An energy storage type UPQC based on improved model predictive control and a switching state predictive control method

Abstract

The application relates to power electronics and power quality treatment technology, and particularly relates to an energy storage type UPQC based on an improved model predictive control and a switching state predictive control method.The device comprises a series active filter, a parallel active filter, a DC bus, an energy storage system and a controller, a dynamic equation of energy storage power and a state of charge is introduced into an MPC prediction model in the control method, and a multi-target cost function is designed.The controller is based on a limited control set model predictive control strategy, candidate switching vectors are predicted and evaluated in each sampling period, and the optimal switching state is output through value function optimization, so that the collaborative optimization of power quality treatment and energy storage energy management is realized.The application significantly improves the dynamic response capability and compensation accuracy of the UPQC under voltage sag / surge, imbalance, harmonic and load fluctuation working conditions, improves the energy storage utilization efficiency and system stability, and is suitable for new energy high penetration and load fluctuation large distribution networks.

Description

Technical Field

[0001] This invention relates to power electronics and power quality management technology, and in particular to an energy storage-type UPQC and a switching state predictive control method based on improved model predictive control. Background Technology

[0002] With the large-scale integration of new energy sources (wind power, photovoltaics, etc.) and the intensification of load fluctuations, power quality problems are becoming increasingly prominent, including voltage sags, voltage rises, flicker, imbalances, and harmonic distortion. These problems seriously affect the stable operation of sensitive industrial loads and distributed power sources.

[0003] The Unified Power Quality Conditioner (UPQC) consists of series and parallel active power filters and a DC energy storage unit. The series and parallel active power filters can simultaneously suppress voltage disturbances and current harmonics. However, traditional UPQCs rely on the capacitors on the DC energy storage unit side for energy support, which is insufficient for energy buffering under long-term disturbances and multi-constraint conditions, leading to a decline in compensation performance. Model predictive control (MPC) can handle system constraints, such as converter capacity and voltage / current limits, and adapt to parameter changes through rolling optimization, enhancing the robustness of UPQC under nonlinear loads and grid disturbances.

[0004] Currently, existing technologies incorporate Battery Energy Storage Systems (BESS) to enhance the energy support capabilities of UPQC (Upgraded Product Quality Control). For example, patent CN120710060A discloses a model predictive control method and device for a two-stage coupled battery energy storage grid-connected converter. This innovatively models the bidirectional DC / DC converter and inverter modules in the energy storage system together, and based on the model predictive control method, constructs a unified cost function that comprehensively considers AC side current (inverter output current) and capacitor voltage, achieving multi-objective coordinated optimization control. Alternatively, patent CN120566388A discloses a method based on the voltage and current ripple resistance of the power converter in a hybrid energy storage system. The dynamic mismatch prediction and control method first samples the voltage and current information of the battery and supercapacitor in the hybrid energy storage system to update the system's state space in real time. Then, it calculates the current reference values ​​of the battery and supercapacitor based on the power loop and cost function of the MPC. Based on the current reference value of the supercapacitor, it determines the current dynamic or steady-state condition of the system and generates a trigger signal. The trigger signal is used to drive the active model mismatch mechanism to dynamically adjust the inductance and capacitance values ​​of the power loop and current loop of the MPC. At the same time, it triggers an adaptive constraint algorithm to optimize the online constraint conditions. Finally, it generates a control signal to regulate the system operation.

[0005] However, existing control methods do not fully consider the state of charge (SOC), power constraints, and energy management of energy storage, resulting in a lack of effective coordination between energy storage and compensation, leading to dynamic response delays or insufficient compensation accuracy. Summary of the Invention

[0006] Purpose of the invention: In order to overcome the shortcomings of the prior art, this invention proposes an energy storage-type UPQC based on improved model predictive control, which solves the problems of low compensation accuracy and poor system stability under conditions of high penetration of new energy sources and load fluctuation. This invention also provides a switching state predictive control method based on energy storage-type UPQC.

[0007] Technical Solution: In a first aspect, the present invention provides an energy storage-type UPQC based on improved model predictive control. The UPQC includes: a parallel active filter, a series active filter, and a DC bus. The series active filter is connected in series between the power grid and the nonlinear load, and its output is connected to the power grid through an isolation transformer to inject voltage compensation into the load. The parallel active filter is connected in parallel between the power grid and the nonlinear load. The parallel active filter and the series active filter share a single DC bus. The invention also includes:

[0008] An energy storage system, comprising a battery pack and a bidirectional DC / DC converter, wherein the bidirectional DC / DC converter connects the energy storage system to the DC bus;

[0009] An energy management system, one end of which is connected to the energy storage system, is used to monitor the state of charge of the energy storage system in real time, dynamically adjust control parameters according to the state of charge and operating mode of the energy storage system, and output reference power or target value of state of charge to the controller.

[0010] The controller, with one end connected to the energy management system and the other end bidirectionally connected to the series active filter, is used to introduce reference power and state-of-charge dynamic equations into the model predictive control (MPC) and iteratively determine the minimum vector as the optimal switching state by combining the constructed cost function.

[0011] Furthermore, it also includes a DC-side support circuit, which includes two electrolytic capacitors connected in series. The midpoint of the two electrolytic capacitors is connected to the zero potential point of the series active filter, and the two electrolytic capacitors together maintain the stability of the DC bus voltage.

[0012] On the other hand, the present invention also provides a switch state prediction control method based on energy storage UPQC, the control method comprising:

[0013] Collect the three-phase AC phase voltage provided by the power grid and current Electrolytic capacitors in the DC-side support circuit and The two voltages of the DC bus Three-phase AC phase voltage on a series active filter Three-phase AC phase voltage on parallel active converter Three-phase AC phase voltage on the load side and current , and the state of charge of the energy storage system;

[0014] The phase of the grid voltage is obtained through a PLL phase-locked loop. Clarke and Park transforms are used to convert the acquired three-phase voltages and currents to... αβ or dq A coordinate system is established, which in turn generates voltage compensation reference values ​​and current compensation reference values;

[0015] Determine the state equation of the energy storage system and the dynamic equation of the energy storage power of the energy storage system;

[0016] For the

[0017] k Any candidate switch in each sampling period Predicted k +1 stage cost function;

[0018] Based on the cost function, candidate switch vectors are predicted and evaluated within a finite control set in each sampling period, and are gradually screened by combining hard constraints and soft constraints.

[0019] When the prediction step size is At that time, determine the candidate switch The total cost function is calculated, and the cost function value is used to prune candidate vectors that do not meet the hard constraints in advance; the vector that minimizes the total cost function is selected as the optimal switching state.

[0020] The optimal switching vector is applied to the parallel active converter and the series active converter to perform synergistic optimization of power quality compensation and energy storage management.

[0021] Furthermore, including:

[0022] The dynamic equation for the energy storage power is expressed as:

[0023] ;

[0024] in, Indicates the first k +1 sampling period Dynamic equations Indicates the first k Each sampling period Dynamic equations; Sampling time, This refers to the battery current in the energy storage system. For the power of the energy storage system, The rated energy of the battery in the energy storage system.

[0025] Furthermore, it includes: the prediction to k The stage cost function of +1 is expressed as: ;

[0026] in, As a current reference, As a voltage reference, As a DC reference, Error Standard Power reference Switch action reference , This represents the weighting factor, which is dynamically adjusted according to the operating mode of the energy management system. This refers to the power loss during switching and conduction.

[0027] Furthermore, including:

[0028] The process of predicting and evaluating candidate switch vectors within a finite control set, and progressively screening them using both hard and soft constraints, includes:

[0029] Hard constraints and soft constraints are set in the cost function. The hard constraints are constraints that must be executed and cannot be violated. The soft constraints are used to control the DC bus deviation and voltage injection upper limit, and are implemented through relaxation variables.

[0030] The hard constraint is expressed as: ;

[0031] Among them, a violation is considered an infeasibility of the candidate switch, that is, the total cost function is set to... And prune the branches in advance;

[0032] The soft constraints are expressed as follows: relaxation variables are used for DC bus deviation, upper limit of injected voltage, etc. .

[0033] Furthermore, including:

[0034] When the prediction step size is At that time, determine the candidate switch The total cost function includes:

[0035] Using prediction k The stage cost function with +1 represents the current total cost function, as follows: ;

[0036] in, The current prediction step size, Indicates the first The stage cost function of the step. As slack variables, Let be the first-order penalty vector applied to the slack variables.

[0037] Furthermore, including:

[0038] The method further includes: establishing discrete mathematical models corresponding to parallel active converters and series active converters based on generated voltage compensation reference values ​​and current compensation reference values. The discrete mathematical models corresponding to parallel active converters and series active converters include: Euler discretized state equations for series branches, Euler discretized state equations for parallel branch currents, and models of the capacitor voltage of the DC-side support circuit in the discrete time domain during the dynamic process.

[0039] Furthermore, including:

[0040] The Euler discretized state equation for the series branch is expressed as: ;

[0041] The Euler discretized state equation for the parallel branch current is expressed as: ;

[0042] in, For load-side filter inductance, For grid-side filter inductance;

[0043] The model of the capacitor voltage of the DC-side support circuit in the discrete-time domain during the dynamic process is expressed as: ;

[0044] in, This refers to the instantaneous active power on the DC energy storage side. This refers to the power loss during switching and conduction.

[0045] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0046] The control method of this invention introduces dynamic equations of energy storage power and state of charge into the MPC prediction model, and designs a multi-objective cost function that integrates voltage and current compensation, DC bus voltage stability, energy storage SOC balance, and switching losses. The controller, based on a finite control set model predictive control strategy, predicts and evaluates candidate switching vectors in each sampling period and optimizes them through the value function to output the optimal switching state. This significantly improves the dynamic response capability and compensation accuracy of existing UPQC under voltage sag / boost, imbalance, harmonic, and load fluctuation conditions, while also improving energy storage utilization efficiency and system stability. It is suitable for distribution networks with high renewable energy penetration and large load fluctuations.

[0047] By constructing a total cost function that includes multi-objective trade-offs and soft constraint penalties, this invention can quickly determine the optimal switching state in each sampling period. This improves the accuracy of voltage and current compensation while ensuring the safe operation of the DC side and energy storage system, achieving high dynamic performance and robust operation of UPQC.

[0048] The multi-objective cost function of this invention considers multiple objectives such as parallel current measurement, series voltage measurement, DC bus stability, SOC dynamic equation, switching frequency penalty, and losses, so as to more comprehensively realize the synergistic optimization of power quality management and energy storage management, thereby overcoming the shortcomings of insufficient energy support and weak robustness of traditional UPQC. Attached Figure Description

[0049] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein:

[0050] Figure 1 This is a block diagram of the predictive control of an energy storage-type UPQC model provided in an embodiment of the present invention;

[0051] Figure 2 A flowchart of the energy storage-type UPQC process provided in an embodiment of the present invention;

[0052] Figure 3 This is a diagram of the energy storage-type UPQC topology provided in an embodiment of the present invention;

[0053] Figure 4 The ideal state FCS-MPC action flowchart provided in the embodiments of the present invention;

[0054] Figure 5 This is a waveform diagram of UPQC voltage compensation during grid voltage sag / sag provided in an embodiment of the present invention;

[0055] Figure 6 This is a waveform diagram of UPQC voltage compensation when the voltage of phase A of the power grid is unbalanced, provided in an embodiment of the present invention.

[0056] Figure 7 The UPQC voltage compensation waveform and THD diagram provided for grid-side harmonic injection are embodiments of the present invention.

[0057] Figure 8 The UPQC current compensation waveform and THD diagram are provided for the grid-side harmonic injection in the embodiments of the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Example 1: As Figure 3 As shown in the figure, this embodiment of the invention provides an energy storage-type UPQC based on improved model predictive control. The UPQC includes: a parallel active filter, a series active filter, and a DC bus. The series active filter is connected in series between the power grid and the nonlinear load, and its output is connected to the power grid through an isolation transformer to inject voltage compensation into the load. The parallel active filter is connected in parallel between the power grid and the nonlinear load. The parallel active filter and the series active filter share a single DC bus. The three-phase AC phase voltage provided to the power grid; For the IGBT switching transistors on the three-phase bridge arm of the series inverter, For the IGBT switching transistors on the three-phase bridge arm of the parallel inverter; The three-phase AC phase voltages on the series converter. The three-phase AC phase voltage on the parallel converter; For power grid compensation voltage of series transformers; This refers to the voltage on the series converter side. This refers to the three-phase AC phase voltage on the load side. This is the compensation current output by the series converter; and These are the output three-phase line currents of the series converter and the parallel converter, respectively; This refers to the three-phase current on the load side. and For the filter inductors of the series converter and the parallel converter; For the filter capacitor of the series converter; and It serves as a DC-side energy storage support capacitor. This is the DC bus voltage.

[0060] An energy storage system, comprising a battery pack and a bidirectional DC / DC converter, wherein the bidirectional DC / DC converter connects the energy storage system to the DC bus;

[0061] An energy management system, one end of which is connected to the energy storage system, is used to monitor the state of charge of the energy storage system in real time, dynamically adjust control parameters according to the state of charge and operating mode of the energy storage system, and output reference power or target value of state of charge to the controller.

[0062] The controller, with one end connected to the energy management system and the other end bidirectionally connected to the series active filter, is used to introduce reference power and state-of-charge dynamic equations into the model predictive control (MPC) and iteratively determine the minimum vector as the optimal switching state by combining the constructed cost function.

[0063] The DC-side support circuit includes two electrolytic capacitors connected in series. The midpoint of the two electrolytic capacitors is connected to the zero potential point of the series active filter. The two electrolytic capacitors together maintain the stability of the DC bus voltage.

[0064] In a preferred embodiment, a series active power filter (SAPF) is connected between the power grid and the nonlinear load to compensate for voltage disturbances. In this embodiment, a T-type three-level inverter is used as the power conversion unit, and the output is connected to the power grid through an isolation transformer to inject voltage compensation into the load.

[0065] Parallel active power filter (PAPF) is connected in parallel between the power grid and the nonlinear load to suppress harmonic currents and compensate reactive power. In this embodiment, a T-type three-level inverter is also used as the power conversion unit, which is connected to the power grid through a filter inductor to ensure the smoothness of the current waveform.

[0066] The DC bus and supporting capacitors are shared by SAPF and PAPF. DC supporting capacitors are set up to balance transient voltages. The DC voltage is used as a control reference quantity to maintain the system energy stability.

[0067] The energy storage system (BESS) includes a battery pack and a bidirectional DC / DC converter. The DC / DC interface connects the energy storage system to the DC bus, enabling bidirectional energy flow. It can release energy during grid disturbances and absorb energy when the grid is stable. Preferably, the energy storage system in this embodiment is a battery energy storage system or a supercapacitor energy storage system, and can switch between charging, discharging, and standby states under different operating modes. More specifically, the energy storage system is a lithium-ion battery pack or a supercapacitor pack.

[0068] The Energy Management System (EMS) monitors the State of Charge (SOC) of the energy storage system in real time and dynamically adjusts control parameters based on the SOC status and operating mode, such as prioritizing compensation or energy storage, and outputs reference power or target SOC value to the controller.

[0069] The controller employs the Finite Control Set Model Predictive Control (FCS-MPC) algorithm, which incorporates energy storage power and SOC dynamic equations into the prediction model. The cost function comprehensively considers: voltage and current tracking errors; DC bus voltage stability; energy storage SOC constraints and energy utilization efficiency; and switching loss optimization.

[0070] In this embodiment, in each sampling period, 27 candidate switch vectors are predicted and calculated, and the optimal switch state is selected to act on the inverter to achieve coordinated optimization of energy storage and compensation.

[0071] Example 2: Figure 1 As shown, the present invention also provides a switch state prediction control method based on energy storage-type UPQC, which includes the following steps:

[0072] Step 1: Collect grid voltage, current and SOC signal.

[0073] In this embodiment, signal acquisition involves acquiring the three-phase AC phase voltages provided by the power grid. and current Electrolytic capacitors in the DC-side support circuit and The two voltages of the DC bus Three-phase AC phase voltage on a series active filter Three-phase AC phase voltage on parallel active converter Three-phase AC phase voltage on the load side and current And the state of charge of the energy storage system; that is state.

[0074] Step 2: Obtain the grid voltage phase through a phase-locked loop, and then use...

[0075] Clarke / Park Transformation calculation voltage reference value or current reference value .

[0076] Specifically, the grid voltage phase is obtained through a PLL (phase-locked loop). The Clarke and Park transformations are used to convert the three-phase voltage and current to... αβ or dq The coordinate system generates voltage compensation reference values ​​and current compensation reference values.

[0077] Introducing energy storage power into the prediction model Dynamic equations;

[0078] Specifically, in this embodiment, the Euler discretized state equations for the series branches are constructed: ;

[0079] In the above formula, For the filter inductor and filter capacitor in the series branch, Sampling time, for Clarke After the change, the series converter outputs three-phase line current. for Clarke The three-phase AC phase voltages on the series converter after the change for Clarke The voltage on the series converter side after the change for Clarke The compensation current output by the series converter;

[0080] The Euler discretized state equation for the parallel branch current is expressed as:

[0081] ;

[0082] In the above formula, For the resistor and filter inductor in the parallel branch, Sampling time, The output three-phase line current of the parallel converter after Clarke variation. for Clarke The three-phase AC phase voltages on the parallel converter after the change for Clarke The three-phase AC phase voltage on the load side after the change;

[0083] The dynamic process of the capacitor voltage on the DC energy storage side is modeled in the discrete-time domain as follows: ;

[0084] in, This refers to the instantaneous active power on the DC energy storage side. For switching and conduction losses, These are the two voltages of the DC bus. Sampling time, This is a DC-side energy storage support capacitor. This is the DC side current. for Clarke The altered power grid provides three-phase AC phase voltages. for Clarke The three-phase line current output by the power grid. The IGBT switching and conduction losses in a series-parallel converter are as follows: when the IGBT is turned on, it continuously generates heat, similar to the heat loss of a resistor; when the IGBT is switched on, it is similar to the energy loss of switching.

[0085] Dynamic equations: ;

[0086] in, This refers to the battery current in the energy storage system. For energy storage unit power, Battery rated energy.

[0087] Energy storage power dynamic equation: ;

[0088] in, For optimal charge / discharge efficiency, in practical implementation, if the battery is approximately directly connected to the bus, a value of [value] can be taken. .

[0089] Considering current, voltage, Multiple factors, including grid-side current and power constraints, are considered. This invention establishes a system that incorporates grid-side current, load voltage, DC link power, and energy storage. The multi-state equations enable unified modeling of power quality and energy flow in UPQC systems. Specifically, the voltage and current equations are used to predict the system's dynamic behavior under various candidate switching states. The equations describe the energy evolution of the energy storage unit, while the DC-link equations maintain energy balance. This modeling method enables the FCS-MPC controller to simultaneously consider compensation accuracy and energy utilization efficiency when solving for optimal switching states, thereby improving stability under long-term disturbances and the collaborative capability of the energy storage system.

[0090] Step 4: Construct a multi-objective cost function, which simultaneously considers voltage and current tracking errors, DC bus voltage deviation, SOC deviation, and switching losses. For the... k Any candidate in each sampling period s Predicted k The stage cost function of +1 is expressed as: ;

[0091] In the formula: The three-phase AC phase voltages on the parallel converter, For reference current, This is the voltage on the series converter side. For reference voltage, This is the DC bus voltage. This is the DC bus reference voltage. for Reference value These are the switching and conduction losses of the IGBT. As a current reference, As a voltage reference, As a DC reference, for Error reference As a power reference, As the reference for switch operation, The weighting factor can be dynamically adjusted according to the EMS operating mode. These are the current tracking weighting coefficient, voltage tracking weighting coefficient, and DC stability weighting coefficient, respectively. Management weight coefficient, switching penalty weight coefficient, and loss-related weight coefficient.

[0092] In this formula, each curly brace indicates the cost function corresponding to each part, which respectively represents the parallel-side current tracking function, the series-side voltage tracking function, the DC bus stability function, the SOC management function, the switching frequency penalty function, and the loss-related function.

[0093] Step 5: Predict and evaluate candidate switch vectors within a finite control set, and gradually screen them by combining hard and soft constraints.

[0094] In this embodiment, the controller sets hard constraints and soft constraints in the cost function: the soft constraints are used to control the DC bus deviation and the upper limit of voltage injection, and are implemented through relaxation variables.

[0095] The constraints are processed, and the hard constraint function is (cannot be violated):

[0096] in, This represents the maximum value of the power constraint. This represents the minimum value of the power constraint. For power constraints, , For the first The predicted current, the maximum DC current is , For the first The predicted DC power is as follows: maximum DC power is The maximum voltage of the DC bus is If this constraint is violated, the candidate state is deemed infeasible.

[0097] Soft constraints (may be slightly violated): Relaxed variables are used for DC bus deviation, upper limit of injected voltage, etc.

[0098]

[0099] in, To predict the DC bus voltage at different step sizes, This is a reference value for the DC bus voltage. For the maximum allowable voltage deviation, For the first The voltage soft-constraint relaxation variable predicted in the step. This represents the predicted value of the series injection voltage. This represents the maximum injected voltage. To inject voltage soft constraint relaxation variables, and to avoid the optimization problem becoming unsolvable due to hard constraints, relaxation variables are introduced. , To achieve a soft constraint mechanism, the relaxation variables are non-negative, reflecting the extent of constraint overruns, and a penalty term is added to the cost function, thereby improving the system's robustness and dynamic feasibility while ensuring safety constraints.

[0100] This invention introduces a soft-constraint penalty term to construct the total cost function based on the stage cost function. Hard constraints are used to limit the feasible search space, while soft constraints reflect the degree of boundary breaches through penalty relaxation variables, thereby achieving flexible adjustment of voltage, current, and energy storage power constraints. By combining stepwise screening with penalty correction, the optimization process balances real-time performance and stability.

[0101] Step 6: Determine the optimal switching vector through value function optimization.

[0102] In this embodiment, within each sampling period, the 27 switch vectors of the finite control set are traversed; the future voltage and current under each switch state are predicted. ;

[0103] Multi-step prediction and rolling optimization, when the prediction step size is At that time, for candidates The total cost function is: ;

[0104] in, Indicates the first The stage cost function of the step. These are slack variables (soft constraint out-of-bounds variables). Let be the first-order penalty vector applied to the slack variables.

[0105] This is a discount factor used to balance the weights of current and future prediction steps. When When the value approaches 1, the controller balances long-term performance and energy balance; when... When the value is small, the controller focuses more on short-time dynamic response. The introduction of this factor can strike a trade-off between rapid compensation and system stability, improving the temporal consistency of multi-step prediction models.

[0106] Calculate the cost function value and prune candidate vectors that do not meet the hard constraints in advance;

[0107] Finally, the vector with the minimum cost function is selected as the optimal switching state.

[0108] Step 7: Apply the optimal switching vector to the series and parallel active filters to perform synergistic optimization of power quality compensation and energy storage management.

[0109] In this embodiment, the optimal switching vector is applied to the inverter; the series APF injects voltage compensation to maintain load voltage stability; the parallel APF outputs compensation current to suppress harmonics and support the DC bus; the energy storage system releases or absorbs energy as needed to collaboratively achieve power quality management.

[0110] To better understand and implement this invention, this embodiment uses a three-phase three-wire system as an example, but the invention is not limited to this and can also be applied to other system topologies. These include:

[0111] In this embodiment of the application, the topology of the energy storage type unified power quality conditioner (UPQC) is as follows: Figure 3 As shown, the structure includes:

[0112] Series converter: Employs a T-type three-level inverter structure, injecting compensation voltage through a series transformer. Parallel converter: Also employs a T-type three-level inverter structure, directly connected in parallel between the grid and the load to compensate for harmonic currents and reactive power, and to stabilize the DC bus voltage.

[0113] DC-side support circuit: consists of two electrolytic capacitors The series configuration is connected with its midpoint linked to the zero potential point of the T-type inverter to jointly maintain the DC bus voltage. Stablize.

[0114] Energy Storage System (BESS): Composed of lithium-ion battery packs, connected to the DC bus of UPQC through a bidirectional DC / DC converter to achieve bidirectional energy flow (charging or discharging).

[0115] The system simulation parameters are shown in Table 1, which can be used as a reference for configuration in actual hardware.

[0116] Table 1 Simulation Parameters

[0117]

[0118] like Figure 2 As shown, the control methods include:

[0119] Signal acquisition: Acquire the three-phase AC phase voltages provided by the power grid. and current Electrolytic capacitors in the DC-side support circuit and DC bus voltage Three-phase AC phase voltage on the series converter Three-phase AC phase voltage on parallel converter Three-phase AC phase voltage on the load side and current And the state of charge of the energy storage system; that is state.

[0120] The phase of the grid voltage is obtained through a PLL phase-locked loop. The Clarke and Park transformations are used to convert the three-phase voltage and current to... αβ or dq The coordinate system generates voltage compensation reference values ​​and current compensation reference values.

[0121] Establish discrete mathematical models for series and parallel APFs:

[0122] Euler discretized state equations for series branches: ;

[0123] Euler discretized state equations for parallel branch currents: ;

[0124] The dynamic process of the capacitor voltage on the DC energy storage side is modeled in the discrete-time domain as follows: ;

[0125] Dynamic equations: ;

[0126] in, This refers to the battery current in the energy storage system. For energy storage unit power, Battery rated energy.

[0127] Energy storage power dynamic equation: ;

[0128] in, For optimal charge / discharge efficiency, in practical implementation, if the battery is approximately directly connected to the bus, a value of [value] can be taken. .

[0129] Considering current, voltage, Multiple factors, including power constraints, etc. For the first... k For any candidate s in each sampling period, the prediction is... k +1 stage cost function:

[0130] ;

[0131] in, As a current reference, As a voltage reference, As a DC reference, for Error reference As a power reference, As the reference for switch operation, The weighting factor can be dynamically adjusted according to the EMS operating mode.

[0132] Optimization and optimization: Within each sampling period, traverse the 27 switching vectors of the finite control set; predict the future voltage, current, and voltage under each switching state. ;

[0133] Multi-step prediction and rolling optimization, when the prediction step size is At that time, for candidates The total cost function is: ;

[0134] Calculate the cost function value and prune candidate vectors that do not meet the hard constraints in advance; select the vector with the minimum cost function as the optimal switching state.

[0135] Control execution: Apply the optimal switching vector to the inverter; inject voltage compensation through series APF to maintain load voltage stability; output compensation current through parallel APF to suppress harmonics and support the DC bus; the energy storage system releases or absorbs energy as needed to collaboratively achieve power quality management.

[0136] To verify the effectiveness of this control method, a system was built in MATLAB / Simulink as follows: Figure 1 The system model shown is used for simulation. The simulation parameters are set according to Table 1.

[0137] like Figure 4 As shown, this is the ideal state FCS-MPC action flowchart provided in this embodiment. State variables are used...

[0138] For example, Figure 4 This demonstrates that, under ideal conditions, the time required for MPC during the model computation phase is extremely short, almost instantaneous, which means that in

[0139] Simultaneously with the completion of system signal sampling, model calculation and optimal state updates are also completed immediately, and within the same sampling period, this optimal state is immediately applied to the current system. Therefore, in At that time, the system had already implemented the reference value of the state variable. It achieves precise approximation and tracking, thereby enabling effective control of the entire system.

[0140] Voltage sag / surge compensation, such as Figure 5 As shown, when the grid voltage experiences a 25% sag or 25% sag within 0.1s to 0.15s, the load voltage remains stable at 310V, exhibiting a smooth transient process and rapid recovery. Figure 5 (in) a The waveform of the three-phase voltage on the grid side during a voltage sag is shown. Figure 5 (in) b The waveform on the load side during a voltage sag is obtained after UPQC compensation. Figure 5 (in) c The waveform represents the three-phase voltage waveform on the grid side during a voltage spurt. Figure 5 (in) d The waveform obtained by UPQC compensation on the load side during voltage spurt is shown.

[0141] Voltage imbalance compensation, such as Figure 6 As shown, when a phase A voltage fault occurs within 0.1s-0.15s, the UPQC can effectively compensate for the negative sequence component, maintaining the balance and stability of the three-phase voltage on the load side. The three lines in the figure represent the three-phase AC power. a , b , c The three-phase voltage sinusoidal curves, the two graphs compare the load-side stable waveform obtained after UPQC compensation when voltage sags, dips, imbalances, and harmonic injections occur on the grid side. Specifically, Figure 6 (in) a )for a When the phase voltages are unbalanced, the three-phase voltage waveforms output from the grid side are as follows: Figure 6 (in) b The waveform obtained by UPQC compensation on the load side when there is voltage imbalance is shown.

[0142] Harmonic suppression, such as Figure 7 and Figure 8 As shown, when the grid voltage is injected with 10% of the 5th and 7th harmonics or the load is a nonlinear load, UPQC can effectively filter out harmonics, making the THD of both the load voltage and the grid current below 2.5%, meeting the national power quality standards. Figure 7 In the diagram, (a) shows the three-phase voltage waveform output from the grid side during harmonic injection. Figure 7 (in) b The waveform shown is the load-side voltage compensated by UPQC during harmonic injection. Figure 7 (in) c This is a schematic diagram of the voltage THD results after UPQC compensation; where Figure 8 (in) aThe waveform represents the three-phase output voltage on the grid side during harmonic injection. Figure 8 (in) b This is a diagram showing the three-phase current waveforms on the load side after UPQC compensation during harmonic injection. Figure 8 (in) c The diagram shows the THD results of the current after UPQC.

[0143] In this embodiment, Total Harmonic Distortion (THD) is an indicator that measures the degree of signal distortion and is mainly used to assess the degree of harmonic pollution in the output of power systems or electronic equipment. It is defined as the ratio of the effective value of each harmonic to the effective value of the fundamental frequency, and is usually expressed as a percentage.

[0144] The simulation results above demonstrate that the energy storage-based UPQC control method proposed in this invention, based on the improved FCS-MPC, can effectively address various power quality issues, achieve synergistic optimization of power quality compensation and energy storage management, and significantly improve the dynamic response speed and operational stability of the system.

[0145] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. "A plurality of" means two or more, unless otherwise explicitly specified.

[0146] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0147] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0148] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0149] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0150] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0151] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0152] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0153] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0154] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. An energy storage based model predictive control based UPQC, the UPQC comprising: Parallel active filter, series active filter and DC bus, the series active filter is connected in series between the power grid and the nonlinear load, the output end is connected to the power grid through an isolation transformer, and a voltage compensation amount is injected into the load, the parallel active filter is connected in parallel between the power grid and the nonlinear load, and the parallel active filter and the series active filter share a DC bus, characterized in that it further comprises: An energy storage system, the energy storage system comprises a battery pack and a bidirectional DC / DC converter, and the bidirectional DC / DC converter is connected to the DC bus; An energy management system, one end of the energy management system is connected to the energy storage system, for real-time monitoring of the state of charge of the energy storage system, dynamic adjustment of control parameters according to the state of charge and the operating mode of the energy storage system, and output of a reference power or a target value of the state of charge to the controller; A controller, one end of the controller is connected to the energy management system, and the other end is in bidirectional communication with the series active filter, which is used for introducing a reference power and a state of charge dynamic equation into model predictive control (MPC), and determining the minimum vector as the optimal switching state through multiple iterations combined with a constructed cost function.

2. The improved model predictive control based energy storage unified power quality conditioner according to claim 1, characterized in that, Further comprising: A DC side support circuit comprising two electrolytic capacitors connected in series, the midpoint of the two electrolytic capacitors being connected to the zero potential point of the series active filter, and the two electrolytic capacitors jointly maintaining the stability of the DC bus voltage.

3. A switching state prediction control method based on energy storage type UPQC, characterized in that, The control method comprises: The three-phase alternating current phase voltages and currents provided by the power grid, the electrolytic capacitor in the DC side support circuit and the voltage of the DC bus, the three-phase alternating current phase voltages on the series active filter, the three-phase alternating current phase voltages on the parallel active current converter, the load side three-phase alternating current phase voltages and currents, and the state of charge of the energy storage system; Obtaining grid voltage phase by PLL , adopt Clarke transform and Park transform, the collected relevant three-phase voltage, current is converted to αβ or dq coordinate system, and then generate voltage compensation reference value and current compensation reference value; Determining the state of charge equation of the energy storage system and the energy storage power dynamic equation of the energy storage system; In the first k any candidate switch for a sampling period , predicted to k +1 stage cost function; In each sampling period, candidate switching vectors are predicted and evaluated in a limited control set, and are gradually screened in combination with hard constraints and soft constraints; When the prediction step is a total cost function of candidate switches is determined, a cost function value is calculated, and a candidate vector that does not satisfy a hard constraint is pruned in advance; a vector that makes the total cost function minimum is selected as an optimal switch state; The optimal switching vector is applied to the parallel active converter and the series active converter to perform coordinated optimization of power quality compensation and energy storage energy management.

4. The switching state prediction control method based on energy storage type UPQC according to claim 3, characterized in that, The energy storage power dynamic equation is expressed as: ; wherein, represents the k + 1 sample period, dynamic equation, represents the k sample period, dynamic equation; is the sampling time, is the battery current in the energy storage system, is the energy storage system power, is the battery rated energy in the energy storage system.

5. The switching state prediction control method based on energy storage type UPQC according to claim 4, characterized in that, The prediction to k +1 phase cost function, denoted as: ; wherein, is a three-phase AC phase voltage on the parallel converter, is a reference current, is a series converter side voltage, is a reference voltage, is a DC bus voltage, is a DC bus reference voltage, is a reference value, is an IGBT switching and conduction loss value, is a current reference, is a voltage reference, is a DC reference, is an error reference, is a power reference, is a switching action reference, the weight factors can be dynamically adjusted depending on the EMS operating mode, are respectively a current tracking weight coefficient, a voltage tracking weight coefficient, a DC stability weight coefficient, a management weight coefficient, a switching penalty weight coefficient and a loss related weight coefficient.

6. The switching state prediction control method based on energy storage type UPQC according to claim 5, characterized in that, The prediction and evaluation of the candidate switching vectors in the limited control set, and the gradual screening in combination with the hard constraints and the soft constraints, comprise: Hard constraints and soft constraints are set in the cost function, the hard constraints are constraints that must be performed and cannot be violated, and the soft constraints are used to control the DC bus deviation and the upper limit of voltage injection, and are realized through slack variables; The hard constraints are expressed as: ; wherein is a maximum value of the power constraint, is a minimum value of the power constraint, is the power constraint, is the current of the first step prediction, is the DC maximum current, is the DC power of the first step prediction, is the DC maximum power, violating the hard constraint, i.e. the current candidate switch state is infeasible. The soft constraint is represented as: the DC bus deviation, the upper limit of the injected voltage, etc., using a relaxation variable: Wherein, is the DC bus voltage under the prediction step, is the reference value of the DC bus voltage, is the maximum allowed voltage deviation, is the first predicted voltage soft constraint relaxation variable, represents the series injected voltage prediction value, is the maximum injected voltage, is the injected voltage soft constraint relaxation variable, to avoid the hard constraint leading to no solution of the optimization problem, a relaxation variable , is introduced to realize the soft constraint mechanism, the relaxation variable is non-negative, reflects the constraint out-of-bounds amplitude, and an additional penalty term is added in the cost function, so as to guarantee the safety constraint while improving the system robustness and dynamic feasibility.

7. The switching state prediction control method of energy storage based UPQC according to claim 6, characterized in that, The total cost function of the candidate switch is determined when the prediction step is ​ The prediction to k The phase cost function of +1 represents the current total cost function as follows: ; wherein, is the step size of the current prediction, denotes the stage cost function of the step, is the relaxation variable, is a penalty vector on the relaxation variable.

8. The switching state prediction control method based on energy storage type UPQC according to claim 3, characterized in that, The method further comprises: establishing a discrete mathematical model corresponding to the parallel active converter and the series active converter based on the generated voltage compensation reference value and the current compensation reference value, the discrete mathematical model corresponding to the parallel active converter and the series active converter comprising: a series branch Euler discretization state equation, a parallel branch current Euler discretization state equation, and a model of capacitor voltage in a dynamic process in a discrete time domain of the DC side support circuit.

9. The switching state prediction control method based on energy storage type UPQC according to claim 8, characterized in that, The series branch Euler-discretized state equation is expressed as: ; wherein, is the filter inductance and filter capacitance in the series branch, is the sampling time, is Clarke is the three-phase line current output by the modified series converter, is Clarke is the three-phase AC phase voltage on the modified series converter, is Clarke is the voltage on the modified series converter side, is Clarke is the compensation current output by the series converter; The parallel branch current Euler discretization state equation is expressed as: ； wherein, is the resistance and filter inductance in the parallel branch, is the sampling time, is the Clarke-transformed parallel converter output three-phase line current, is the Clarke is the Clarke-transformed parallel converter output three-phase line current, is the Clarke is the Clarke-transformed parallel converter output three-phase line current, The model of capacitor voltage in a dynamic process in a discrete time domain of the DC side support circuit is expressed as: ； wherein, is the instantaneous active power of the DC energy storage side, is the switching and conduction losses power, is the voltage of the two sections of the DC bus, is the sampling time, is the DC side energy storage support capacitor, is the DC side current, is Clarke the changed three-phase AC phase voltage provided by the grid, is Clarke the three-phase line current output by the grid.

Images