Self-adaptive dynamic partition type control method and system for energy storage power supply bus

Abstract

The invention belongs to the technical field of power supply, and particularly relates to a self-adaptive dynamic partition type control method and system for an energy storage power supply bus. According to the bus voltage stability control requirement of an energy storage power supply system, actual compensation is carried out on bus current and voltage measurement values through an electrical parameter perturbation dynamic estimator, and a bus voltage control law v is designed by adopting a multi-state dynamic combination function, a softening factor rho and a switching gain function K; the optimal control parameters are adaptively optimized by adopting the softening factor characteristic partition controller g (rho) and the switching gain function characteristic partition controller j (K), so that the energy storage access power converter is ensured to be in the optimal working state, the stability and the anti-interference capability of the bus voltage are improved, and the method has the advantages of being adaptive, strong in robustness, high in stability, easy to implement and the like.

Description

Technical Field

[0001] The present invention belongs to the technical field of power supply, and in particular relates to an adaptive dynamic partitioning control method and system for an energy storage power supply bus. Background Art

[0002] Energy storage power supply system has significant advantages in many fields. It is crucial to improve the stability and efficiency of power system. Battery energy storage system can balance peak and valley power, reduce grid loss, and provide emergency power in case of sudden power failure to ensure the normal operation of the power grid. Energy storage power supply system can also be combined with distributed energy system to balance energy supply and demand, improve the utilization rate of renewable energy, and improve the power quality of the power grid. Battery energy storage system also plays an important role in promoting green travel and building smart grid and microgrid. In the field of transportation, battery energy storage system improves energy utilization efficiency, promotes the popularization of green travel modes such as electric vehicles, and helps save energy and reduce emissions. At the same time, energy storage power supply system provides necessary auxiliary services for smart grid, such as frequency regulation and demand response, and enhances the safety and efficiency of the power grid. In microgrid or off-grid occasions, battery energy storage system, as the main energy source, provides stable and lasting power support for remote areas or islands, and promotes the construction and development of smart grid and microgrid.

[0003] Power supply systems with battery energy storage often face the problem of power volatility, especially when distributed energy sources (such as photovoltaic power generation and wind power generation) are connected. The volatility and instability of these distributed energy sources will cause bus voltage fluctuations and affect the stability of the system. In order to maintain the stability of the bus voltage, effective control strategies are needed to buffer and suppress the impact of power fluctuations on the bus voltage. However, achieving this goal requires high-precision control algorithms and advanced control technologies. Summary of the invention

[0004] The purpose of the present invention is to propose an adaptive dynamic partitioning control method and system for an energy storage power supply bus in response to the demand for stable control of bus voltage in an energy storage power supply system.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions:

[0006] An adaptive dynamic partition control method for an energy storage power supply bus, comprising the following steps:

[0007] S1: Receive the busbar injection current value I0 and the access node voltage value U0 measured by the current and voltage sensors, and run the electrical parameter perturbation dynamic state estimator A(I x ,U x ), to perform actual dynamic compensation for the busbar injected current I and voltage U;

[0008] S2: Calculate the bus voltage error x1 and the bus injection current x2, construct the bus voltage control multi-state dynamic combination function S(x1,x2), and use the softening factor ρ and switch gain K to design the bus voltage dynamic control law v with nonlinear fast convergence characteristics over a large range of voltage errors;

[0009] S3: Design a characteristic partition controller of the softening factor ρ based on the bus voltage control multi-state dynamic combination function S(x1,x2) and the DC bus voltage error x1, and adaptively optimize ρ according to the adaptive optimization strategy corresponding to the partition;

[0010] S4: Design the characteristic partition controller of the switch gain K based on the bus voltage control multi-state dynamic combination function S(x1,x2) value and the DC bus voltage error x1 and its integral, and adaptively optimize K according to the adaptive optimization strategy corresponding to the partition; S5: Substitute the optimal softening factor ρ and switch gain K selected in S3 and S4 into the bus voltage dynamic control law v in S2 to form a control signal for the switch state of the switch components of the power converter to adjust the power supply bus voltage so that it approaches the rated voltage stability value U of the power supply bus. ref .

[0011] Further, as a preferred technical solution of the present invention, the calculation function A(I x ,U x )for

[0012]

[0013] Among them, I x is the estimated value of busbar injection current, U x is the estimated value of bus voltage, L is the bus inductance, v is the bus voltage dynamic control law, and also represents the switch duty cycle of the power converter components connected, b and h are the constant parameters of the dynamic state estimator.

[0014] Further as a preferred technical solution of the present invention, S2 specifically includes the following steps:

[0015] S2.1, bus voltage error x1 and bus injection current x2 are calculated as:

[0016]

[0017] Among them, U ref is the stable value of the busbar rated voltage, R and C are the busbar resistance and capacitance respectively;

[0018] S2.2, using the bus voltage error x1 and the bus injection current x2, construct the bus voltage control multi-state dynamic combination function S(x1,x2):

[0019]

[0020] Wherein, β>0, q, p are positive odd numbers and p>q; tanh() is the inverse tangent trigonometric function;

[0021] S2.3, using the softening factor ρ and the switch gain K, design the bus voltage dynamic control law v with the nonlinear fast convergence characteristics over a wide range of voltage errors:

[0022]

[0023] Further, as a preferred technical solution of the present invention, S3 specifically includes the following steps:

[0024] S3.1. Divide the input value of S(x1,x2) into 7 partitions S1, S2, S3, S4, S5, S6, S7, and divide the value of ρ into 4 partitions, α1, α2, α3, α4, and 0<α1<α2<α3<α4; that is:

[0025]

[0026] S3.2. Establish a partition-based adaptive optimization strategy to adaptively find different softening factor ρ values ​​in different S(x1,x2) value partitions, so as to achieve the effect of improving stability. The partition-based strategy table is set as follows:

[0027]

[0028] S3.3. Design a feature partitioning strategy for input subsets and output subsets based on Gaussian distribution function, and construct a softening factor feature partitioning state function μ Si (z1):

[0029]

[0030] Among them, e is a natural constant, K1 is a positive constant, α 1i is a seven-digit arithmetic progression symmetrical with zero as the midpoint, b1 is a constant and 0<b1<1, max[] is the maximum value function;

[0031] S3.4. Construct the softening factor output function g(ρ) based on the softening factor characteristic partition function:

[0032]

[0033] S3.5. According to the softening factor output function g(ρ), the real-time value of the softening factor ρ is obtained as

[0034] ρ=ρ min +(ρ max -ρmin )·(g(ρ)-α1) / (α4-α1)

[0035] Among them, min and ρ max are the minimum and maximum values ​​of the softening factor ρ, respectively.

[0036] Further, as a preferred technical solution of the present invention, the S4 specifically includes the following steps:

[0037] S4.1. Calculate the bus voltage error integral minimum objective function f(t):

[0038]

[0039] Among them, η1, η2 and η3 are all weights;

[0040] S4.2. Divide the input value of f(t) into 9 partitions, f1, f2, f3, f4, f5, f6, f7, f8, f9, and divide the value of K into 5 partitions, β1, β2, β3, β4, β5, and 0<β1<β2<β3<β4<β5; that is:

[0041]

[0042] S4.3. Establish a partition control corresponding strategy, and adaptively find different switch gain K values ​​in different f(t) value partitions; when the voltage error integral value is large, a larger switch gain value can be found to improve the response speed; when the voltage error integral value is small, a smaller switch gain value can be found to improve the anti-interference ability; the partition strategy table is set as follows:

[0043]

[0044] S4.4. Design a feature partitioning strategy for input subsets and output subsets based on Gaussian distribution function, and construct the following switch gain feature partitioning state function μ fi (z2):

[0045]

[0046] Where K2 is the gain constant, α 2i is a subset of nine-digit arithmetic progressions starting with 0 and ending with K2, b 2i is a subset of the nine-digit difference sequence and 0<b 2i <(K2 / 9);

[0047] S4.5. Construct a switching gain output function j(K) based on the switching gain characteristic partition function:

[0048]

[0049] S4.6. According to the switch gain output function j(K), the real-time value of the switch gain K is obtained as

[0050] K=K min +(K max -K min )·(j(K)-β1) / (β5-β1)

[0051] Among them, K min and K max They are the minimum and maximum values ​​of the switching gain K respectively.

[0052] An adaptive dynamic partitioning control system for an energy storage power supply bus comprises an energy storage unit, a power supply bus, a current and voltage sensor, a high stability controller, an access power converter and a bus access node. The energy storage unit is connected to the power supply bus via the access power converter. The current and voltage sensor is installed between the access power converter and the bus access node. The output of the current and voltage sensor is connected to the high stability controller. The output of the high stability controller is connected to the control end of the power converter. An adaptive characteristic partitioning control method is run inside the high stability controller.

[0053] The adaptive dynamic partition control method and system of the energy storage power supply bus described in the present invention adopts the above technical solution and has the following technical effects compared with the prior art:

[0054] (1) The electrical parameter perturbation dynamic state estimator is used to analyze the electrical parameter characteristics of the power supply bus voltage. The mathematical model of the power supply system is used to eliminate the random errors caused by unknown interference in the actual measurement of current and voltage sensors, and the data accuracy is improved through compensation.

[0055] (2) The bus voltage dynamic control law v is highly nonlinear and requires few parameters for the bus voltage model. It is insensitive to modeling errors, parameter changes, and external interference. Even if the system is subject to large disturbances or parameter changes, the control law v can maintain the stability and robustness of the system.

[0056] (3) The bus voltage dynamic control law v utilizes the high nonlinear slope characteristics of the inverse tangent function tanh() to enhance the strong adaptive performance of the control law v output as the deviation size is adjusted, so that it exhibits a large slope for large deviations and a small slope for small deviations, thereby more quickly and gently suppressing the bus voltage deviation, making the bus voltage stable and converge quickly.

[0057] (4) The bus voltage dynamic control law v introduces an innovative multi-state dynamic combination function S(x1,x2), softening factor ρ and gain coefficient K, which greatly improves the adaptability and anti-interference ability of the control law, and enhances the orderly nonlinear characteristics of the system and the good control effect.

[0058] (5) Adaptive optimization technology and feature partitioning technology are designed. The g(ρ) and j(K) controllers can perform adaptive control on different control intervals, improving the ability of the energy storage power supply system to adaptively find the optimal softening factor ρ and gain coefficient K in complex and diverse working environments. This improves the ability to operate reliably in complex working environments with diverse external environments and load demands. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic diagram of the structure of the present invention;

[0060] Figure 2 It is a schematic diagram of the adaptive feature partition control method of the present invention;

[0061] The reference numerals are: 1-energy storage unit; 2-power supply bus; 3-current and voltage sensor; 4-high stability controller; 5-access power converter; 6-adaptive characteristic partition control method; 7-bus access node; 601-electrical parameter perturbation dynamic state estimator A(I x ,U x ); 602-bus voltage control multi-state dynamic combination function S(x1,x2); 603-softening factor characteristic partition state function μ Si (z1); 604-softening factor output function g(ρ); 605-bus voltage error integral minimum objective function f(t); 606-switching gain characteristic partition state function μ fi (z2); 607-switching gain output function j(K); 608-bus voltage dynamic control law v. DETAILED DESCRIPTION

[0062] The present invention is further explained below in detail with reference to the accompanying drawings so that those skilled in the art can more deeply understand the present invention and be able to implement it. However, the following reference examples are only used to explain the present invention and are not intended to limit the present invention.

[0063] See attached Figure 1 , an adaptive dynamic partitioning control system for an energy storage power supply bus, comprising an energy storage unit 1, a power supply bus 2, a current and voltage sensor 3, a high stability controller 4, an access power converter 5, an adaptive characteristic partitioning control method 6, and a bus access node 7; the energy storage unit 1 is connected to the power supply bus 2 through the access power converter 5, the current and voltage sensor 3 is installed between the access power converter 5 and the bus access node 7, the output of the current and voltage sensor 3 is connected to the high stability controller 4, the output of the high stability controller 4 is connected to the control end of the power converter 5, and the adaptive characteristic partitioning control method 6 runs inside the high stability controller 4.

[0064] See attached Figure 2 The adaptive characteristic partition control method 6 includes: an electric parameter perturbation dynamic state estimator A(I x ,U x )601、Multi-state dynamic combination function of bus voltage control S(x1,x2)602、Softening factor characteristic partition state function μ Si (z1) 603, softening factor output function g(ρ) 604, bus voltage error integral minimum objective function f(t) 605, switch gain characteristic partition state function μ fi (z2) 606, switch gain output function j(K) 607, bus voltage dynamic control law v 608.

[0065] An adaptive dynamic partition control method for an energy storage power supply bus, comprising the following steps:

[0066] S1: Receive the busbar injection current value I0 and the access node voltage value U0 measured by the current and voltage sensors, and run the electrical parameter perturbation dynamic state estimator A(I x ,U x ), to perform actual dynamic compensation for the busbar injected current I and voltage U;

[0067] S2: Calculate the bus voltage error x1 and the bus injection current x2, construct the bus voltage control multi-state dynamic combination function S(x1,x2), and use the softening factor ρ and switch gain K to design the bus voltage dynamic control law v with nonlinear fast convergence characteristics over a large range of voltage errors;

[0068] S3: Design a characteristic partition controller of the softening factor ρ based on the bus voltage control multi-state dynamic combination function S(x1,x2) and the DC bus voltage error x1, and adaptively optimize ρ according to the adaptive optimization strategy corresponding to the partition;

[0069] S4: Design the characteristic partition controller of the switch gain K based on the bus voltage control multi-state dynamic combination function S(x1,x2) value and the DC bus voltage error x1 and its integral, and adaptively optimize K according to the adaptive optimization strategy corresponding to the partition; S5: Substitute the optimal softening factor ρ and switch gain K selected in S3 and S4 into the bus voltage dynamic control law v in S2 to form a control signal for the switch state of the switch components of the power converter to adjust the power supply bus voltage so that it approaches the rated voltage stability value U of the power supply bus. ref .

[0070] The calculation function A(I x ,U x )for

[0071]

[0072] Among them, I x is the estimated value of busbar injection current, U x is the estimated value of bus voltage, L is the bus inductance, v is the bus voltage dynamic control law, and also represents the switch duty cycle of the power converter components connected, b and h are the constant parameters of the dynamic state estimator.

[0073] S2 specifically includes the following steps:

[0074] S2.1, bus voltage error x1 and bus injection current x2 are calculated as:

[0075]

[0076] Among them, U ref is the stable value of the busbar rated voltage, R and C are the busbar resistance and capacitance respectively;

[0077] According to the circuit principle that the energy storage unit is connected to the power supply bus through the access power converter, the change rates of the bus voltage error x1 and the bus injection current x2 can be calculated as follows:

[0078]

[0079] S2.2, using the bus voltage error x1 and the bus injection current x2, construct the bus voltage control multi-state dynamic combination function S(x1,x2):

[0080]

[0081] Wherein, β>0, q, p are positive odd numbers and p>q; tanh() is the inverse tangent trigonometric function;

[0082] The value of the multi-state dynamic combination function S(x1,x2) represents the degree to which the bus voltage error x1 and the bus injection current x2 are away from the equilibrium zero point. That is to say, when the bus voltage error x1=0 and the bus injection current x2=0, S(x1,x2)=0, which means that the bus voltage has stabilized and reached the rated voltage stability value U ref , the connected power converter stops injecting current into the bus, and the entire system reaches a balanced zero point.

[0083] In order to make the multi-state dynamic combination function S(x1,x2) converge to the equilibrium zero point quickly and orderly, the convergence rate trajectory is set as:

[0084]

[0085] Where ρ is the softening factor, K is the switching gain, ρ>0, K>0;

[0086] S2.3, using the softening factor ρ and the switch gain K, design the bus voltage dynamic control law v with the nonlinear fast convergence characteristics over a wide range of voltage errors:

[0087]

[0088] S3 specifically includes the following steps:

[0089] S3.1. Divide the input value of S(x1,x2) into 7 partitions S1, S2, S3, S4, S5, S6, S7, and divide the value of ρ into 4 partitions, α1, α2, α3, α4, and 0<α1<α2<α3<α4; that is:

[0090]

[0091] S3.2. Establish a partition-based adaptive optimization strategy to adaptively find different softening factor ρ values ​​in different S(x1,x2) value partitions, so as to achieve the effect of improving stability. The partition-based strategy table is set as follows:

[0092]

[0093] S3.3. Design a feature partitioning strategy for input subsets and output subsets based on Gaussian distribution function, and construct a softening factor feature partitioning state function μ Si (z1):

[0094]

[0095] Among them, e is a natural constant, K1 is a positive constant, α 1i is a seven-digit arithmetic progression symmetrical with zero as the midpoint, b1 is a constant and 0<b1<1, max[] is the maximum value function;

[0096] S3.4. Construct the softening factor output function g(ρ) based on the softening factor characteristic partition function:

[0097]

[0098] S3.5. According to the softening factor output function g(ρ), the real-time value of the softening factor ρ is obtained as

[0099] ρ=ρ min +(ρ max -ρ min )·(g(ρ)-α1) / (α4-α1)

[0100] Among them, min and ρ max are the minimum and maximum values ​​of the softening factor ρ, respectively.

[0101] S4 specifically includes the following steps:

[0102] S4.1. Calculate the bus voltage error integral minimum objective function f(t):

[0103]

[0104] Among them, η1, η2 and η3 are all weights;

[0105] S4.2. Divide the input value of f(t) into 9 partitions, f1, f2, f3, f4, f5, f6, f7, f8, f9, and divide the value of K into 5 partitions, β1, β2, β3, β4, β5, and 0<β1<β2<β3<β4<β5; that is:

[0106]

[0107] S4.3. Establish a partition control corresponding strategy, and adaptively find different switch gain K values ​​in different f(t) value partitions; when the voltage error integral value is large, a larger switch gain value can be found to improve the response speed; when the voltage error integral value is small, a smaller switch gain value can be found to improve the anti-interference ability; the partition strategy table is set as follows:

[0108]

[0109] S4.4. Design a feature partitioning strategy for input subsets and output subsets based on Gaussian distribution function, and construct the following switch gain feature partitioning state function μ fi (z2):

[0110]

[0111] Where K2 is the gain constant, α 2i is a subset of nine-digit arithmetic progressions starting with 0 and ending with K2, b 2i is a subset of the nine-digit difference sequence and 0<b 2i <(K2 / 9);

[0112] S4.5. Construct a switching gain output function j(K) based on the switching gain characteristic partition function:

[0113]

[0114] S4.6. According to the switch gain output function j(K), the real-time value of the switch gain K is obtained as

[0115] K=K min +(K max -K min)·(j(K)-β1) / (β5-β1)

[0116] Among them, K min and K max They are the minimum and maximum values ​​of the switching gain K respectively.

[0117] According to the characteristic partition control method, the present invention adaptively optimizes the softening factor ρ and the switch gain K in the bus voltage control law v, so that the high-stability controller outputs the dynamically optimal control value v, ensuring that the switch element of the connected power converter is in the optimal working state when providing the energy storage unit electric energy to the power supply bus, thereby improving the stability and anti-interference ability of the bus voltage.

[0118] The specific implementation scheme described above further describes in detail the purpose, technical scheme and beneficial effects of the present invention. It should be understood that the above is only a specific implementation scheme of the present invention and is not intended to limit the scope of the present invention. Any equivalent changes and modifications made by any technician in the field without departing from the concept and principle of the present invention should fall within the scope of protection of the present invention.

Claims

1. An adaptive dynamic partition control method for an energy storage power supply bus, characterized in that: The following steps are involved: S1: Receive the busbar injection current value I0 and the access node voltage value U0 measured by the current and voltage sensors, and run the electrical parameter perturbation dynamic state estimator A(I x ,U x ), to perform actual dynamic compensation for the busbar injected current I and voltage U; S2: Calculate the bus voltage error x1 and the bus injection current x2, construct the bus voltage control multi-state dynamic combination function S(x1,x2), and use the softening factor ρ and switch gain K to design the bus voltage dynamic control law v with nonlinear fast convergence characteristics over a large range of voltage errors; S3: Design a characteristic partition controller of the softening factor ρ based on the bus voltage control multi-state dynamic combination function S(x1,x2) and the DC bus voltage error x1, and adaptively optimize ρ according to the adaptive optimization strategy corresponding to the partition; S4: Design a characteristic partition controller of the switch gain K based on the bus voltage control multi-state dynamic combination function S(x1,x2) value and the DC bus voltage error x1 and its integral, and adaptively optimize K according to the adaptive optimization strategy corresponding to the partition; S5: The optimal softening factor ρ and switch gain K selected in S3 and S4 are introduced into the bus voltage dynamic control law v in S2 to form a control signal for the switch state of the switch components of the power converter to adjust the power supply bus voltage to make it close to the rated voltage stability value U of the power supply bus. ref .

2. The adaptive dynamic partition control method of the energy storage power supply bus according to claim 1 is characterized in that: The calculation function A(I x ,U x )for Among them, I x is the estimated value of busbar injection current, U x is the estimated value of bus voltage, L is the bus inductance, v is the bus voltage dynamic control law, and also represents the switch duty cycle of the power converter components connected, b and h are the constant parameters of the dynamic state estimator.

3. The adaptive dynamic partition control method of the energy storage power supply bus according to claim 2 is characterized in that: The S2 specifically includes the following steps: S2.1, bus voltage error x1 and bus injection current x2 are calculated as: Among them, U ref is the stable value of the busbar rated voltage, R and C are the busbar resistance and capacitance respectively; S2.2, using the bus voltage error x1 and the bus injection current x2, construct the bus voltage control multi-state dynamic combination function S(x1,x2): Wherein, β>0, q, p are positive odd numbers and p>q; tanh() is the inverse tangent trigonometric function; S2.3, using the softening factor ρ and the switch gain K, design the bus voltage dynamic control law v with the nonlinear fast convergence characteristics over a wide range of voltage errors:

4. The adaptive dynamic partition control method of the energy storage power supply bus according to claim 3 is characterized in that: The S3 specifically includes the following steps: S3.

1. Divide the input value of S(x1,x2) into 7 partitions S1, S2, S3, S4, S5, S6, S7, and divide the value of ρ into 4 partitions, α1, α2, α3, α4, and 0<α1<α2<α3<α4; that is: S3.

2. Establish a partition-based adaptive optimization strategy to adaptively find different softening factor ρ values ​​in different S(x1,x2) value partitions, so as to achieve the effect of improving stability. The partition-based strategy table is set as follows: S3.

3. Design a feature partitioning strategy for input subsets and output subsets based on Gaussian distribution function, and construct a softening factor feature partitioning state function μ Si (z1): Among them, e is a natural constant, K1 is a positive constant, α 1i is a seven-digit arithmetic progression symmetrical with zero as the midpoint, b1 is a constant and 0<b1<1, max[] is the maximum value function; S3.

4. Construct the softening factor output function g(ρ) based on the softening factor characteristic partition function: S3.

5. According to the softening factor output function g(ρ), the real-time value of the softening factor ρ is obtained as p=p min +(r max -r min )·(g(ρ)-α1) / (α4-α1) Among them, min and ρ max are the minimum and maximum values ​​of the softening factor ρ, respectively.

5. The adaptive dynamic partition control method of the energy storage power supply bus according to claim 4 is characterized in that: The S4 specifically comprises the following steps: S4.

1. Calculate the bus voltage error integral minimum objective function f(t): Among them, η1, η2 and η3 are all weights; S4.

2. Divide the input value of f(t) into 9 partitions, f1, f2, f3, f4, f5, f6, f7, f8, f9, and divide the value of K into 5 partitions, β1, β2, β3, β4, β5, and 0<β1<β2<β3<β4<β5; that is: S4.

3. Establish a partition control corresponding strategy, and adaptively find different switch gain K values ​​in different f(t) value partitions; when the voltage error integral value is large, a larger switch gain value can be found to improve the response speed; when the voltage error integral value is small, a smaller switch gain value can be found to improve the anti-interference ability; the partition strategy table is set as follows: S4.

4. Design a feature partitioning strategy for input subsets and output subsets based on Gaussian distribution function, and construct the following switch gain feature partitioning state function μ fi (z2): Where K2 is the gain constant, α 2i is a subset of nine-digit arithmetic progressions starting with 0 and ending with K2, b 2i is a subset of the nine-digit difference sequence and 0<b 2i <(K2 / 9); S4.

5. Construct a switching gain output function j(K) based on the switching gain characteristic partition function: S4.

6. According to the switch gain output function j(K), the real-time value of the switch gain K is obtained as K=K min +(K max -K min )·(j(K)-β1) / (β5-β1) Among them, K min and K max They are the minimum and maximum values ​​of the switching gain K respectively.

6. A control system based on an adaptive dynamic partitioning control method of an energy storage power supply bus according to any one of claims 1 to 5, comprising an energy storage unit, a power supply bus, a current and voltage sensor, a high stability controller, an access power converter and a bus access node, characterized in that: The energy storage unit is connected to the power supply bus through an access power converter. The current and voltage sensors are installed between the access power converter and the bus access node. The output of the current and voltage sensors is connected to a high stability controller. The output of the high stability controller is connected to the control end of the power converter. The high stability controller runs an adaptive characteristic partitioning control method inside.

Images