CN113949096A - Energy storage system controller design method based on reverse model predictive control - Google Patents

Energy storage system controller design method based on reverse model predictive control Download PDF

Info

Publication number
CN113949096A
CN113949096A CN202111192134.4A CN202111192134A CN113949096A CN 113949096 A CN113949096 A CN 113949096A CN 202111192134 A CN202111192134 A CN 202111192134A CN 113949096 A CN113949096 A CN 113949096A
Authority
CN
China
Prior art keywords
current
voltage
value
formula
model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111192134.4A
Other languages
Chinese (zh)
Other versions
CN113949096B (en
Inventor
张自伟
程振华
伏祥运
于跃
韩伟
岳付昌
李响
谭恒兵
黄淮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Donghai Power Supply Branch Of State Grid Jiangsu Electric Power Co ltd
Lianyungang Power Supply Co of State Grid Jiangsu Electric Power Co Ltd
Original Assignee
Donghai Power Supply Branch Of State Grid Jiangsu Electric Power Co ltd
Lianyungang Power Supply Co of State Grid Jiangsu Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Donghai Power Supply Branch Of State Grid Jiangsu Electric Power Co ltd, Lianyungang Power Supply Co of State Grid Jiangsu Electric Power Co Ltd filed Critical Donghai Power Supply Branch Of State Grid Jiangsu Electric Power Co ltd
Priority to CN202111192134.4A priority Critical patent/CN113949096B/en
Publication of CN113949096A publication Critical patent/CN113949096A/en
Application granted granted Critical
Publication of CN113949096B publication Critical patent/CN113949096B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/28—Arrangements for balancing of the load in a network by storage of energy
    • H02J3/32—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Inverter Devices (AREA)
  • Control Of Ac Motors In General (AREA)

Abstract

The application provides an energy storage system controller design method based on backstepping model predictive control, which is characterized by comprising the following steps: establishing an energy storage grid-connected inverter mathematical model; utilizing backward control to take the virtual control quantity as a reference current to be fed into the inner ring controller; and optimizing the prediction model. The method provided by the application directly takes the virtual control quantity as the reference current to be sent to the inner ring controller, so that not only is multiple derivation on the virtual control quantity avoided, but also the complexity of controller design is reduced.

Description

Energy storage system controller design method based on reverse model predictive control
Technical Field
The application relates to the technical field of energy storage controllers of power systems, in particular to an energy storage system controller design method based on reverse model predictive control.
Background
With the rapid development of the economy of China and the deep revolution of the urban power supply and utilization form, the distributed energy-friendly access with high reliability power supply and high permeability puts higher requirements on the safe and reliable operation of a power grid. However, because a large amount of electric energy cannot be stored, the traditional power grid operation mode is a so-called 'instant generation and instant use' state, the generated energy and the power consumption need to be kept in a dynamic balance, and when disturbance occurs in a system, power imbalance is caused, so that the safe and stable operation of a power grid is influenced under severe conditions. The stored energy plays a role in lightening and weighting each part of the modern power grid such as power generation, power transmission and power distribution because of the characteristics of electric energy storage, flexible configuration, peak clipping and valley filling and the like. The capacity of responding to the power grid requirement can be effectively improved through the rapid and reliable control of the energy storage converter, and the reliability and flexibility of the power grid are enhanced. Therefore, research on energy storage control technology is being focused on by more and more scholars.
Maintaining the voltage stability of the direct current side of the energy storage system is a precondition for stable power transmission. Therefore, improving voltage stability and dynamic responsiveness is a major direction for complex energy storage system controller optimization. Research shows that the double-closed-loop PI control is adopted to control the DC/AC converter, system PQ decoupling control is achieved, control is simple and convenient to achieve, and dynamic response speed and control accuracy need to be improved. The model prediction control based on the grid-connected inverter is provided by existing research, so that the complexity of control strategy design can be effectively reduced, and the problems of low system dynamic response speed and the like can be solved. And an inner loop current controller is designed based on model prediction, so that the dynamic response speed of the system is improved, but the problems of integral saturation, low voltage stability and the like of an outer loop PI controller are not considered. In order to avoid integral saturation, a controller is designed by adopting reverse control, so that the problem of large voltage fluctuation of a direct current side is effectively solved, but the control operation amount is increased by carrying out multiple derivation on the virtual control amount. The inductance value of the inverter is always set to be a fixed value in modeling simulation analysis, and when a large current flows through the inductor, the inductance value changes, and the fixed inductance has influence on the control precision of the system.
Although the above methods all solve the problems in some aspects to a certain extent, the problems of large direct-current voltage fluctuation and low dynamic response speed of the energy storage inverter have not been effectively solved, the problems of large direct-current voltage fluctuation and low dynamic response speed of the energy storage inverter have been solved to a certain extent in time, and other problems such as too large operation amount are introduced at the same time.
Disclosure of Invention
The application provides an energy storage system controller design method based on reverse model predictive control, which can be used for solving the technical problems of large direct-current voltage fluctuation and low dynamic response speed of an energy storage inverter.
The application provides an energy storage system controller design method based on reverse model predictive control, which comprises the following steps:
establishing an energy storage grid-connected inverter mathematical model;
utilizing backward control to take the virtual control quantity as a reference current to be fed into the inner ring controller;
and optimizing the prediction model.
Optionally, the establishing of the energy storage grid-connected inverter mathematical model includes:
Figure BDA0003301608170000021
in the formula (1), R is equivalent resistance, UskFor the grid side voltage, ikFor grid side current, UrkThe voltage is the AC side voltage of the converter; wherein k is a, b, cA, b and c respectively represent three phases of voltage, and L is inductance;
and (3) carrying out dq coordinate transformation on the formula (1) to obtain a mathematical model of the converter under a dq coordinate system as follows:
Figure BDA0003301608170000022
in the formula (2), iqIs the q-axis component of the current, idBeing d-axis component of current, UsdBeing d-axis component of voltage, UsqBeing the q-axis component of the voltage, UrdIs the d-axis voltage value, U, of the AC side of the converterrqThe q-axis voltage value of the alternating current side of the converter;
u under steady statesqWhen the resistance value R is smaller, the loss can be ignored, and the active power and the reactive power absorbed by the converter from the alternating current side are as follows:
Figure BDA0003301608170000023
in formula (3), P is active power, and Q is no power;
the power conservation formula of the grid-connected inverter under the condition of neglecting loss is as follows:
Figure BDA0003301608170000024
in formula (4), UdcIs the DC side bus voltage, and C is the DC side capacitance.
Optionally, the feeding the virtual control amount as a reference current into the inner-loop controller by using a reverse control includes:
introducing reverse control in the outer loop control to obtain a current reference value;
and utilizing the current reference value, and performing predictive control on the inner loop controller through a predictive model.
Optionally, introducing a back-stepping control in the outer loop control to obtain a current reference value, including:
defined as the dc voltage tracking error:
z1＝Udc-Udcref (5)
in the formula (5), z1For DC voltage tracking errors, UdcIs the actual value of the DC voltage, UdcrefIs a direct current voltage reference value;
the inverse of the dc voltage tracking error is:
Figure BDA0003301608170000031
in the formula (6), the first and second groups,
Figure BDA0003301608170000032
is the derivative of the dc voltage tracking error,
Figure BDA0003301608170000033
is the derivative of the dc voltage reference;
setting the positive Lyapunov function can be expressed as:
Figure BDA0003301608170000034
in formula (7), V1 is a positive definite Lyapunov function;
derivation is performed on equation (7) to obtain:
Figure BDA0003301608170000035
in the formula (8), idrefIs a current reference value;
processing the formula (8) to obtain the current reference value;
Figure BDA0003301608170000036
in equation (9), the current reference value is used as a reference value of the d-axis current of the inner ring.
Optionally, using the current reference value, the inner loop controller performs predictive control through a predictive model, including:
the discrete model is:
Figure BDA0003301608170000037
Figure BDA0003301608170000038
in the formulas (10) and (11), TsIs the sampling period; u shaperdq(k-1) is a voltage sampling value at the k-1 moment of the AC side of the converter; u shapesdq(k) Is the grid side voltage at time k, idq(k) Is the current sample value at time k; i.e. idq(k +1) is the predicted value of the current at the moment k + 1; delta Urdq(k) The output voltage increment of the converter at the moment k, omega is angular frequency, Iq is a q-axis component of current, Urq is a q-axis voltage value of the alternating current side of the converter, and Urd is a d-axis voltage value of the alternating current side of the converter;
the prediction model is a neural network model;
the output expression of the neural network model is as follows:
Figure BDA0003301608170000039
in the formula (12), wiIs weight of neural network, xiFor the network input signal, corresponding to the current sample value at time k, o (w)i,xi) An output stimulus function for the network;
the weight value adjusting algorithm of the least mean square algorithm of the neural network model comprises the following steps:
wi(k+1)＝wi(k)+2ηxi(d(k)-O(k)) (13)
in the formula (13), wi(k +1) is the weight of the neural network model least mean square algorithm at the moment k +1, wi(k) And the weight value of the neural network model least mean square algorithm at the moment k is obtained.
Optionally, optimizing the prediction model includes:
the current prediction error is defined as:
Figure BDA0003301608170000041
in the formula (15), h1Is one of the error correction coefficients, h2Is a second error correction coefficient idq(k) Is the sampled value of the current of the converter at the time k,
Figure BDA0003301608170000042
predicting the current value at k time after compensation;
the output of the corrected prediction model is:
Figure BDA0003301608170000043
performing rolling optimization to enable the difference value between the current predicted value and the current reference value to be the target of the rolling optimization, and controlling the target function as follows:
Figure BDA0003301608170000044
in formula (17), λ1,2Is the weight coefficient, ξ, of the current error in the objective function1,2Is a weight coefficient, i, of the incremental error of the control voltage in the target functiondqrefIs the dq-axis reference current.
The method provided by the application aims at the problems of large direct-current voltage fluctuation, low dynamic response speed and the like of the energy storage inverter and designs the outer loop reverse-thrust inner loop model prediction controller. Firstly, an energy storage grid-connected inverter mathematical model is established. Then, aiming at the problems of large voltage fluctuation, difficult setting of PI parameters and the like in the traditional outer ring PI control, reverse control is applied to replace the traditional outer ring PI control, and the virtual control quantity is directly taken as a reference current to be fed into an inner ring controller, so that not only is the multiple derivation of the virtual control quantity avoided, but also the design complexity of the controller is reduced; secondly, model predictive control is introduced into the inner loop controller aiming at the problem of low dynamic response speed of the traditional PI inner loop control, and uncertain inductance parameters are identified on line by adopting a neural network algorithm, so that the model predictive control precision is improved.
Drawings
Fig. 1 is a schematic flow chart of a method for designing an energy storage system controller based on a reverse model predictive control according to an embodiment of the present disclosure;
fig. 2 is a topological diagram of a grid-connected structure of an energy storage converter according to an embodiment of the present application;
fig. 3 is a schematic block diagram of a design method of an energy storage system controller based on a reverse model predictive control according to an embodiment of the present application;
FIG. 4 is a block diagram of an outer loop back-stepping control provided in an embodiment of the present application;
fig. 5 is a block diagram of an inner loop model predictive control provided in an embodiment of the present application;
fig. 6 is a schematic diagram of bus voltages under a prediction model and a reverse-thrust control according to an embodiment of the present application;
FIG. 7 is a schematic diagram illustrating a bus voltage comparison according to an embodiment of the present disclosure;
fig. 8 is a schematic diagram of an inductance under neural network parameter identification according to an embodiment of the present disclosure;
fig. 9 is a second comparative schematic diagram of bus voltage provided in the embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
First, referring to fig. 1, a schematic flow chart of a method for designing an energy storage system controller based on a reverse model predictive control according to an embodiment of the present application is shown.
Please refer to fig. 2, which is a topological diagram of a grid-connected structure of an energy storage converter according to an embodiment of the present application. The energy storage converter grid connection mainly comprises a storage battery and a grid connection converter. The storage battery provides energy, and the grid-connected converter is responsible for transmitting the energy.
Fig. 3 is a schematic block diagram of a method for designing an energy storage system controller based on a reverse-thrust model predictive control according to an embodiment of the present application.
The method provided by the application comprises the following steps:
and S101, establishing an energy storage grid-connected inverter mathematical model.
According to kirchhoff's law, a mathematical model of the energy storage grid-connected inverter is as follows:
Figure BDA0003301608170000051
in the formula (1), R is equivalent resistance, UskFor the grid side voltage, ikFor grid side current, UrkIs the AC side voltage of the converter. Where k denotes a, b, c, a, b, and c denote three phases of voltage, and L denotes an inductor.
And (3) carrying out dq coordinate transformation on the formula (1) to obtain a mathematical model of the converter under a dq coordinate system as follows:
Figure BDA0003301608170000052
in the formula (2), iqIs the q-axis component of the current, idBeing d-axis component of current, UsdBeing d-axis component of voltage, UsqBeing the q-axis component of the voltage, UrdIs the d-axis voltage value, U, of the AC side of the converterrqIs the value of the q-axis voltage on the alternating current side of the converter.
U under steady statesqWhen the resistance value R is smaller, the loss can be ignored, and the active power and the reactive power absorbed by the converter from the alternating current side are as follows:
Figure BDA0003301608170000061
in formula (3), P is active power and Q is no power.
The power conservation formula of the grid-connected inverter under the condition of neglecting loss is as follows:
Figure BDA0003301608170000062
in formula (4), UdcIs the DC side bus voltage, and C is the DC side capacitance.
And step S102, utilizing backward control to take the virtual control quantity as a reference current to be sent to the inner ring controller.
In order to better realize the stability of the voltage at the direct current side of the energy storage system, a reverse control is introduced into the outer loop controller, and because the outer loop controller adopts a reverse control design, the design method is the same and the design steps are similar, the design description is only carried out by taking a constant voltage charging controller as an example. In order to improve the dynamic responsiveness of the system and quickly track the instruction power, the inner loop controller is controlled by a prediction model.
Fig. 4 is a block diagram of an outer loop back-stepping control provided in an embodiment of the present application.
Specifically, step S102 includes:
and introducing reverse control in the outer loop control to obtain a current reference value.
In the embodiment of the present application, the tracking error is defined as a dc voltage tracking error:
z1＝Udc-Udcref (5)
in the formula (5), z1For DC voltage tracking errors, UdcIs the actual value of the DC voltage, UdcrefIs a dc voltage reference.
The inverse of the dc voltage tracking error is:
Figure BDA0003301608170000063
in the formula (6), the first and second groups,
Figure BDA0003301608170000064
is the derivative of the dc voltage tracking error,
Figure BDA0003301608170000065
is the derivative of the dc voltage reference.
Setting the positive Lyapunov function can be expressed as:
Figure BDA0003301608170000066
in equation (7), V1 is a positive definite Lyapunov function.
Derivation is performed on equation (7) to obtain:
Figure BDA0003301608170000067
in the formula (8), idrefIs a current reference value.
And (4) processing the formula (8) to obtain a current reference value.
Figure BDA0003301608170000071
In equation (9), the current reference value is used as a reference value for the d-axis current of the inner ring.
Step S102 further includes: the inner loop controller predicts control by a predictive model using the current reference value.
Fig. 5 is a block diagram of an inner-loop model predictive control according to an embodiment of the present application.
Specifically, the formula (2) is discretized, the first-order euler method processing is performed on the derivative of the current in the formula, and after the equation is rearranged, the following discrete model is obtained:
the discrete model is:
Figure BDA0003301608170000072
Figure BDA0003301608170000073
in the formulas (10) and (11), TsIs the sampling period. U shaperdqAnd (k-1) is a voltage sampling value at the time k-1 on the alternating current side of the converter. U shapesdq(k) Is the grid side voltage at time k, idq(k) Is the current sample at time k. i.e. idq(k +1) is the predicted value of the current at the time k + 1. Delta Urdq(k) And the increment of the output voltage of the converter at the moment k, omega is angular frequency, Iq is a q-axis component of current, Urq is a q-axis voltage value of the alternating current side of the converter, and Urd is a d-axis voltage value of the alternating current side of the converter.
The prediction model is a neural network model.
It should be noted that, in the current research analysis, it is generally assumed that the inductance L is a constant. However, in actual operation, when a large current flows through the discrete model, the inductance saturation phenomenon is easily caused, so that the inductance value changes, and the prediction accuracy of the discrete model is directly influenced. In order to obtain the accurate value of the inductor in actual operation, the inductor in the model is subjected to online parameter identification by adopting a neural network algorithm, so that the model prediction precision is improved.
The output expression of the neural network model is as follows:
Figure BDA0003301608170000074
in the formula (12), wiIs weight of neural network, xiFor the network input signal, corresponding to the current sample value at time k, o (w)i,xi) The function is excited for the output of the network.
The weight value adjusting algorithm of the least mean square algorithm of the neural network model comprises the following steps:
wi(k+1)＝wi(k)+2ηxi(d(k)-O(k)) (13)
in the formula (13), wi(k +1) is the weight of the neural network model least mean square algorithm at the moment k +1, wi(k) And the weight value of the neural network model least mean square algorithm at the moment k is obtained.
Step S103, optimizing the prediction model.
Specifically, the current prediction error is defined as:
Figure BDA0003301608170000081
in the formula (15), h1Is one of the error correction coefficients, h2Is a second error correction coefficient idq(k) Is the sampled value of the current of the converter at the time k,
Figure BDA0003301608170000082
the predicted current value at time k after compensation.
The output of the corrected prediction model is:
Figure BDA0003301608170000083
performing rolling optimization to enable the difference value between the current predicted value and the current reference value to be the target of the rolling optimization, and controlling the target function as follows:
Figure BDA0003301608170000084
in formula (17), λ1,2Is the weight coefficient, ξ, of the current error in the objective function1,2Is a weight coefficient, i, of the incremental error of the control voltage in the target functiondqrefIs the dq-axis reference current.
And performing partial derivation on the target function to obtain the optimal control quantity.
In order to further illustrate the examples provided herein, reference will now be made to specific examples. As shown in table 1, are simulation parameters.
Table 1: simulation parameters
Figure BDA0003301608170000085
Table 2 shows comparison of effects under various control methods provided in the present application.
Table 2: comparison of control effects of multiple control methods
Figure BDA0003301608170000086
To verify the effect and dynamic response performance of the designed controller in suppressing the voltage fluctuation of the direct current bus, the active power required by the power grid is suddenly reduced from 0.5MW to-0.2 MW at the time of setting 0.4 s. The active power value increased from-0.2 MW to 0.3MW at 0.72s, at 0.4s and 0.72s, respectively.
Fig. 6 is a schematic diagram of bus voltage under a prediction model and a reverse-thrust control according to an embodiment of the present application. Fig. 7 is a schematic diagram of bus voltage comparison provided in the embodiment of the present application.
As can be seen from fig. 6 and 7, under the condition of small disturbance, the double closed loop PI controller has larger overshoot and undershoot of 32V and 12V, respectively. The MPC controller overshoot and undershoot were, in turn, 28V and 18V. The overshoot of the reverse push controller is 18kV, and the undershoot is 17V. The overshoot and undershoot of the designed controller are 10V and 7V, the direct-current voltage stabilizing capacity of the controller is strongest, and the robustness of the controller is optimal. And the minimum dynamic response time is about 0.01s, compared with the double closed-loop PI controller 0.07s, the MPC controller 0.05s and the reverse push controller 0.06s are all small, the designed controller has the optimal response speed. Therefore, the robustness and the dynamic response performance of the designed controller under the condition that the active power required by the power grid is changed are better than those of the other three controllers.
In order to verify the effect of induction parameter identification on the voltage stabilization of the direct current bus in the inner loop model predictive control of the designed controller, the condition that the voltage of the direct current bus fluctuates under the condition of parameter identification and the condition of no parameter identification is verified in a simulation mode. The simulation results are shown in fig. 8 and 9.
FIG. 9 illustrates DC bus voltage fluctuations with and without parameter identification for the inverse-model predictive control. It can be seen from the figure that the overshoot of the parameter-identified back-model predictive controller is 10V, while the overshoot of the parameter-identified back-model predictive controller is 22V, and the response time is 0.03s longer than the response time of the parameter-identified back-model predictive controller, which is 0.01 s. Therefore, the induction of inductance parameter neural network identification in the inner loop model predictive control can improve the system response time and inhibit the fluctuation of direct current voltage.
The outer ring reverse-thrust inner ring model prediction controller is designed aiming at the problems that the direct-current voltage fluctuation of the energy storage inverter is large, the dynamic response speed is low and the like. Firstly, an energy storage grid-connected inverter mathematical model is established. Then, aiming at the problems of large voltage fluctuation, difficult setting of PI parameters and the like existing in the traditional outer ring PI control, the traditional outer ring PI control is replaced by the reverse control, and the virtual control quantity is directly taken as the reference current to be fed into the inner ring controller, so that not only is the multiple derivation of the virtual control quantity avoided, but also the complexity of the design of the controller is reduced. Secondly, model predictive control is introduced into an inner loop controller aiming at the problem of low dynamic response speed of the traditional PI inner loop control, and uncertain inductance parameters are identified on line by adopting a neural network algorithm, so that the model predictive control precision is improved.
Those skilled in the art will clearly understand that the techniques of the embodiments of the present application may be implemented in software plus a required general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present application may be essentially or partially implemented in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present application.
The same and similar parts in the various embodiments in this specification may be referred to each other. In particular, for the embodiments of the service construction apparatus and the service loading apparatus, since they are substantially similar to the embodiments of the method, the description is simple, and the relevant points can be referred to the description in the embodiments of the method.
The above-described embodiments of the present application do not limit the scope of the present application.

Claims (6)

1. A method for designing an energy storage system controller based on a reverse model predictive control is characterized by comprising the following steps:
establishing an energy storage grid-connected inverter mathematical model;
utilizing backward control to take the virtual control quantity as a reference current to be fed into the inner ring controller;
and optimizing the prediction model.
2. The method of claim 1, wherein building a mathematical model of an energy storage grid-connected inverter comprises:
Figure FDA0003301608160000011
in the formula (1), R is equivalent resistance, UskFor the grid side voltage, ikFor grid side current, UrkIs the AC side voltage of the converter; wherein k is a, b, c, a, b, c respectively represent three phases of voltage, and L is inductance;
and (3) carrying out dq coordinate transformation on the formula (1) to obtain a mathematical model of the converter under a dq coordinate system as follows:
Figure FDA0003301608160000012
in the formula (2), iqIs the q-axis component of the current, idBeing d-axis component of current, UsdBeing d-axis component of voltage, UsqBeing the q-axis component of the voltage, UrdIs the d-axis voltage value, U, of the AC side of the converterrqThe q-axis voltage value of the alternating current side of the converter;
u under steady statesqWhen the resistance value R is smaller, the loss can be ignored, and the active power and the reactive power absorbed by the converter from the alternating current side are as follows:
Figure FDA0003301608160000013
in formula (3), P is active power, and Q is no power;
the power conservation formula of the grid-connected inverter under the condition of neglecting loss is as follows:
Figure FDA0003301608160000014
in formula (4), UdcIs the DC side bus voltage, and C is the DC side capacitance.
3. The method of claim 1, wherein feeding the virtual control quantity as a reference current to the inner-loop controller using a back-stepping control comprises:
introducing reverse control in the outer loop control to obtain a current reference value;
and utilizing the current reference value, and performing predictive control on the inner loop controller through a predictive model.
4. The method of claim 3, wherein introducing a back-stepping control in the outer loop control to obtain the current reference comprises:
defined as the dc voltage tracking error:
z1＝Udc-Udcref (5)
in the formula (5), z1For DC voltage tracking errors, UdcIs the actual value of the DC voltage, UdcrefIs a direct current voltage reference value;
the inverse of the dc voltage tracking error is:
Figure FDA0003301608160000021
in the formula (6), the first and second groups,
Figure FDA0003301608160000022
is the derivative of the dc voltage tracking error,
Figure FDA0003301608160000023
is the derivative of the dc voltage reference;
setting the positive Lyapunov function can be expressed as:
Figure FDA0003301608160000024
in formula (7), V1 is a positive definite Lyapunov function;
derivation is performed on equation (7) to obtain:
Figure FDA0003301608160000025
in the formula (8), idrefIs a current reference value;
processing the formula (8) to obtain the current reference value;
Figure FDA0003301608160000026
in equation (9), the current reference value is used as a reference value of the d-axis current of the inner ring.
5. The method of claim 3, wherein using the current reference, the inner loop controller is predictively controlled by a predictive model, comprising:
the discrete model is:
Figure FDA0003301608160000027
Figure FDA0003301608160000028
in the formulas (10) and (11), TsIs the sampling period; u shaperdq(k-1) is a voltage sampling value at the time of k-1 on the AC side of the converter; u shapesdq(k) Is the grid side voltage at time k, idq(k) Is the current sample value at time k; i.e. idq(k +1) is the predicted value of the current at the moment k + 1; delta Urdq(k) The output voltage increment of the converter at the moment k, omega is angular frequency, Iq is a q-axis component of current, Urq is a q-axis voltage value of the alternating current side of the converter, and Urd is a d-axis voltage value of the alternating current side of the converter;
the prediction model is a neural network model;
the output expression of the neural network model is as follows:
Figure FDA0003301608160000031
in the formula (12), wiIs weight of neural network, xiFor the network input signal, corresponding to the current sample value at time k, o (w)i,xi) An output stimulus function for the network;
the weight value adjusting algorithm of the least mean square algorithm of the neural network model comprises the following steps:
wi(k+1)＝wi(k)+2ηxi(d(k)-O(k)) (13)
in the formula (13), wi(k +1) is the weight of the neural network model least mean square algorithm at the moment k +1, wi(k) And the weight value of the neural network model at the moment k is obtained.
6. The method of claim 1, wherein optimizing the predictive model comprises:
the current prediction error is defined as:
Figure FDA0003301608160000032
in the formula (15), h1Is one of the error correction coefficients, h2Is a second error correction coefficient idq(k) Is the sampled value of the current of the converter at the moment k,
Figure FDA0003301608160000033
predicting the current value at k time after compensation;
the output of the corrected prediction model is:
Figure FDA0003301608160000034
and performing rolling optimization to enable the difference value between the current predicted value and the current reference value to be the target of the rolling optimization, wherein the control objective function is as follows:
Figure FDA0003301608160000035
in formula (17), λ1,2Is the weight coefficient, ξ, of the current error in the objective function1,2Is a weight coefficient, i, of the control voltage delta error in an objective functiondqrefIs the dq-axis reference current.
CN202111192134.4A 2021-10-13 2021-10-13 Energy storage system controller design method based on inverse model predictive control Active CN113949096B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111192134.4A CN113949096B (en) 2021-10-13 2021-10-13 Energy storage system controller design method based on inverse model predictive control

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111192134.4A CN113949096B (en) 2021-10-13 2021-10-13 Energy storage system controller design method based on inverse model predictive control

Publications (2)

Publication Number Publication Date
CN113949096A true CN113949096A (en) 2022-01-18
CN113949096B CN113949096B (en) 2024-04-09

Family

ID=79330287

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111192134.4A Active CN113949096B (en) 2021-10-13 2021-10-13 Energy storage system controller design method based on inverse model predictive control

Country Status (1)

Country Link
CN (1) CN113949096B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160329714A1 (en) * 2015-05-08 2016-11-10 The Board Of Trustees Of The University Of Alabama Systems and methods for providing vector control of a grid connected converter with a resonant circuit grid filter
US20160377306A1 (en) * 2015-10-08 2016-12-29 Johnson Controls Technology Company Building control systems with optimization of equipment life cycle economic value while participating in ibdr and pbdr programs
CN109217363A (en) * 2018-08-28 2019-01-15 南京理工大学 A kind of virtual synchronous generator control method based on model current prediction
CN110021960A (en) * 2019-04-23 2019-07-16 华南理工大学 A kind of isolated island micro-capacitance sensor voltage control method for frequency based on Model Predictive Control
CN110086207A (en) * 2019-04-29 2019-08-02 国网江苏省电力有限公司电力科学研究院 A kind of grid-connected converter Control method, apparatus of energy storage and computer storage medium
CN112104000A (en) * 2020-09-18 2020-12-18 天津理工大学 Novel double-closed-loop linear active disturbance rejection control method for energy storage grid-connected inverter
CN112688307A (en) * 2020-12-04 2021-04-20 国网江苏省电力有限公司连云港供电分公司 Alternating current-direct current hybrid microgrid controller and control method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160329714A1 (en) * 2015-05-08 2016-11-10 The Board Of Trustees Of The University Of Alabama Systems and methods for providing vector control of a grid connected converter with a resonant circuit grid filter
US20160377306A1 (en) * 2015-10-08 2016-12-29 Johnson Controls Technology Company Building control systems with optimization of equipment life cycle economic value while participating in ibdr and pbdr programs
CN109217363A (en) * 2018-08-28 2019-01-15 南京理工大学 A kind of virtual synchronous generator control method based on model current prediction
CN110021960A (en) * 2019-04-23 2019-07-16 华南理工大学 A kind of isolated island micro-capacitance sensor voltage control method for frequency based on Model Predictive Control
CN110086207A (en) * 2019-04-29 2019-08-02 国网江苏省电力有限公司电力科学研究院 A kind of grid-connected converter Control method, apparatus of energy storage and computer storage medium
CN112104000A (en) * 2020-09-18 2020-12-18 天津理工大学 Novel double-closed-loop linear active disturbance rejection control method for energy storage grid-connected inverter
CN112688307A (en) * 2020-12-04 2021-04-20 国网江苏省电力有限公司连云港供电分公司 Alternating current-direct current hybrid microgrid controller and control method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
于明;王毅;李永刚;: "基于预测方法的直流微网混合储能虚拟惯性控制", 电网技术, no. 05 *

Also Published As

Publication number Publication date
CN113949096B (en) 2024-04-09

Similar Documents

Publication Publication Date Title
CN109149620B (en) Self-energy-storage multi-terminal flexible-straight system control method and system
CN110289618B (en) Grid-connected power quality compensation control method of multifunctional energy storage converter
Meng et al. A novel intelligent nonlinear controller for dual active bridge converter with constant power loads
CN112688307B (en) AC/DC hybrid micro-grid controller and control method
KR101367522B1 (en) Appatatus and method for controlling static synchronous compensator
WO2022121446A1 (en) Control system, reactive voltage control method and device, medium, and calculation device
CN113517721A (en) Multi-element power distribution network cooperative regulation and control method and system containing high-proportion distributed power supply
Swain et al. Stabilized frequency response of a microgrid using a two-degree-of-freedom controller with African vultures optimization algorithm
Xu et al. Design of three-stage start-up optimal strategy based on fuzzy fractional-order proportion integration differentiation controller of pumped storage unit under low water head
KR101367467B1 (en) Appatatus and method for controlling btb static synchronous compensator
CN113949096B (en) Energy storage system controller design method based on inverse model predictive control
CN105162159B (en) Photovoltaic grid inverter controller parameter identification method based on differential evolution algorithm
CN112510685A (en) Urban power grid risk avoidance prediction control method and device
CN117458534A (en) Novel liquid flow energy storage peak regulation and frequency modulation method and device
WO2022218270A1 (en) Method and apparatus for controlling output power of power generator, and power generator system
CN112271759B (en) Method, system and device for optimizing wind power frequency modulation parameters under multi-frequency modulation resource framework
CN115912515A (en) Load power real-time control method considering voltage out-of-limit
CN115036929A (en) Parallel APF control method and device
CN113922384A (en) Wind power plant distributed reactive voltage optimization coordination control method
Weng et al. Integrated bus voltage control method for DC microgrids based on adaptive virtual inertia control
Rafia et al. Ann-pi controller for the grid-side converters in wind energy conversions systems under voltage dips
Serra et al. Control of DC-DC boost converter in discontinuous conduction mode feeding a constant power load
Jiang et al. No-load cutting-in control of the doubly fed induction generator based on grey prediction PI control
CN118554487B (en) Frequency modulation method based on hydroelectric generating set-photovoltaic-hybrid energy storage system
CN117543728A (en) Control method and system of DRU-MMC (DRU-Modular multilevel converter) sending-out system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant