will harm the converter device. For application scenarios such as distributed systems and microgrids, the smooth transition between isolated grid / grid-connected modes is particularly important, and the current smooth transition capability of the grid-type control is not ideal. Finally, in the future, a huge number of grid-type converters will be connected to the power grid. How to achieve coordinated control of converters under the conditions of complex system dynamic characteristics is an urgent problem to be solved. Summary of the invention

[0004] In order to solve the problem that distribution feeder scheduling is difficult and it is difficult to provide voltage and frequency regulation at the same time, the purpose of the present invention is to provide a grid-type optimization control method for a battery energy storage system that can realize the dispatchability of the distribution feeder of a grid-type control unit and can simultaneously realize voltage regulation and frequency regulation.

[0005] To achieve the above object, the present invention adopts the following technical solution: a grid-type optimization control method for a battery energy storage system, comprising the following steps in order:

[0006] (1) Day-ahead stage: Establish a day-ahead optimization model to calculate the power offset and frequency regulation droop control coefficient;

[0007] (2) Intraday stage: The droop control coefficient is adjusted according to the power offset and frequency, and the power set point of the battery energy storage system is calculated using the model predictive control algorithm;

[0008] (3) Real-time control stage: Perform set point conversion of the grid-connected inverter, convert the power set point obtained in the intraday stage into a feasible frequency set point of the grid-connected inverter, establish the inverter capability curve, perform real-time problem formulation, solve the optimization problem, and obtain the optimal reference frequency and optimal reference voltage of the inverter that are compatible with the inverter capability curve.

[0009] The step (1) specifically refers to:

[0010]

[0011]

[0012] The constraints of the day-ahead optimization model are:

[0013]

[0014]

[0015]

[0016]

[0017] Where T is the total scheduling window time, N represents the discretization time step, n = 0, ..., -1; and They represent the upper and lower limits of the feeder consumption prediction value at the corresponding time step, respectively. 0 Represents the power deviation curve of the battery energy storage system; σ f represents the frequency regulation droop control coefficient, and They represent the integral of the upper limit of the frequency deviation and the integral of the lower limit of the frequency deviation over a period of time, respectively. SOE max and SOE min They represent the upper and lower limits of the state of charge of the battery energy storage system, P max and P min They represent the upper and lower limits of the power of the battery energy storage system, E nom represents the nominal energy of the battery energy storage system, SOE0 represents the state of charge of the battery energy storage system at time 0; F represents the power offset, F i Indicates the offset of the power corresponding to the step, F n represents the power offset when i=n, represents the upper limit of the feeder consumption prediction value when i=n, It represents the lower limit of the feeder consumption prediction value when i=n.

[0018] The step (2) specifically comprises the following steps in order:

[0019] (2a) The sampling time of the model predictive control algorithm is 10 seconds. K = 0, 1, 2, ..., K-1 is introduced to represent the rolling 10-second interval, where K = 8640 represents the number of 10-second periods in 24 hours, and the consumption set value retrieved from the scheduling plan for the current 5-minute period is It is expressed by the time step k as:

[0020]

[0021] Among them, [·] represents the nearest integer to the parameter; Represents a scheduling plan;

[0022] Use the first 10 seconds and the last 10 seconds of the current 5 minutes respectively and Indicates that we get:

[0023]

[0024]

[0025] (2b) By averaging the existing power before k, the average composite power flow G at the common coupling point is obtained. k :

[0026]

[0027] Among them, L j represents the feeder power consumption at time j, P j represents the bidirectional power of the battery energy storage system at time j;

[0028] (2c) Calculate the expected average composite flow at the common coupling point at the end of the 5-minute window

[0029]

[0030] in, represents the feeder power consumption at k-1;

[0031] Energy error e between reality and target within 5 minutes time interval k for:

[0032] in, It indicates the deviation caused by the frequency control of the grid-connected inverter, and its formula is:

[0033]

[0034] Among them, f j is the measured value of the frequency at time j;

[0035] (2d) Based on the physical constraints of the battery energy storage system’s SOC, DC voltage, and current operating limits, a model predictive control is developed to enable e k Minimization, the optimization problem is stated as:

[0036]

[0037]

[0038] The constraints are:

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046] in, is the calculated control action trajectory, 1 represents a column vector of all 1s, and the symbol ≤ represents a component inequality; in the above formula, except for the matrix H and the column vector 1, all bold symbols represent the sequence obtained by superimposing the reference variables in the column vector in time; are the battery DC voltage and current, respectively, α = 10 / 3600 is the conversion factor from the 10-second average power to energy expressed in kilowatt-hours, φ v , Both represent transfer matrices, x k represents the state vector of the three-time constant electrical equivalent circuit model, v min 、v max They represent the minimum and maximum voltage of the battery energy storage system respectively, and SOC represents the battery state of charge of the battery energy storage system; φ soc and are transient matrices obtained from the SOC model of the battery energy storage system, SOE max and SOE min They represent the upper and lower limits of the state of charge of the battery energy storage system, Indicates k time The state of charge of the battery energy storage system at time T is the total dispatch window time; e k Indicates the energy error between reality and target within a 5-minute time interval; i min and i max Respectively represent the minimum and maximum current of the battery energy storage system; Δi min and Δi max Respectively represent the change rate of the minimum and maximum current of the battery energy storage system; SOC k Represents the state of charge of the battery energy storage system at time k; the matrix The formula is:

[0047]

[0048] (2e) Solve the optimization problem at each time step k and get Considering only the first component of the current control trajectory as the drive, we get Will Converted to a power set point The calculation formula is as follows:

[0049]

[0050] Among them, v k represents the voltage at k, represents the current at time k obtained by solving the optimization problem.

[0051] In step (3), the set point conversion of the grid-type inverter specifically refers to: the real-time controller converts the power set point into a frequency reference set point to feed back to the grid-type inverter, and the power output P of the grid-type inverter is:

[0052] P=σ f (ff nom )+σ f (f nom -f ref,t )=P fcr +P set,t

[0053] Among them, f nom Indicates the nominal frequency, P fcr represents the output power of the frequency containment action, f is the grid frequency; σ f Indicates the frequency regulation droop control coefficient, P set,t represents the active power set point; fref,t represents the reference frequency at time t;

[0054] The active power set point P of the grid-connected inverter at time t set,t With the input reference frequency f ref,t The linear relationship is:

[0055] P set,t =σ f ·(f nom -f ref,t )

[0056] Similarly, for the reactive power set point Q at time t set,t :

[0057] Q set,t =σ v ·(v nom -v ref,t )

[0058] Among them, v nom represents the nominal voltage, σ v represents the voltage regulation droop control coefficient, v ref,t Represents the reference voltage at time t.

[0059] In step (3), the real-time problem statement specifically refers to: given an active power set point P set,t , reactive power set point Q set,t , the optimal reference of the grid-connected inverter is calculated by solving the optimization problem;

[0060] The optimization problem is:

[0061]

[0062] in, They represent the optimal reference frequency and optimal reference voltage at time t respectively, They represent the optimal active power set point and the optimal reactive power set point, respectively, P , Q Respectively represent active power control coefficient and reactive power control coefficient; P set,t Indicates the active power set point; Q set,t Indicates reactive power set point;

[0063] The solution to the optimization problem is Optimal reference frequency and the optimal reference voltage They are:

[0064]

[0065]

[0066] In the formula, f nom represents the nominal frequency, v nom represents the nominal voltage, σ f represents the frequency regulation droop control coefficient, σ v Represents the voltage regulation droop control coefficient.

[0067] It can be seen from the above technical scheme that the beneficial effects of the present invention are: first, in the day-ahead stage, the present invention determines the scheduling plan at intervals of 5 minutes through production and sales forecasts and calculation of the decline rate of frequency control reserves, thereby solving the problem of optimizing the allocation of power and energy budget of the battery energy storage system; second, the present invention uses a battery energy storage system to realize the dispatchability of the distribution feeder of the grid-connected control unit; third, the present invention uses a battery energy storage system to realize the function of simultaneously providing voltage regulation and frequency regulation of the grid-connected control unit; fourth, the present invention is used for a battery energy storage system using a grid-connected inverter interface, which solves the current situation of relying on the grid-following control strategy to provide the best service, namely voltage regulation and frequency regulation; fifth, the present invention realizes the maximum utilization of the battery energy storage system under the conditions of high randomness of power demand, distributed power generation grid connection, and grid frequency uncertainty. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 Schematic diagram of the feeder and corresponding power flow applied by the present invention.

[0069] Figure 2 is a flow chart of the method of the present invention;

[0070] Figure 3 It is the control time axis of the present invention. DETAILED DESCRIPTION

[0071] like Figure 2 As shown, a grid-based optimization control method for a battery energy storage system includes the following steps in sequence:

[0072] (1) Day-ahead stage: Establish a day-ahead optimization model to calculate the power offset and frequency regulation droop control coefficient;

[0073] (2) Intraday stage: The droop control coefficient is adjusted according to the power offset and frequency, and the power set point of the battery energy storage system is calculated using the model predictive control algorithm;

[0074] (3) Real-time control stage: Perform set point conversion of the grid-connected inverter, convert the power set point obtained in the intraday stage into a feasible frequency set point of the grid-connected inverter, establish the inverter capability curve, perform real-time problem formulation, solve the optimization problem, and obtain the optimal reference frequency and optimal reference voltage of the inverter that are compatible with the inverter capability curve.

[0075] The step (1) specifically refers to:

[0076]

[0077]

[0078] The constraints of the day-ahead optimization model are:

[0079]

[0080]

[0081]

[0082]

[0083] Where T is the total scheduling window time, N represents the discretization time step, n = 0, ..., -1; and They represent the upper and lower limits of the feeder consumption prediction value at the corresponding time step, respectively. 0 Represents the power deviation curve of the battery energy storage system; σ f represents the frequency regulation droop control coefficient, and They represent the integral of the upper limit of the frequency deviation and the integral of the lower limit of the frequency deviation over a period of time, respectively. SOE max and SOE min They represent the upper and lower limits of the state of charge of the battery energy storage system, P max and P min They represent the upper and lower limits of the power of the battery energy storage system, E nom represents the nominal energy of the battery energy storage system, SOE0 represents the state of charge of the battery energy storage system at time 0; F represents the power offset, F i Indicates the offset of the power corresponding to the step, F n represents the power offset when i=n, represents the upper limit of the feeder consumption prediction value when i=n, It represents the lower limit of the feeder consumption prediction value when i=n.

[0084] The step (2) specifically comprises the following steps in order:

[0085] (2a) Figure 3 As shown, the sampling time of the model predictive control algorithm is 10 seconds, and K = 0, 1, 2, ..., K-1 is introduced to represent the rolling 10-second interval, where K = 8640 represents the number of 10-second cycles within 24 hours, and the consumption set value retrieved from the scheduling plan of the current 5-minute period It is expressed by the time step k as:

[0086]

[0087] Among them, [·] represents the nearest integer to the parameter; Represents a scheduling plan;

[0088] Use the first 10 seconds and the last 10 seconds of the current 5 minutes respectively and Indicates that we get:

[0089]

[0090]

[0091] (2b) By averaging the existing power before k, the average composite power flow G at the common coupling point is obtained. k :

[0092]

[0093] Among them, L j represents the feeder power consumption at time j, P j represents the bidirectional power of the battery energy storage system at time j;

[0094] (2c) Calculate the expected average composite flow at the common coupling point at the end of the 5-minute window

[0095]

[0096] in, represents the feeder power consumption at k-1;

[0097] Energy error e between reality and target within 5 minutes time interval k for:

[0098]

[0099] in, It indicates the deviation caused by the frequency control of the grid-connected inverter, and its formula is:

[0100]

[0101] Among them, f j is the measured value of the frequency at time j;

[0102] (2d) Based on the physical constraints of the battery energy storage system’s SOC, DC voltage, and current operating limits, a model predictive control is developed to enable e k Minimization, the optimization problem is stated as:

[0103]

[0104]

[0105] The constraints are:

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113] in, is the calculated control action trajectory, 1 represents a column vector of all 1s, and the symbol ≤ represents a component inequality; in the above formula, except for the matrix H and the column vector 1, all bold symbols represent the sequence obtained by superimposing the reference variables in the column vector in time; are the battery DC voltage and current, respectively, α = 10 / 3600 is the conversion factor from the 10-second average power to energy expressed in kilowatt-hours, φ v , Both represent transfer matrices, x k represents the state vector of the three-time constant electrical equivalent circuit model, v min 、v max They represent the minimum and maximum voltage of the battery energy storage system respectively, and SOC represents the battery state of charge of the battery energy storage system; φ soc and are transient matrices obtained from the SOC model of the battery energy storage system, SOE max and SOE min They represent the upper and lower limits of the state of charge of the battery energy storage system, Indicates k time The state of charge of the battery energy storage system at time T is the total dispatch window time; e k Indicates the energy error between reality and target within a 5-minute time interval; i min and i max Respectively represent the minimum and maximum current of the battery energy storage system; Δi min and Δi max Respectively represent the change rate of the minimum and maximum current of the battery energy storage system; SOC k Represents the state of charge of the battery energy storage system at time k; the matrix The formula is:

[0114]

[0115] (2e) Solve the optimization problem at each time step k and get Considering only the first component of the current control trajectory as the drive, we get Will Converted to a power set point The calculation formula is as follows:

[0116]

[0117] Among them, v k represents the voltage at k, represents the current at time k obtained by solving the optimization problem.

[0118] In step (3), the set point conversion of the grid-type inverter specifically refers to: the real-time controller converts the power set point into a frequency reference set point to feed back to the grid-type inverter, and the power output P of the grid-type inverter is:

[0119] P=σ f (ff nom )+σ f (f nom -f ref,t )=P fcr +P set,t

[0120] Among them, f nom Indicates the nominal frequency, P fcr represents the output power of the frequency containment action, f is the grid frequency; σ f Indicates the frequency regulation droop control coefficient, P set,t represents the active power set point; f ref,t represents the reference frequency at time t;

[0121] The active power set point P of the grid-connected inverter at time t set,tWith the input reference frequency f ref,t The linear relationship is:

[0122] P set,t =σ f ·(f nom -f ref,t )

[0123] Similarly, for the reactive power set point Q at time t set,t :

[0124] Q set,t =σ v ·(v nom -v ref,t )

[0125] Among them, v nom represents the nominal voltage, σ v represents the voltage regulation droop control coefficient, v ref,t Represents the reference voltage at time t.

[0126] In step (3), the real-time problem statement specifically refers to: given an active power set point P set,t , reactive power set point Q set,t , the optimal reference of the grid-connected inverter is calculated by solving the optimization problem;

[0127] The optimization problem is:

[0128]

[0129] in, They represent the optimal reference frequency and optimal reference voltage at time t respectively, They represent the optimal active power set point and the optimal reactive power set point, respectively, P , Q Respectively represent active power control coefficient and reactive power control coefficient; P set,t Indicates the active power set point; Q set,t Indicates reactive power set point;

[0130] The solution to the optimization problem is Optimal reference frequency and the optimal reference voltage They are:

[0131]

[0132]

[0133] In the formula, f nom represents the nominal frequency, v nomrepresents the nominal voltage, σ f represents the frequency regulation droop control coefficient, σ v Represents the voltage regulation droop control coefficient.

[0134] Embodiment 1

[0135] like Figure 1 As shown, a 20kV distribution feeder equipped with a battery energy storage system at the experimental site includes a group of buildings characterized by a 140 kW base load, carrying a 105 kWh rooftop photovoltaic installation and a 720 kWh / 500 kWh grid-type lithium titanate battery energy storage system. The battery energy storage system is connected to the grid through a three-phase two-level inverter with an IGBT / diode pair, and its characteristics are: rated power 720kVA; nominal AC voltage 300V; DC voltage 500~890V; DC rated current 1360A; round-trip efficiency ≥97%. The target grid is a radial topology, characterized by a coaxial cable line with a cross-section of 95mm 2 , with a length of several hundred meters, so network losses are negligible. The measurement system consists of a distributed sensing infrastructure based on phasor measurement units. Due to the fast reporting rate (50 frames per second) and high accuracy (1 standard deviation equals 0.001 degree) of the phasor measurement units, the measurement infrastructure allows to obtain real-time and accurate power flow information.

[0136] In summary, the present invention solves the problem of optimizing the allocation of electricity and energy budget of the battery energy storage system by predicting production and sales and calculating the decline rate of the frequency control reserve at intervals of 5 minutes in the day-ahead stage; realizes the dispatchability of the distribution feeder of the grid-forming control unit; realizes the function of the grid-forming control unit to provide voltage regulation and frequency regulation at the same time; solves the current situation of relying on the grid-following control strategy to provide the best service, namely voltage regulation and frequency regulation; realizes the maximum utilization of the battery energy storage system under the conditions of high randomness of electricity demand, distributed generation grid connection and grid frequency and other uncertainties.

Claims

1. A grid-based optimization control method for a battery energy storage system, characterized in that: The method comprises the following steps in order: (1) Day-ahead stage: Establish a day-ahead optimization model to calculate the power offset and frequency regulation droop control coefficient; (2) Intraday stage: The droop control coefficient is adjusted according to the power offset and frequency, and the power set point of the battery energy storage system is calculated using the model predictive control algorithm; (3) Real-time control stage: Perform set point conversion of the grid-connected inverter, convert the power set point obtained in the intraday stage into a feasible frequency set point of the grid-connected inverter, establish the inverter capability curve, perform real-time problem formulation, solve the optimization problem, and obtain the optimal reference frequency and optimal reference voltage of the inverter that are compatible with the inverter capability curve.

2. The grid-based optimization control method for a battery energy storage system according to claim 1, characterized in that: The step (1) specifically refers to: The constraints of the day-ahead optimization model are: Where T is the total scheduling window time, N represents the discretized time step, n = 0, ..., N-1; and They represent the upper and lower limits of the feeder consumption prediction value at the corresponding time step, respectively. 0 Represents the power deviation curve of the battery energy storage system; σ f represents the frequency regulation droop control coefficient, and They represent the integral of the upper limit of the frequency deviation and the integral of the lower limit of the frequency deviation over a period of time, respectively. SOE max and SOE min They represent the upper and lower limits of the state of charge of the battery energy storage system, P max and P min They represent the upper and lower limits of the power of the battery energy storage system, E nom represents the nominal energy of the battery energy storage system, SOE0 represents the state of charge of the battery energy storage system at time 0; F represents the power offset, F i Indicates the offset of the power corresponding to the step, F n represents the power offset when i=n, represents the upper limit of the feeder consumption prediction value when i=n, It represents the lower limit of the feeder consumption prediction value when i=n.

3. The grid-based optimization control method for a battery energy storage system according to claim 1, characterized in that: The step (2) specifically comprises the following steps in order: (2a) The sampling time of the model predictive control algorithm is 10 seconds. K = 0, 1, 2, ..., K-1 is introduced to represent the rolling 10-second interval, where K = 8640 represents the number of 10-second periods in 24 hours, and the consumption set value retrieved from the scheduling plan for the current 5-minute period is It is expressed by the time step k as: Among them, [·] represents the nearest integer to the parameter; Represents a scheduling plan; Use the first 10 seconds and the last 10 seconds of the current 5 minutes respectively k and Indicates that we get: (2b) By averaging the existing power before k, the average composite power flow G at the common coupling point is obtained. k : Among them, L j represents the feeder power consumption at time j, P j represents the bidirectional power of the battery energy storage system at time j; (2c) Calculate the expected average composite flow at the common coupling point at the end of the 5-minute window in, represents the feeder power consumption at k-1; Energy error e between reality and target within 5 minutes time interval k for: in, It indicates the deviation caused by the frequency control of the grid-connected inverter, and its formula is: Among them, f j is the measured value of the frequency at time j; (2d) Based on the physical constraints of the battery energy storage system’s SOC, DC voltage, and current operating limits, a model predictive control is developed to enable e k Minimization, the optimization problem is stated as: The constraints are: in, is the calculated control action trajectory, 1 represents a column vector of all 1s, and the symbol is a component inequality; in the above formula, except for the matrix H and the column vector 1, all bold symbols represent the sequence obtained by superimposing the reference variable in the column vector in time; are the battery DC voltage and current, respectively, α = 10 / 3600 is the conversion factor from the 10-second average power to energy expressed in kilowatt-hours, φ v , Both represent transfer matrices, x k represents the state vector of the three-time constant electrical equivalent circuit model, v min 、v max They represent the minimum and maximum voltage of the battery energy storage system respectively, and SOC represents the battery state of charge of the battery energy storage system; φ soc and are transient matrices obtained from the SOC model of the battery energy storage system, SOE max and SOE min They represent the upper and lower limits of the state of charge of the battery energy storage system, Indicates k time The state of charge of the battery energy storage system at time T is the total dispatch window time; e k Indicates the energy error between reality and target within a 5-minute time interval; i min and i max Respectively represent the minimum and maximum current of the battery energy storage system; Δi min and Δi max Respectively represent the change rate of the minimum and maximum current of the battery energy storage system; SOC k Represents the state of charge of the battery energy storage system at time k; the matrix The formula is: (2e) Solve the optimization problem at each time step k and get Considering only the first component of the current control trajectory as the drive, we get Will Converted to a power set point The calculation formula is as follows: Among them, v k represents the voltage at k, represents the current at time k obtained by solving the optimization problem.

4. The grid-based optimization control method for a battery energy storage system according to claim 1, characterized in that: In step (3), the set point conversion of the grid-type inverter specifically refers to: the real-time controller converts the power set point into a frequency reference set point to feed back to the grid-type inverter, and the power output P of the grid-type inverter is: P=σ f (ff nom )+σ f (f nom -f ref，t )=P fcr +P set，t Among them, f nom Indicates the nominal frequency, P fcr represents the output power of the frequency containment action, f is the grid frequency; σ f Indicates the frequency regulation droop control coefficient, P set，t represents the active power set point; f ref，t represents the reference frequency at time t; The active power set point P of the grid-connected inverter at time t set，t With the input reference frequency f ref，t The linear relationship is: P set，t =σ f ·(f nom -f ref，t ) Similarly, for the reactive power set point Q at time t set，t : Q set，t =σ v ·(v nom -v ref，t ) Among them, v nom represents the nominal voltage, σ v represents the voltage regulation droop control coefficient, v ref，t Represents the reference voltage at time t.

5. The grid-based optimization control method for a battery energy storage system according to claim 1, characterized in that: In step (3), the real-time problem statement specifically refers to: given an active power set point P set，t , reactive power set point Q set，t , the optimal reference of the grid-connected inverter is calculated by solving the optimization problem; The optimization problem is: in, They represent the optimal reference frequency and optimal reference voltage at time t respectively, They represent the optimal active power set point and the optimal reactive power set point, respectively, P , Q Respectively represent active power control coefficient and reactive power control coefficient; P set，t Indicates the active power set point; Q set，t Indicates reactive power set point; The solution to the optimization problem is Optimal reference frequency and the optimal reference voltage They are: In the formula, f nom represents the nominal frequency, v nom represents the nominal voltage, σ f represents the frequency regulation droop control coefficient, σ v Represents the voltage regulation droop control coefficient.

Images