Rapid power control method and system for off-grid micro-grid containing grid-forming distributed power supply

Abstract

The invention discloses a rapid power control method and system for an off-grid micro-grid containing a grid-forming distributed power supply. The method comprises the following steps: based on a microgrid information physical fusion model, sensing and identifying parameters and operation states of each part of the microgrid through a finite state machine; carrying out detailed modeling and description on each state of the system, and clearly displaying state conversion information and conditions of the micro-grid system; primary control based on a network construction type converter is introduced to control the power of the system, active and reactive power output is adjusted, and when the system senses and identifies corresponding parameter changes, each stage of control is triggered to act; secondary control is introduced to carry out rapid power control adjustment on the micro-grid, frequency and voltage excursion caused by primary control is made up, external disturbance is suppressed, and the power of each part of equipment is reasonably distributed; model predictive control based on data driving is introduced, accurate power distribution and global optimal control are achieved, and the system operates stably.

Description

Technical Field

[0001] The present invention belongs to the technical field of microgrid information-physical fusion, and in particular relates to a fast power control method and system for an off-grid microgrid containing a grid-connected distributed power source. Background Art

[0002] The electric energy structure is shifting towards a high proportion of distributed access to new energy. However, the large-scale access to renewable energy such as wind power and photovoltaics has had a significant impact on the structure and operation mode of the power system. Microgrid is an emerging research topic in the field of power systems today. As a component of the smart grid, microgrid is a small power system that facilitates the effective integration of distributed generators. Existing microgrid systems are difficult to clearly identify and reflect the transitions between system states, and it is difficult to associate power control with system states. Grid-Forming Distributed Generation (Grid-Forming DG for short) refers to distributed power sources that can provide the frequency and voltage support required by the power grid in microgrids or distributed energy systems. This type of power supply usually has the ability to operate autonomously and can provide a stable power supply to the load even without a main grid connection.

[0003] Distributed power generation with grid-connected structure is one of the key technologies for microgrids to achieve autonomous operation and island operation. However, for off-grid microgrids with distributed power generation with grid-connected structure, since they are not operated in parallel with the large power grid, their operating state will be more complicated and more difficult to control. Therefore, there is an urgent need for a method to control the off-grid microgrid and improve the power supply stability. Summary of the invention

[0004] Purpose of the invention: To address the deficiencies in the prior art, the present invention provides a method and system for fast power control of an off-grid microgrid containing a distributed power source, which introduces a finite state machine tool to clearly display the distinction between the various states of the system and their respective conversion conditions, and uses the primary and secondary control principles to quickly control and adjust the power between the various devices in the microgrid.

[0005] Technical solution: In the first aspect, a fast power control method for an off-grid microgrid including a grid-connected distributed power source comprises the following steps:

[0006] Based on the microgrid information-physical fusion model, the state of each distributed power source in the microgrid is modeled to obtain a finite state machine that describes the information and conditions of the state transition of each distributed power source; the operating parameters of each distributed power source are obtained, and the finite state machine is used to perceive and identify the operating state according to the operating parameters, and the state transition logic is applied to manage the operation of the microgrid;

[0007] When the state of the finite state machine changes, the corresponding parameters of the distributed power source change. In response to the change of the parameters, the control strategies of each level pre-established in the microgrid system are triggered to control the power of the microgrid. The control strategies of each level include primary control based on the grid-type converter, secondary control and data-driven model predictive control. The primary control is used to adjust the active and reactive output of the system; the secondary control is used to compensate for the frequency and voltage deviations caused by the primary control and prevent external disturbances, so that the power of each part of the equipment is reasonably distributed; the data-driven model predictive control is used to achieve precise power distribution and global optimal control of the voltage vector, so that the system can operate stably.

[0008] Furthermore, the distributed power source in the microgrid includes a wind turbine unit, a photovoltaic unit, and an energy storage unit;

[0009] The state modeling of each distributed power source in the microgrid is performed to obtain a finite state machine describing information and conditions of state transition of each distributed power source, including:

[0010] The wind turbine unit, the photovoltaic unit, and the energy storage unit are respectively regarded as a system, and the operating state of each system is divided into several discrete states. According to the state transition law of each system, a corresponding state transition diagram is established, where the state is defined as: n Indicates the state of the system; e ij Indicates that the system changes from state S i Transition to state S j The required conditions are: n=1,2,3..., i,j∈n.

[0011] Furthermore, the state modeling of the fan unit is expressed as follows:

[0012] S 1 :P WT =0 Shutdown

[0013] S 2 :P WT =P N Constant power output

[0014] S 3 : Maximum Power Output

[0015] e 12 :v n ≤v<v out

[0016] e 13 :v in ≤v<v n

[0017] e21 :v≥v out

[0018] e 23 :v in ≤v<v n

[0019] e 31 :v<v in

[0020] e 32 :v n ≤v<v out

[0021] Among them, P WT is the actual power of the fan; P N is the rated power of the fan; v is the actual wind speed; v in is the cut-in wind speed; v n is the rated wind speed; v out To cut out wind speed.

[0022] Furthermore, the state modeling of the photovoltaic unit is expressed as follows:

[0023] S 1 :P PV =P ref Constant power output

[0024] S 2 :P PV =P max Maximum Power Output

[0025] e 12 :P max <P ref

[0026] e 21 :P max ≥P ref

[0027] Among them, P PV is the actual power of the photovoltaic unit; P ref P is the constant power reference value issued by the system; max It is the maximum power that the photovoltaic can generate at this time.

[0028] Furthermore, the state modeling of the energy storage unit is expressed as follows:

[0029] S 1 :Discharge

[0030] S 2 :Charge

[0031] S 3 :Overcharge shutdown

[0032] S 4 :idle

[0033] S 5 : Insufficient charging shutdown

[0034] e 14 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0035] e 15 :SOC(t)<SOC min

[0036] e 23 :SOC>SOC max

[0037] e 24 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0038] e 31 :P input (t)<P load (t)SOC(t)≥SOC max

[0039] e 41 :P input (t)<P load (t)&SOC min <SOC(t)<SOC max

[0040] e 42 :P input (t)>P load (t)&SOC min <SOC(t)<SOC max

[0041] e 52 :P input (t)>P load (t)SOC(t)≤SOC min

[0042] Among them, P input Represents the sum of the power input to the energy storage unit; P load Indicates the total power consumed by the load; SOC min Indicates the minimum allowable value of the state of charge; SOCmax It represents the maximum allowable value of the state of charge, and SOC(t) represents the state of charge at time t.

[0043] Furthermore, the primary control based on the grid-connected converter includes:

[0044] After collecting the filter capacitor voltage and filter inductor current at the converter port, they are fed into the inner loop through dq0 transformation on the one hand, and input into the droop control link through the power calculation link on the other hand to generate the dq component of the reference voltage and feed it into the voltage loop in the voltage-current dual closed-loop control. The dq component of the current is calculated together with the dq component of the measured voltage and fed into the current loop. After abc transformation, it is converted into a three-phase reference voltage, which is modulated into a gate signal to drive the converter through the PWM link. The phase angle used for dq transformation and inverse transformation is autonomously generated by the grid-type control.

[0045] Furthermore, the droop control is expressed by the following formula:

[0046]

[0047] In the formula, m P and n Q are the droop slopes of the frequency droop control and voltage droop control of the equipment respectively; P and Q are the actual values ​​of the active power and reactive power of the equipment respectively; ω and V are the frequency and voltage reference values ​​of the equipment respectively; ω ref and V ref are the frequency and voltage ratings of the equipment, respectively;

[0048] Furthermore, the secondary control is a distributed control introduced between various energy storage units, and each energy storage unit obtains its own information and the information of adjacent energy storage units, and calculates its own reference value according to a consistency algorithm.

[0049] Furthermore, the data-driven model predictive control adopts a dual-vector model predictive control strategy to achieve global optimal control of the voltage vector of the three-phase two-level inverter;

[0050] The mathematical model of the three-phase two-level inverter in the dq axis coordinate system is expressed as:

[0051]

[0052] In the formula, u od 、u oq is the dq axis component of the inverter output voltage, i Ld 、i Lq is the dq axis component of the inductor current, u id 、u iq is the dq axis component of the inverter port voltage; L fis the line coupling inductance; R is the line resistance; ω is the system angular frequency; assuming the sampling period is T S , then the system current prediction model expression at the moment is:

[0053]

[0054] In the formula, x(k) is the variable at t k The sampling value at time t, x(k+1) is the variable at t k+1 The sampling value at the moment;

[0055] The following value function is used to evaluate the prediction error, and the predicted values ​​are substituted in turn to compare the error size. The voltage vector that minimizes the value function is selected as the optimal voltage vector:

[0056] f g =(i dref -i Ld (k+1)) 2 +(i qref -i Lq (k+1)) 2

[0057] i dref 、i qref is the dq value component of the current reference value;

[0058] On the basis of selecting an optimal voltage vector by single vector MPC, another vector selection is performed. The selection principle is that after the two voltage vectors act together for one sampling cycle, the error between the dq axis current and the given value is the smallest;

[0059] Then the optimal reference voltage vector is:

[0060]

[0061] In the formula, u d1 、u d2 are the d-axis components of the first and second optimal voltage vectors respectively; u q1 、u q2 are the q-axis components of the first and second optimal voltage vectors respectively; t 1 ,t 2 are the action time of the first and second voltage vectors respectively.

[0062] In a second aspect, an off-grid microgrid fast power control system including a grid-connected distributed power source comprises:

[0063] The state modeling and management module is used to perform state modeling on each distributed power source in the microgrid based on the microgrid information-physical fusion model, and obtain a finite state machine that describes the information and conditions of the state transition of each distributed power source; obtain the operating parameters of each distributed power source, use the finite state machine to perceive and identify the operating state according to the operating parameters, and apply the state transition logic to manage the operation of the microgrid;

[0064] The power control module is used to respond to the change of the parameters when the state of the finite state machine changes and the corresponding parameters of the distributed power source change, so that the control strategies of various levels pre-established in the microgrid system are triggered to control the power of the microgrid. The control strategies of various levels include primary control based on the grid-type converter, secondary control and data-driven model predictive control. The primary control is used to adjust the active and reactive output of the system; the secondary control is used to compensate for the frequency and voltage offsets caused by the primary control and prevent external disturbances, so that the power of each part of the equipment is reasonably distributed; the data-driven model predictive control is used to achieve accurate power distribution and global optimal control of the voltage vector, so that the system can operate stably.

[0065] Furthermore, the distributed power source in the microgrid includes a wind turbine unit, a photovoltaic unit, and an energy storage unit;

[0066] The state modeling and management module performs state modeling on each distributed power source in the microgrid to obtain a finite state machine describing information and conditions of state transition of each distributed power source, including:

[0067] The wind turbine unit, the photovoltaic unit, and the energy storage unit are respectively regarded as a system, and the operating state of each system is divided into several discrete states. According to the state transition law of each system, a corresponding state transition diagram is established, where the state is defined as: n Indicates the state of the system; e ij Indicates that the system changes from state S i Transition to state S j The required conditions are: n=1,2,3..., i,j∈n.

[0068] Furthermore, the state modeling of the fan unit is expressed as follows:

[0069] S 1 :P WT =0 Shutdown

[0070] S 2 :P WT =P N Constant power output

[0071] S 3 : Maximum Power Output

[0072] e 12 :v n ≤v<v out

[0073] e 13 :v in ≤v<v n

[0074] e 21 :v≥v out

[0075] e 23 :v in ≤v<v n

[0076] e 31 :v<v in

[0077] e 32 :v n ≤v<v out

[0078] Among them, P WT is the actual power of the fan; P N is the rated power of the fan; v is the actual wind speed; v in is the cut-in wind speed; v n is the rated wind speed; v out To cut out wind speed.

[0079] Furthermore, the state modeling of the photovoltaic unit is expressed as follows:

[0080] S 1 :P PV =P ref Constant power output

[0081] S 2 :P PV =P max Maximum Power Output

[0082] e 12 :P max <P ref

[0083] e 21 :P max ≥P ref

[0084] Among them, P PV is the actual power of the photovoltaic unit; P ref P is the constant power reference value issued by the system; max It is the maximum power that the photovoltaic can generate at this time.

[0085] Furthermore, the state modeling of the energy storage unit is expressed as follows:

[0086] S 1 :Discharge

[0087] S 2 :Charge

[0088] S 3 :Overcharge shutdown

[0089] S 4 :idle

[0090] S 5 : Insufficient charging shutdown

[0091] e 14 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0092] e 15 :SOC(t)<SOC min

[0093] e 23 :SOC>SOC max

[0094] e 24 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0095] e 31 :P input (t)<P load (t)SOC(t)≥SOC max

[0096] e 41 :P input (t)<P load (t)&SOC min <SOC(t)<SOC max

[0097] e 42 :P input (t)>P load (t)&SOC min <SOC(t)<SOC max

[0098] e 52 :Pinput (t)>P load (t)SOC(t)≤SOC min

[0099] Among them, P input Represents the sum of the power input to the energy storage unit; P load Indicates the total power consumed by the load; SOC min Indicates the minimum allowable value of the state of charge; SOC max It represents the maximum allowable value of the state of charge, and SOC(t) represents the state of charge at time t.

[0100] Furthermore, the primary control based on the grid-connected converter includes:

[0101] After collecting the filter capacitor voltage and filter inductor current at the converter port, they are fed into the inner loop through dq0 transformation on the one hand, and input into the droop control link through the power calculation link on the other hand to generate the dq component of the reference voltage and feed it into the voltage loop in the voltage-current dual closed-loop control. The dq component of the current is calculated together with the dq component of the measured voltage and fed into the current loop. After abc transformation, it is converted into a three-phase reference voltage, which is modulated into a gate signal to drive the converter through the PWM link. The phase angle used for dq transformation and inverse transformation is autonomously generated by the grid-type control.

[0102] Furthermore, the droop control is expressed by the following formula:

[0103]

[0104] In the formula, m P and n Q are the droop slopes of the frequency droop control and voltage droop control of the equipment respectively; P and Q are the actual values ​​of the active power and reactive power of the equipment respectively; ω and V are the frequency and voltage reference values ​​of the equipment respectively; ω ref and V ref are the frequency and voltage ratings of the equipment, respectively;

[0105] Furthermore, the secondary control is a distributed control introduced between various energy storage units, and each energy storage unit obtains its own information and the information of adjacent energy storage units, and calculates its own reference value according to a consistency algorithm.

[0106] Furthermore, the data-driven model predictive control adopts a dual-vector model predictive control strategy to achieve global optimal control of the voltage vector of the three-phase two-level inverter;

[0107] The mathematical model of the three-phase two-level inverter in the dq axis coordinate system is expressed as:

[0108]

[0109] In the formula, uod 、u oq is the dq axis component of the inverter output voltage, i Ld 、i Lq is the dq axis component of the inductor current, u id 、u iq is the dq axis component of the inverter port voltage; L f is the line coupling inductance; R is the line resistance; ω is the system angular frequency; assuming the sampling period is T S , then the system current prediction model expression at the moment is:

[0110]

[0111] In the formula, x(k) is the variable at t k The sampling value at time t, x(k+1) is the variable at t k+1 The sampling value at the moment;

[0112] The following value function is used to evaluate the prediction error, and the predicted values ​​are substituted in turn to compare the error size. The voltage vector that minimizes the value function is selected as the optimal voltage vector:

[0113] f g =(i dref -i Ld (k+1)) 2 +(i qref -i Lq (k+1)) 2

[0114] i dref 、i qref is the dq value component of the current reference value;

[0115] On the basis of selecting an optimal voltage vector by single vector MPC, another vector selection is performed. The selection principle is that after the two voltage vectors act together for one sampling cycle, the error between the dq axis current and the given value is the smallest;

[0116] Then the optimal reference voltage vector is:

[0117]

[0118] In the formula, u d1 、u d2 are the d-axis components of the first and second optimal voltage vectors respectively; u q1 、u q2 are the q-axis components of the first and second optimal voltage vectors respectively; t 1 ,t 2 are the action time of the first and second voltage vectors respectively.

[0119] The present invention also provides a computer device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the off-grid microgrid fast power control method containing grid-connected distributed power sources as described above are implemented.

[0120] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the off-grid microgrid fast power control method containing a grid-connected distributed power source as described above are implemented.

[0121] Beneficial effects:

[0122] (1) Based on the information-physical fusion model of the microgrid, the present invention firstly senses and identifies the system parameters through the finite state machine tool, and determines the triggering conditions of the system state transition. By introducing the finite state machine tool, the distinction between the various states of the system and their respective transition conditions can be clearly displayed, laying the foundation for the subsequent optimal control strategy for each operating state.

[0123] (2) In terms of model control prediction, it is necessary to reasonably use information data to predict subsequent behaviors and control system power. The present invention is based on the principle of primary and secondary control. It can quickly control and adjust the system power while preventing disturbances. The primary control can quickly respond to load changes, adjust the output power of the generator, maintain the stability of the grid frequency and voltage, improve the stability and reliability of the grid, and initially realize automatic adjustment. Secondary control is used to quickly control and adjust the power of the microgrid to compensate for the frequency and voltage offsets caused by primary control, prevent external disturbances, and reasonably distribute the power of each part of the equipment. The dual-vector model predictive control strategy is introduced to achieve global optimal control of the voltage vector. BRIEF DESCRIPTION OF THE DRAWINGS

[0124] Figure 1 It is a flow chart of a fast power control method of an off-grid microgrid containing a grid-connected distributed power source according to the present invention.

[0125] Figure 2 It is a state transition diagram of an off-grid microgrid information-physical system including a grid-connected distributed power source according to the present invention.

[0126] Figure 3 This is a primary control structure diagram of an off-grid microgrid including a grid-connected distributed power source according to the present invention.

[0127] Figure 4 It is a secondary control principle diagram of an off-grid microgrid including a grid-connected distributed power source according to the present invention. DETAILED DESCRIPTION

[0128] The present application is further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present application.

[0129] like Figure 1 As shown, a fast power control method of an off-grid microgrid containing a grid-connected distributed power source of the present invention comprises the following steps:

[0130] Step 1: Based on the microgrid information-physical fusion model, the online dynamic parameter perception and identification technology is introduced through the finite state machine to perceive and identify the parameters and operating status of each part of the microgrid.

[0131] The microgrid mainly includes wind, solar, energy storage and flexible loads. Since the loads are generally not modeled separately, in the solution of the present invention, the information-physical model of the microgrid considers the wind, solar, energy storage part, that is, the wind turbine unit, photovoltaic unit, and energy storage unit contained in the distributed power source DG (distributed edge generators). Each part is equipped with a finite state machine to facilitate the conversion and monitoring of the system state. On this basis, in the finite state machine tool (such as stateflow), the voltage, current, power and other parameters of each distributed power source in the microgrid can be obtained in real time through the reserved interface, and the monitoring data can be feature extracted and analyzed using filters and signal processing tools, that is, the perception and identification of parameters can be realized. If the system conversion conditions are met, the finite state machine will drive the microgrid system (or equipment) to convert the state.

[0132] The present invention embodies the information characteristics of the microgrid based on a finite state machine. The wind turbine unit, photovoltaic unit, and energy storage unit in the microgrid are respectively regarded as a system. Each system contains countable states. When the corresponding state event occurs, the system will transfer from the current state to the corresponding state. The system can not only perform state conversion and monitoring by this method under normal circumstances, but also identify the corresponding parameters when disturbed. Here is an example: the operating state of the microgrid is divided into several discrete states, such as normal operation, load fluctuation, off-grid accidents, etc. According to the state transition law, a state transition diagram is established to describe the state evolution process of the microgrid. Under normal operating conditions, the voltage, current, power and other parameters of the microgrid are in a stable state, while under load fluctuation conditions, these parameters may show obvious changes and fluctuations. Therefore, different processing methods can be used to sample and analyze the relevant parameters. Under normal operating conditions, filtering and averaging methods can be used to stabilize the parameters; while under load fluctuation conditions, time-frequency analysis and wavelet transform methods can be used to extract and analyze the parameters. According to the results of parameter perception and identification, the state estimation and control strategy of the microgrid are updated.

[0133] The premise of using finite state machines is to model and describe each state of the system in detail, and clearly display the information and conditions of the state transition of the microgrid system.

[0134] Here, the state of the system is defined: S n (n=1,2,3...) indicates the state of the system; e ij (i,j∈n) represents the system from state S i Transition to state S j The required conditions.

[0135] Figure 2 The state transition diagram of the off-grid microgrid cyber-physical system containing a distributed power source is shown. For each part of the system, it can be expressed as follows:

[0136] 1) Wind turbine unit (WT):

[0137] S 1 :P WT =0 Shutdown

[0138] S 2 :P WT =P N Constant power output

[0139] S 3 : Maximum Power Output

[0140] e 12 :v n ≤v<v out

[0141] e 13 :v in ≤v<v n

[0142] e 21 :v≥v out

[0143] e 23 :v in ≤v<v n

[0144] e 31 :v<v in

[0145] e 32 :v n ≤v<v out

[0146] Among them, P WT is the actual power of the fan; P N is the rated power of the fan; v is the actual wind speed; vin is the cut-in wind speed; v n is the rated wind speed; v out To cut out wind speed.

[0147] 2) Photovoltaic unit (PV):

[0148] S 1 :P PV =P ref Constant power output

[0149] S 2 :P PV =P max Maximum Power Output

[0150] e 12 :P max <P ref

[0151] e 21 :P max ≥P ref

[0152] Among them, P PV is the actual power of the photovoltaic unit; P ref P is the constant power reference value issued by the system; max It is the maximum power that the photovoltaic can generate at this time.

[0153] 3) Energy storage unit (BSS):

[0154] For an energy storage unit, an important indicator to measure its performance is its state of charge (SOC).

[0155] S 1 :Discharge

[0156] S 2 :Charge

[0157] S 3 :Overcharge shutdown

[0158] S 4 :idle

[0159] S 5 : Insufficient charging shutdown

[0160] e 14 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0161] e 15 :SOC(t)<SOCmin

[0162] e 23 :SOC>SOC max

[0163] e 24 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0164] e 31 :P input (t)<P load (t)SOC(t)≥SOC max

[0165] e 41 :P input (t)<P load (t)&SOC min <SOC(t)<SOC max

[0166] e 42 :P input (t)>P load (t)&SOC min <SOC(t)<SOC max

[0167] e 52 :P input (t)>P load (t)SOC(t)≤SOC min

[0168] Among them, P input Indicates the sum of the power input to the BSS at this moment; P load Indicates the total power consumed by the load; SOC min Indicates the minimum allowable value of the state of charge; SOC max Indicates the maximum allowable value of the state of charge.

[0169] Step 2: Use the primary control based on the grid-connected converter to control the power of the system and adjust the active and reactive output.

[0170] When the system senses and identifies changes in corresponding parameters, each level of control will be triggered to act. The present invention introduces primary control and secondary control between each device as well as data-driven model predictive control, wherein the primary control is a droop control based on a grid-type converter. Since the droop control adjusts the output of active power and reactive power by adjusting the amplitude and frequency of the voltage when a disturbance occurs in the power grid, this will cause deviations in the amplitude and frequency of the output voltage, so secondary control is introduced. Secondary control performs distributed control between energy storages, requiring energy storage units to communicate and exchange information with each other to provide reference values ​​for parameters for primary control. Based on data-driven model predictive control. When the system senses and identifies changes in the external environment or related parameters, it will trigger each level of control to act, quickly adjust the output of active and reactive power, and achieve rapid power control.

[0171] The synchronization method of the grid-type converter is similar to that of the synchronous machine. It adjusts the frequency of the reference voltage by exchanging power with the grid, thereby achieving synchronization with the grid. It belongs to voltage-type control and generally shows voltage source characteristics. Figure 3 As shown in the figure, after the system collects the filter capacitor voltage and filter inductor current at the converter port, it is fed into the inner loop through dq0 transformation on the one hand, and input into the droop control link through the power calculation link on the other hand, generating the dq component of the reference voltage and feeding it into the voltage loop in the voltage and current dual closed-loop control, and calculating the current dq component together with the dq component of the measured voltage and feeding it into the current loop, and then transformed into a three-phase reference voltage through abc, and modulated into a gate signal to drive the converter through the PWM link. The phase angle used for the dq transformation and the inverse transformation is generated autonomously by the grid-type control, without the need for a phase-locked loop link.

[0172] Primary control can quickly respond to load changes, adjust generator output power, maintain grid frequency and voltage stability, and improve grid stability and reliability. It can initially achieve automatic adjustment, reduce manual intervention costs, improve the operation efficiency of microgrids, and thus reduce overall operating costs. The introduction of primary control can also enhance the autonomy and self-adaptation of microgrids, giving microgrids the ability to self-regulate and adapt, improving their resistance to external interference and enhancing the autonomy of microgrids.

[0173] The droop control of the system equipment is expressed by the following formula:

[0174]

[0175] In the formula, m P and n Q are the droop slopes of the frequency droop control and voltage droop control of the equipment respectively; P and Q are the actual values ​​of the active power and reactive power of the equipment respectively; ω and V are the frequency and voltage reference values ​​of the equipment respectively. ref and V refare the frequency and voltage ratings of the equipment, respectively.

[0176] After droop control is performed, the voltage and current dual closed-loop module is introduced to calculate a stable voltage reference value, which is returned to the inverter to control the corresponding equipment.

[0177] Step 4: Use secondary control to quickly adjust the power of the microgrid to compensate for the frequency and voltage deviation caused by primary control, prevent external disturbances, and reasonably distribute the power of each device:

[0178] Introducing distributed secondary control between various energy storages can not only continuously detect deviations in grid frequency and voltage, and eliminate these deviations through internal adjustments to restore frequency and voltage to standard values, but also coordinate the operation of the energy storage system and distributed power generation equipment within the microgrid, and adjust the charging and discharging power of the energy storage system according to grid demand; actively adjust the power of controllable loads to achieve active load management.

[0179] like Figure 4 As shown in the figure, distributed control is designed based on graph theory and collaborative control. Each distributed power source DG only needs to obtain its own and adjacent DG information, and according to the consistency algorithm, it can calculate its own reference value to make the voltage amplitude and frequency of each DG consistent. Usually one or several DGs are set as leader nodes, accept the set reference value information, and propagate the reference value. Non-leader nodes update their own status according to the status of adjacent nodes. Through the iterative process, all nodes gradually reach the same reference value as the leader node, thereby achieving consistency in voltage amplitude and frequency. Finally, the voltage of each node can reach the rated value.

[0180] The quadratic controller with a multi-agent consensus algorithm with a leader node is used for design. Taking the quadratic frequency controller as an example, its control law is:

[0181]

[0182] Among them, ω i and ω j Represents the angular frequency of the output voltage of device i and device j respectively; N i is the set of devices adjacent to device i on the communication network; C w >0 is control gain; a ij is an element in the adjacency matrix. If device j transmits information to device i, then a ij =1, otherwise it is 0; g i is the control gain, which represents the ability to receive the reference signal ω ref If it is a leader node, then g i=1, otherwise it is 0. For the control law of the secondary voltage controller, the voltage angular frequency in the above formula can be replaced by the voltage amplitude.

[0183] Step 5: Use data-driven model predictive control to achieve accurate power distribution and global optimal control of the voltage vector to ensure stable system operation.

[0184] If the microgrid is regulated only by droop control and secondary control, due to problems such as line impedance mismatch, droop control will often fail to achieve accurate power distribution, thereby affecting the stable operation of the parallel system. Data-driven model predictive control is widely used in the optimal control of voltage source inverters due to its advantages such as simple control structure and fast dynamic response speed. In model predictive control, multi-vector model predictive control is mostly used to improve system response speed and suppress current harmonics. To this end, quickly and accurately calculating the action time of the voltage vector is the key to determining its control effect. The present invention introduces a dual-vector model predictive control strategy to achieve global optimal control of the voltage vector.

[0185] The mathematical model of the three-phase two-level inverter in the dq axis coordinate system can be expressed as:

[0186]

[0187] In the formula, u od 、u oq is the dq axis component of the inverter output voltage, i Ld 、i Lq is the dq axis component of the inductor current, u id 、u iq is the dq axis component of the inverter port voltage; L f is the line coupling inductance; R is the line resistance; ω is the system angular frequency. Assume that the sampling period is T S , then the system current prediction model expression at the moment is:

[0188]

[0189] In the formula, x(k) is the variable at t k The sampling value at time t, x(k+1) is the variable at t k+1 The sampling value at the time.

[0190] The following value function is used to evaluate the prediction error, and the predicted values ​​are substituted in turn to compare the error size. The voltage vector that minimizes the value function is selected as the optimal voltage vector:

[0191] f g =(i dref -i Ld (k+1)) 2 +(iqref -i Lq (k+1)) 2

[0192] In the formula, i dref 、i qref is the dq value component of the current reference value. On the basis of selecting an optimal voltage vector by single vector MPC, another vector selection is performed. The selection principle is that after the two voltage vectors act together for a sampling cycle, the error between the dq axis current and the given value is the smallest.

[0193] Then the optimal reference voltage vector is:

[0194]

[0195] In the formula, u d1 、u d2 are the d-axis components of the first and second optimal voltage vectors respectively; u q1 、u q2 are the q-axis components of the first and second optimal voltage vectors respectively; t 1 ,t 2 are the action time of the first and second voltage vectors respectively.

[0196] By introducing primary and secondary control and data-driven model predictive control, not only can disturbances and voltage frequency deviations be effectively prevented, but power can also be quickly adjusted between devices. When the system senses and identifies changes in the external environment or related parameters, each level of control will be triggered, and the equipment can respond quickly to achieve rapid power control of the microgrid.

[0197] According to the power control method of the present invention, when applied, a finite state machine tool is installed in each device to monitor the system state in real time and drive the system state transition. When the system parameters change, the conditions required for the system state transition can be quickly sensed and identified. Taking the change of the power instruction issued by the system as an example, when the wind turbine changes its output, the system can sense and identify the change in the parameters of the wind turbine, and convert the state of the entire microgrid system by confirming the trigger conditions. On this basis, the primary control, secondary control and model predictive control modules will act quickly, immediately give the optimal reference voltage, adjust the active and reactive output of each unit, and realize the rapid regulation and distribution of system power.

[0198] The present invention also provides an off-grid microgrid fast power control system containing a grid-connected distributed power source, comprising:

[0199] The state modeling and management module is used to perform state modeling on each distributed power source in the microgrid based on the microgrid information-physical fusion model, and obtain a finite state machine that describes the information and conditions of the state transition of each distributed power source; obtain the operating parameters of each distributed power source, use the finite state machine to perceive and identify the operating state according to the operating parameters, and apply the state transition logic to manage the operation of the microgrid;

[0200] The power control module is used to respond to the change of the parameters when the state of the finite state machine changes and the corresponding parameters of the distributed power source change, so that the control strategies of various levels pre-established in the microgrid system are triggered to control the power of the microgrid. The control strategies of various levels include primary control based on the grid-type converter, secondary control and data-driven model predictive control. The primary control is used to adjust the active and reactive output of the system; the secondary control is used to compensate for the frequency and voltage offsets caused by the primary control and prevent external disturbances, so that the power of each part of the equipment is reasonably distributed; the data-driven model predictive control is used to achieve accurate power distribution and global optimal control of the voltage vector, so that the system can operate stably.

[0201] The distributed power source in the microgrid includes a wind turbine unit, a photovoltaic unit, and an energy storage unit;

[0202] The state modeling and management module performs state modeling on each distributed power source in the microgrid to obtain a finite state machine describing information and conditions of state transition of each distributed power source, including:

[0203] The wind turbine unit, photovoltaic unit, and energy storage unit are regarded as a system respectively, and the operating state of each system is divided into several discrete states. According to the state transition law of each system, the corresponding state transition diagram is established, where the state is defined as: n Indicates the state of the system; e ij Indicates that the system changes from state S i Transition to state S j The required conditions are: n=1,2,3..., i,j∈n.

[0204] The state modeling of the fan unit is expressed as follows:

[0205] S 1 :P WT =0 Shutdown

[0206] S 2 :P WT =P N Constant power output

[0207] S 3 : Maximum Power Output

[0208] e 12 :v n ≤v<v out

[0209] e 13 :v in ≤v<v n

[0210] e 21 :v≥v out

[0211] e 23 :v in ≤v<v n

[0212] e 31 :v<v in

[0213] e 32 :v n ≤v<v out

[0214] Among them, P WT is the actual power of the fan; P N is the rated power of the fan; v is the actual wind speed; v in is the cut-in wind speed; v n is the rated wind speed; v out To cut out wind speed.

[0215] The state modeling of the photovoltaic unit is expressed as follows:

[0216] S 1 :P PV =P ref Constant power output

[0217] S 2 :P PV =P max Maximum Power Output

[0218] e 12 :P max <P ref

[0219] e 21 :P max ≥P ref

[0220] Among them, P PV is the actual power of the photovoltaic unit; P ref P is the constant power reference value issued by the system; max It is the maximum power that the photovoltaic can generate at this time.

[0221] The state modeling of the energy storage unit is expressed as follows:

[0222] S 1 :Discharge

[0223] S 2 :Charge

[0224] S 3 :Overcharge shutdown

[0225] S 4 :idle

[0226] S 5 : Insufficient charging shutdown

[0227] e 14 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0228] e 15 :SOC(t)<SOC min

[0229] e 23 :SOC>SOC max

[0230] e 24 :P input (t) = P load (t)&SOC min <SOC(t)<SOC max

[0231] e 31 :P input (t)<P load (t)SOC(t)≥SOC max

[0232] e 41 :P input (t)<P load (t)&SOC min <SOC(t)<SOC max

[0233] e 42 :P input (t)>P load (t)&SOC min <SOC(t)<SOC max

[0234] e 52 :P input (t)>P load(t)SOC(t)≤SOC min

[0235] Among them, P input Represents the sum of the power input to the energy storage unit; P load Indicates the total power consumed by the load; SOC min Indicates the minimum allowable value of the state of charge; SOC max It represents the maximum allowable value of the state of charge, and SOC(t) represents the state of charge at time t.

[0236] The primary control based on the grid-type converter includes:

[0237] After collecting the filter capacitor voltage and filter inductor current at the converter port, they are fed into the inner loop through dq0 transformation on the one hand, and input into the droop control link through the power calculation link on the other hand to generate the dq component of the reference voltage and feed it into the voltage loop in the voltage-current dual closed-loop control. The dq component of the current is calculated together with the dq component of the measured voltage and fed into the current loop. After abc transformation, it is converted into a three-phase reference voltage, which is modulated into a gate signal to drive the converter through the PWM link. The phase angle used for dq transformation and inverse transformation is autonomously generated by the grid-type control.

[0238] The droop control is expressed by the following equation:

[0239]

[0240] In the formula, m P and n Q are the droop slopes of the frequency droop control and voltage droop control of the equipment respectively; P and Q are the actual values ​​of the active power and reactive power of the equipment respectively; ω and V are the frequency and voltage reference values ​​of the equipment respectively; ω ref and V ref are the frequency and voltage ratings of the equipment, respectively;

[0241] The secondary control is a distributed control introduced between various energy storage units. Each DG obtains its own and adjacent DG information and calculates its own reference value according to the consistency algorithm, so that the voltage amplitude and frequency of each DG are consistent.

[0242] The data-driven model predictive control adopts a dual-vector model predictive control strategy to achieve global optimal control of the voltage vector;

[0243] The mathematical model of the three-phase two-level inverter in the dq axis coordinate system is expressed as:

[0244]

[0245] In the formula, u od 、u oqis the dq axis component of the inverter output voltage, i Ld 、i Lq is the dq axis component of the inductor current, u id 、u iq is the dq axis component of the inverter port voltage; L f is the line coupling inductance; R is the line resistance; ω is the system angular frequency. Assume that the sampling period is T S , then the system current prediction model expression at the moment is:

[0246]

[0247] In the formula, x(k) is the variable at t k The sampling value at time t, x(k+1) is the variable at t k+1 The sampling value at the moment;

[0248] The following value function is used to evaluate the prediction error, and the predicted values ​​are substituted in turn to compare the error size. The voltage vector that minimizes the value function is selected as the optimal voltage vector:

[0249] f g =(i dref -i Ld (k+1)) 2 +(i qref -i Lq (k+1)) 2

[0250] i dref 、i qref is the dq value component of the current reference value;

[0251] On the basis of selecting an optimal voltage vector by single vector MPC, another vector selection is performed. The selection principle is that after the two voltage vectors act together for one sampling cycle, the error between the dq axis current and the given value is the smallest;

[0252] Then the optimal reference voltage vector is:

[0253]

[0254] In the formula, u d1 、u d2 are the d-axis components of the first and second optimal voltage vectors respectively; u q1 、u q2 are the q-axis components of the first and second optimal voltage vectors respectively; t 1 ,t 2 are the action time of the first and second voltage vectors respectively.

[0255] It should be understood that the off-grid microgrid fast power control system containing grid-connected distributed power sources in the embodiment of the present invention can implement all the technical solutions in the above-mentioned method embodiment, and the functions of its various functional modules can be specifically implemented according to the methods in the above-mentioned method embodiments. The specific implementation process can refer to the relevant description in the above-mentioned embodiments, which will not be repeated here.

[0256] The present invention also provides a computer device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the off-grid microgrid fast power control method containing grid-connected distributed power sources as described above are implemented.

[0257] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the off-grid microgrid fast power control method containing a grid-connected distributed power source as described above are implemented.

[0258] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, devices (systems), computer equipment or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0259] The present invention is described with reference to a flowchart of a method according to an embodiment of the present invention. It should be understood that each process in the flowchart and a combination of processes in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A device that specifies functions in a process or multiple processes.

[0260] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A function specified in a process or multiple processes.

[0261] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 The steps of a specified function in a process or multiple processes.

Claims

1. A fast power control method for an off-grid microgrid containing a distributed power source, characterized in that: The method comprises: Based on the microgrid information-physical fusion model, the state of each distributed power source in the microgrid is modeled to obtain a finite state machine that describes the information and conditions of the state transition of each distributed power source; Acquire operating parameters of each distributed power source, use the finite state machine to sense and identify the operating state according to the operating parameters, and apply state transition logic to manage the operation of the microgrid; When the state of the finite state machine changes, the corresponding parameters of the distributed power source change. In response to the change of the parameters, the control strategies of each level pre-established in the microgrid system are triggered to control the power of the microgrid. The control strategies of each level include primary control based on the grid-type converter, secondary control and data-driven model predictive control. The primary control is used to adjust the active and reactive output of the system; the secondary control is used to compensate for the frequency and voltage deviations caused by the primary control and prevent external disturbances, so that the power of each part of the equipment is reasonably distributed; the data-driven model predictive control is used to achieve precise power distribution and global optimal control of the voltage vector, so that the system can operate stably.

2. The method according to claim 1, characterized in that The distributed power source in the microgrid includes a wind turbine unit, a photovoltaic unit, and an energy storage unit; The state modeling of each distributed power source in the microgrid is performed to obtain a finite state machine describing information and conditions of state transition of each distributed power source, including: The wind turbine unit, the photovoltaic unit, and the energy storage unit are respectively regarded as a system, and the operating state of each system is divided into several discrete states. According to the state transition law of each system, a corresponding state transition diagram is established, where the state is defined as: n Indicates the state of the system; e ij Indicates that the system changes from state S i Transition to state S j The required conditions are: n=1,2,3..., i,j∈n.

3. The method according to claim 2, characterized in that The state modeling of the fan unit is expressed as follows: S1:P WT =0 Shutdown S2:P WT =P N Constant power output S3: Maximum Power Output yes 12 :v n ≤v<v out yes 13 :v in ≤v<v n and 21 :v≥v out yes 23 :v in ≤v<v n yes 31 :v<v in yes 32 :v n ≤v<v out Among them, P WT is the actual power of the fan; P N is the rated power of the fan; v is the actual wind speed; v in is the cut-in wind speed; v n is the rated wind speed; v out To cut out wind speed.

4. The method according to claim 2, characterized in that: The state modeling of the photovoltaic unit is expressed as follows: S1:P PV =P ref Constant power output S2:P PV =P max Maximum Power Output e 12 :P max <P ref And 21 :P max ≥P ref Among them, P PV is the actual power of the photovoltaic unit; P ref P is the constant power reference value issued by the system; max It is the maximum power that the photovoltaic can generate at this time.

5. The method according to claim 2, characterized in that: The state modeling of the energy storage unit is expressed as follows: S1: Discharge S2: Charging S3: Overcharge shutdown S4: Idle S5: Insufficient charging shutdown from 14 :P input (t)=P load (t)&SOC min <SOC(t)<SOC max in 15 :SOC(t)<SOC min in 23 :SOC>SOC max from 24 :P input (t)=P load (t)&SOC min <SOC(t)<SOC max e 31 :P input (t)<P load (t)SOC(t)≥SOC max from 41 :P input (t)<P load (t)&SOC min <SOC(t)<SOC max from 42 :P input (t)>P load (t)&SOC min <SOC(t)<SOC max e 52 :P input (t)>P load (t)SOC(t)≤SOC min Among them, P input Represents the sum of the power input to the energy storage unit; P load Indicates the total power consumed by the load; SOC min Indicates the minimum allowable value of the state of charge; SOC max It represents the maximum allowable value of the state of charge, and SOC(t) represents the state of charge at time t.

6. The method according to claim 1, characterized in that The primary control based on the grid-type converter includes: After collecting the filter capacitor voltage and filter inductor current at the converter port, they are fed into the inner loop through dq0 transformation on the one hand, and input into the droop control link through the power calculation link on the other hand to generate the dq component of the reference voltage and feed it into the voltage loop in the voltage-current dual closed-loop control. The dq component of the current is calculated together with the dq component of the measured voltage and fed into the current loop. After abc transformation, it is converted into a three-phase reference voltage, which is modulated into a gate signal to drive the converter through the PWM link. The phase angle used for dq transformation and inverse transformation is autonomously generated by the grid-type control.

7. The method according to claim 6, characterized in that The droop control is expressed by the following equation: In the formula, m P and n Q are the droop slopes of the frequency droop control and voltage droop control of the equipment respectively; P and Q are the actual values ​​of the active power and reactive power of the equipment respectively; ω and V are the frequency and voltage reference values ​​of the equipment respectively; ω ref and V ref are the frequency and voltage ratings of the equipment, respectively.

8. The method according to claim 1, characterized in that: The secondary control is a distributed control introduced between various energy storage units. Each energy storage unit obtains information about itself and adjacent energy storage units and calculates its own reference value according to a consistency algorithm.

9. The method according to claim 1, characterized in that: The data-driven model predictive control adopts a dual-vector model predictive control strategy to achieve global optimal control of the voltage vector of the three-phase two-level inverter; The mathematical model of the three-phase two-level inverter in the dq axis coordinate system is expressed as: In the formula, u od 、u oq is the dq axis component of the inverter output voltage, i Ld 、i Lq is the dq axis component of the inductor current, u id 、u iq is the dq axis component of the inverter port voltage; L f is the line coupling inductance; R is the line resistance; ω is the system angular frequency; assuming the sampling period is T S , then the system current prediction model expression at the moment is: In the formula, x(k) is the variable at t k The sampling value at time t, x(k+1) is the variable at t k+1 The sampling value at the moment; The following value function is used to evaluate the prediction error, and the predicted values ​​are substituted in turn to compare the error size. The voltage vector that minimizes the value function is selected as the optimal voltage vector: f g =(i dref -i Ld (k+1)) 2 +(i qref -i Lq (k+1)) 2 i dref 、i qref is the dq value component of the current reference value; On the basis of selecting an optimal voltage vector by single vector MPC, another vector selection is performed. The selection principle is that after the two voltage vectors act together for one sampling cycle, the error between the dq axis current and the given value is the smallest; Then the optimal reference voltage vector is: In the formula, u d1 、u d2 are the d-axis components of the first and second optimal voltage vectors respectively; u q1 、u q2 are the q-axis components of the first and second optimal voltage vectors respectively; t1 and t2 are the action times of the first and second voltage vectors respectively.

10. An off-grid microgrid fast power control system containing a grid-connected distributed power source, characterized in that: include: The state modeling and management module is used to perform state modeling on each distributed power source in the microgrid based on the microgrid information-physical fusion model, and obtain a finite state machine that describes the information and conditions of the state transition of each distributed power source; obtain the operating parameters of each distributed power source, use the finite state machine to perceive and identify the operating state according to the operating parameters, and apply the state transition logic to manage the operation of the microgrid; The power control module is used to respond to the change of the parameters when the state of the finite state machine changes and the corresponding parameters of the distributed power source change, so that the control strategies of various levels pre-established in the microgrid system are triggered to control the power of the microgrid. The control strategies of various levels include primary control based on the grid-type converter, secondary control and data-driven model predictive control. The primary control is used to adjust the active and reactive output of the system; the secondary control is used to compensate for the frequency and voltage offsets caused by the primary control and prevent external disturbances, so that the power of each part of the equipment is reasonably distributed; the data-driven model predictive control is used to achieve accurate power distribution and global optimal control of the voltage vector, so that the system can operate stably.

11. The system according to claim 10, characterized in that The distributed power source in the microgrid includes a wind turbine unit, a photovoltaic unit, and an energy storage unit; The state modeling and management module performs state modeling on each distributed power source in the microgrid to obtain a finite state machine describing information and conditions of state transition of each distributed power source, including: The wind turbine unit, the photovoltaic unit, and the energy storage unit are respectively regarded as a system, and the operating state of each system is divided into several discrete states. According to the state transition law of each system, a corresponding state transition diagram is established, where the state is defined as: n Indicates the state of the system; e ij Indicates that the system changes from state S i Transition to state S j The required conditions are: n=1,2,3..., i,j∈n.

12. The system according to claim 11, characterized in that The state modeling of the fan unit is expressed as follows: S1:P WT =0 Shutdown S2:P WT =P N Constant power output S3: Maximum Power Output yes 12 :v n ≤v<v out yes 13 :v in ≤v<v n and 21 :v≥v out yes 23 :v in ≤v<v n yes 31 :v<v in yes 32 :v n ≤v<v out Among them, P WT is the actual power of the fan; P N is the rated power of the fan; v is the actual wind speed; v in is the cut-in wind speed; v n is the rated wind speed; v out To cut out wind speed.

13. The system according to claim 11, characterized in that The state modeling of the photovoltaic unit is expressed as follows: S1:P PV =P ref Constant power output S2:P PV =P max Maximum Power Output e 12 :P max <P ref And 21 :P max ≥P ref Among them, P PV is the actual power of the photovoltaic unit; P ref P is the constant power reference value issued by the system; max It is the maximum power that the photovoltaic can generate at this time.

14. The system according to claim 11, characterized in that The state modeling of the energy storage unit is expressed as follows: S1: Discharge S2: Charging S3: Overcharge shutdown S4: Idle S5: Insufficient charging shutdown from 14 :P input (t)=P load (t)&SOC min <SOC(t)<SOC max in 15 :SOC(t)<SOC min in 23 :SOC>SOC max from 24 :P input (t)=P load (t)&SOC min <SOC(t)<SOC max e 31 :P input (t)<P load (t)SOC(t)≥SOC max from 41 :P input (t)<P load (t)&SOC min <SOC(t)<SOC max from 42 :P input (t)>P load (t)&SOC min <SOC(t)<SOC max e 52 :P input (t)>P load (t)SOC(t)≤SOC min Among them, P input Represents the sum of the power input to the energy storage unit; P load Indicates the total power consumed by the load; SOC min Indicates the minimum allowable value of the state of charge; SOC max It represents the maximum allowable value of the state of charge, and SOC(t) represents the state of charge at time t.

15. The system according to claim 10, characterized in that The primary control based on the grid-type converter includes: After collecting the filter capacitor voltage and filter inductor current at the converter port, they are fed into the inner loop through dq0 transformation on the one hand, and input into the droop control link through the power calculation link on the other hand to generate the dq component of the reference voltage and feed it into the voltage loop in the voltage-current dual closed-loop control. The dq component of the current is calculated together with the dq component of the measured voltage and fed into the current loop. After abc transformation, it is converted into a three-phase reference voltage, which is modulated into a gate signal to drive the converter through the PWM link. The phase angle used for dq transformation and inverse transformation is autonomously generated by the grid-type control.

16. The system according to claim 15, characterized in that The droop control is expressed by the following equation: In the formula, m P and n Q are the droop slopes of the frequency droop control and voltage droop control of the equipment respectively; P and Q are the actual values ​​of the active power and reactive power of the equipment respectively; ω and V are the frequency and voltage reference values ​​of the equipment respectively; ω ref and V ref are the frequency and voltage ratings of the equipment, respectively.

17. The system according to claim 10, characterized in that The secondary control is a distributed control introduced between various energy storage units. Each energy storage unit obtains information about itself and adjacent energy storage units and calculates its own reference value according to a consistency algorithm.

18. The system according to claim 10, characterized in that The data-driven model predictive control adopts a dual-vector model predictive control strategy to achieve global optimal control of the voltage vector of the three-phase two-level inverter; The mathematical model of the three-phase two-level inverter in the dq axis coordinate system is expressed as: In the formula, u od 、u oq is the dq axis component of the inverter output voltage, i Ld 、i Lq is the dq axis component of the inductor current, u id 、u iq is the dq axis component of the inverter port voltage; L f is the line coupling inductance; R is the line resistance; ω is the system angular frequency; assuming the sampling period is T S , then the system current prediction model expression at the moment is: In the formula, x(k) is the variable at t k The sampling value at time t, x(k+1) is the variable at t k+1 The sampling value at the moment; The following value function is used to evaluate the prediction error, and the predicted values ​​are substituted in turn to compare the error size. The voltage vector that minimizes the value function is selected as the optimal voltage vector: f g =(i dref -i Ld (k+1)) 2 +(i qref -i Lq (k+1)) 2 i dref 、i qref is the dq value component of the current reference value; On the basis of selecting an optimal voltage vector by single vector MPC, another vector selection is performed. The selection principle is that after the two voltage vectors act together for one sampling cycle, the error between the dq axis current and the given value is the smallest; Then the optimal reference voltage vector is: In the formula, u d1 、u d2 are the d-axis components of the first and second optimal voltage vectors respectively; u q1 、u q2 are the q-axis components of the first and second optimal voltage vectors respectively; t1 and t2 are the action times of the first and second voltage vectors respectively.

19. A computer device, characterized in that: include: one or more processors; Memory; And one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and when the programs are executed by the processors, the steps of the off-grid microgrid fast power control method containing grid-connected distributed power sources as described in any one of claims 1-9 are implemented.

20. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the off-grid microgrid fast power control method containing a grid-connected distributed power source as described in any one of claims 1-9 are implemented.

Images