Direct current micro-grid passive model prediction control method based on predefined time feedforward compensation

Abstract

The invention provides a direct-current micro-grid passive model prediction control method based on predefined time feed-forward compensation. The design process of the control method specifically comprises the following steps of: firstly, performing online estimation and feed-forward compensation on external interference by designing a predefined time nonlinear disturbance observer, so that the robustness of the system is remarkably improved; in order to obtain the optimal control structure, system parameters are estimated online instead of strictly assumed interference signals. Then, an adaptive passive model predictive control (APMPC) composed of a passive control voltage loop and a model predictive control current loop is designed, and the adaptive passive model predictive control (APMPC) has the advantages of simple structure, high robustness, rapid voltage regulation and the like. In addition, before the model predicts and controls the current loop, simple current limiting can be realized through a saturation current limiting link. The predefined time disturbance observer provided by the invention ensures that the error signal converges in the freely set time upper limit.

Description

Technical Field

[0001] The present invention relates to the technical field of direct current microgrids, and in particular to a direct current microgrid passive model predictive control method based on predefined time feedforward compensation. Background Art

[0002] In recent years, microgrids have attracted extensive attention as small power systems that effectively integrate renewable energy, energy storage systems, distributed generation units, and modern power electronic loads. Compared with AC microgrids, the control structure of DC microgrids is relatively simple, which can avoid reactive power, phase synchronization, and frequency regulation issues. In addition, DC / DC boost converters are one of the most common interfaces of DC microgrids, which can reduce the redundant energy conversion stages of most DC power sources and loads. Therefore, it is very important to explore the control strategy of DC microgrid interface converters that can achieve fast bus voltage regulation and stabilize constant power loads.

[0003] In DC microgrids, modern power electronic loads usually include resistive loads and constant power loads. The former are passive components that can provide positive damping for DC microgrid systems. In contrast, constant power loads are strictly regulated power electronic converter loads that consume constant power, causing the input to present a negative incremental impedance characteristic, thereby interacting with the source converter to reduce system damping and even leading to limit cycle oscillation behavior at the equilibrium point. Therefore, high-penetration constant power loads pose a huge threat to the stable operation of DC microgrids. To address the above issues, most studies are based on small-signal stability and large-signal stability, respectively. Among them, the control strategies based on small-signal stability are mainly divided into two categories: passive damping by adding passive devices to the power circuit and active damping by modifying the control loop to reshape the system impedance. However, the former increases the weight and cost of the system, and the latter relies on a linearized small-signal model and can only ensure system stability within a small range of the equilibrium point.

[0004] Considering the highly nonlinearity of the converter and the negative impedance characteristics of the constant power load, many advanced control techniques have been proposed to ensure the global stability of the system, such as passive control, backstepping control, model predictive control, sliding mode control, and deep reinforcement learning control. Although the above controls ensure the stability of the system, the design of the nonlinear controller depends on the accurate system model construction, which will cause steady-state deviation of the system control target under random disturbances at the source and load ends. In order to ensure the error-free regulation of the nonlinear control, a composite nonlinear control based on integral compensation is proposed. However, the feedback regulation mechanism leads to a slow dynamic response speed of the controller when facing disturbances. It is worth noting that the disturbance observer technology has attracted widespread attention as an online estimation and feedforward compensation technology independent of the nonlinear controller design. However, traditional nonlinear disturbance observers, finite-time disturbance observers, and fixed-time disturbance observers can only converge in asymptotic time, finite time limited by the initial state, and inaccurate fixed time with difficult parameter adjustment. Summary of the invention

[0005] The purpose of the present invention is to solve the problems in the prior art to meet the needs of actual situations, and proposes a DC microgrid passive model predictive control method based on predefined time feedforward compensation.

[0006] The present invention is realized by the following technical scheme. The present invention proposes a DC microgrid passive model predictive control method based on predefined time feedforward compensation. The control method firstly estimates and feedforwards the external disturbance online by designing a predefined time nonlinear disturbance observer, which is specifically used to accurately estimate the time-varying and unknown load power and input voltage within a predetermined time; then, an adaptive passive model predictive controller APMPC composed of a passive control voltage loop and a model predictive control current loop is designed to control a DC / DC boost converter with a constant power load; finally, before the model predictive control current loop, current limiting is achieved through a saturation current limiting link.

[0007] Furthermore, the DC / DC boost converter is mathematically modeled. Without loss of generality, the average dynamic model of the DC / DC boost converter is expressed as:

[0008]

[0009] Among them, L, C, E, v o ,i L , μ, R and P CPL are the inductance value, output side capacitance value, input voltage, output voltage, inductor current, switch duty cycle, resistive load and constant power load power of the DC / DC boost converter, δ1=E and δ2=-Po represent the state variables that need to be estimated;

[0010] In order to take the output power in the dynamic model (1) as a disturbance form and satisfy the Brunovsky standard form, the exact feedback linearization technique is used for coordinate transformation, which is expressed as

[0011]

[0012] In formula (3), E0 is the rated input voltage, ψ1 and ψ2 represent the disturbance forms under different modeling respectively;

[0013] The input voltage in the dynamic model (1) is regarded as a disturbance and converted into a form that complies with the Brunovsky criterion, expressed as

[0014]

[0015] Furthermore, the design of the predefined time nonlinear disturbance observer is as follows: In order to estimate δ1 in equation (1) within a predetermined time, a reference auxiliary variable is introduced, and its expression is:

[0016]

[0017] In the above formula, γ1 is a known positive constant; the predefined time nonlinear disturbance observer for estimating the input voltage δ1 is designed as

[0018]

[0019] In the above formula, is the estimated value of the input voltage δ1; to ensure that the designed predefined time nonlinear disturbance observer satisfies Lyapunov stability, the function Designed to

[0020]

[0021] In the above formula, To estimate the error signal, 0<ξ<1, T o1 >0, α1>0, α2>0 and T o1 represents a predefined time coefficient that can be freely set, i.e. for any time constant t>T o1 Achievable and

[0022] In order to estimate δ2 in equation (1), a reference auxiliary variable is introduced, and its expression is

[0023]

[0024] In the above formula, γ2 is a known positive constant; the predefined time nonlinear disturbance observer for estimating the output power δ2 is designed as

[0025]

[0026] In the above formula, is the estimated value of the output power δ2; to ensure that the designed predefined time nonlinear disturbance observer satisfies Lyapunov stability, the function Designed to

[0027]

[0028] In the above formula, To estimate the error signal, 0<ξ<1, T o2 >0, α1>0, α2>0 and T o2 represents a predefined time coefficient that can be freely set, i.e. for any time constant t>T o2 Achievable and

[0029] Furthermore, in the passive control voltage loop design, in order to facilitate the design of the adaptive passive model predictive controller, the dynamic model (1) is expressed in the following matrix form:

[0030]

[0031] Where G = -G T , R = R T , Formula (12) satisfies the characteristics of the Euler-Lagrange model and provides a model basis for the design of the adaptive passive model predictive controller.

[0032] Furthermore, the design of the adaptive passive model predictive controller is divided into two stages:

[0033] (1) Energy reshaping stage: Adopt Transform and reshape formula (11) into

[0034]

[0035] In the above formula, X d and represent the reference value and interference value of X respectively;

[0036] (2) Damping injection stage: In this stage, the virtual damping matrix is ​​injected into both ends of equation (13): To modify the system dissipation function, expressed as

[0037]

[0038] Furthermore, to ensure that the system injected with the virtual damping matrix satisfies Lyapunov stability, there exists a global stable equilibrium point of the system The energy function of the system is selected as follows:

[0039]

[0040] The time derivative of the energy function (16) is expressed as

[0041]

[0042] Furthermore, the control law of the passive control voltage loop is designed as

[0043]

[0044] Substituting the estimated values ​​of equations (6) and (9) into (18) we can express it as

[0045]

[0046] Furthermore, in the process of model prediction control current loop design, the average state model (1) and the inductor current waveform when the duty cycle μ=1 or 0 are combined to calculate the current slope as

[0047]

[0048] According to the inductor current i at the current sampling time L (k), the inductor current i at the next moment L The predicted value of (k+1) is expressed as

[0049]

[0050] In the above formula, T s is the switching cycle; the cost function constructed based on the single-cycle predicted current and reference current is expressed as

[0051] J=(i L,ref -i L (k+1)) 2 (twenty two)

[0052] To obtain the optimal variable, the partial derivative of μ in equation (22) is obtained:

[0053]

[0054] The calculation formula of duty cycle μ can be derived as

[0055]

[0056] Substituting the estimated value of formula (6) into (24), we can obtain

[0057]

[0058] The present invention proposes an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a DC microgrid passive model predictive control method based on predefined time feedforward compensation are implemented.

[0059] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of a DC microgrid passive model predictive control method based on predefined time feedforward compensation.

[0060] Beneficial effects of the present invention:

[0061] The present invention proposes an adaptive passive model predictive control (APMPC) method based on a predefined time nonlinear disturbance observer, aiming to achieve excellent transient and steady-state performance under a wide range of operation with a more concise and practical control structure. The proposed controller includes a predefined time feedforward compensation loop and a nonlinear control feedback loop. First, the robustness of the system is significantly improved by designing a predefined time nonlinear disturbance observer to estimate and feedforward the external disturbance online. In order to obtain the optimal control structure, the system parameters (output power and input voltage) will be estimated online instead of strictly assumed interference signals. Then, the APMPC consisting of a passive control voltage loop and a model predictive control current loop is designed, which has the advantages of simple structure, high robustness, fast voltage regulation, etc. In addition, before the model predictive control current loop, simple current limiting can also be achieved through a saturation current limiting link. The predefined time disturbance observer proposed in the present invention ensures that the error signal converges within a freely set time upper bound. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0063] Figure 1 It is a typical DC microgrid structure and its equivalent circuit schematic diagram;

[0064] Figure 2 It is the waveform of inductor current and duty cycle in a single switching cycle;

[0065] Figure 3It is a structural block diagram of a controller in the adaptive passive model predictive control method of the present invention;

[0066] Figure 4 is the different gain coefficient T o The simulation waveform of the predefined time nonlinear disturbance observer when ;

[0067] Figure 5 It is a dynamic response simulation waveform diagram of switching from open-loop operation to the control method proposed by the present invention;

[0068] Figure 6 It is the experimental waveform diagram of the resistive load change (160Ω→80Ω→160Ω) under the control method of the present invention;

[0069] Figure 7 It is the experimental waveform diagram of constant power load change (200W→600W→200W) under the control method of the present invention;

[0070] Figure 8 It is an experimental waveform diagram of input voltage change (100V→125V→100V) under the control method of the present invention;

[0071] Fig. 9 It is an experimental waveform diagram of input voltage change (100V→75V→100V) under the control method of the present invention;

[0072] Fig.10 It is an experimental waveform diagram of the change of the output voltage reference value under the control method of the present invention;

[0073] Fig.11 It is a schematic diagram of the overall structure of the controller with a DC microgrid structure. DETAILED DESCRIPTION

[0074] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] The present invention proposes a passive model predictive control method for a DC microgrid based on predefined time feedforward compensation. The method proposes an adaptive control strategy with strong robustness, current limiting and fast voltage regulation functions. The proposed control method mainly combines passive control, model predictive control and observer technology. Firstly, a predefined time nonlinear disturbance observer is designed to accurately estimate the time-varying and unknown load power and input voltage within a predetermined time. Then, an adaptive passive model prediction technology is proposed based on the estimated state variables to control a boost converter with a constant power load, thereby significantly improving the robustness and dynamic response characteristics of the controller. In addition, the proposed nonlinear control loop avoids the complexity of obtaining nominal parameters and multi-degree-of-freedom parameter adjustment, and thus has a relatively simple and practical control structure.

[0076] Specifically, combined Figure 1-Figure 11 The present invention proposes a DC microgrid passive model predictive control method based on predefined time feedforward compensation. The control method firstly estimates and feedforwards the external disturbance online by designing a predefined time nonlinear disturbance observer, which is specifically used to accurately estimate the time-varying and unknown load power and input voltage within a predetermined time; then, an adaptive passive model predictive controller APMPC consisting of a passive control voltage loop and a model predictive control current loop is designed to control a DC / DC boost converter with a constant power load; finally, current limiting is achieved through a saturation current limiting link before the model predictive control current loop.

[0077] The present invention is based on Figure 1 The general DC microgrid structure and its equivalent circuit are shown in Figure 1. The source bus is fed by renewable energy, distributed generation, energy storage system and AC grid, while the DC / DC boost converter is used to increase the voltage and provide energy to the load bus. Resistive load and constant power load as load absorb electric energy from the load bus, while the negative impedance characteristic of constant power load is not conducive to system stability. Therefore, it is necessary to design a suitable controller for the interface converter to achieve stable operation of the DC microgrid.

[0078] The method of the present invention firstly needs to mathematically model the DC / DC boost converter. Without loss of generality, the average dynamic model of the DC / DC boost converter is expressed as:

[0079]

[0080] Among them, L, C, E, v o ,i L , μ, R and P CPLare the inductance value, output side capacitance value, input voltage, output voltage, inductor current, switch duty cycle, resistive load and constant power load power of the DC / DC boost converter, δ1=E and δ2=-Po represent the state variables that need to be estimated;

[0081] In order to take the output power in the dynamic model (1) as a disturbance form and satisfy the Brunovsky standard form, the exact feedback linearization technique is used for coordinate transformation, which is expressed as

[0082]

[0083] In formula (3), E0 is the rated input voltage, ψ1 and ψ2 represent the disturbance forms under different modeling. From formulas (3) and (4), it can be seen that when the sensor is used to measure the input voltage E in real time, the disturbance form obtained after coordinate transformation is ψ1 = -P o =δ2; when the rated input voltage E0 is used, the disturbance form obtained after coordinate transformation is ψ2 = -P o +ΔEi L ≠δ2, it is impossible to accurately estimate the output power. Therefore, online estimation and feedforward compensation of input voltage are very necessary.

[0084] The input voltage in the dynamic model (1) is regarded as a disturbance and converted into a form that complies with the Brunovsky criterion, expressed as

[0085]

[0086] The control method proposed in the present invention is designed in three parts, including the design of a predefined time nonlinear disturbance observer, a passive control voltage loop and a model predictive control current loop. First, the designed predefined time nonlinear disturbance observer not only significantly improves the robustness of the controller through the predefined time characteristics, but also only requires a control parameter that is independent of the initial state to set the upper limit of the convergence time. In addition, compared with the sensor acquisition signal, it can also improve reliability and reduce system cost. Secondly, an adaptive passive control voltage loop is proposed to achieve global equilibrium point stability (v o -v ref =0). Then, an offline model predictive control current loop with a single-step prediction level is proposed, which does not require additional control parameters and is robust with the assistance of a predefined time nonlinear disturbance observer. Finally, the adaptive passive model predictive control strategy with strong robustness, current limiting and fast voltage regulation functions proposed in this invention is proposed in combination with the three parts of the design content.

[0087] The design of the predefined time nonlinear disturbance observer is as follows: In order to estimate δ1 in equation (1) within the predetermined time, a reference auxiliary variable is introduced, whose expression is:

[0088]

[0089] In the above formula, γ1 is a known positive constant; the predefined time nonlinear disturbance observer for estimating the input voltage δ1 is designed as

[0090]

[0091] In the above formula, is the estimated value of the input voltage δ1; to ensure that the designed predefined time nonlinear disturbance observer satisfies Lyapunov stability, the function Designed to

[0092]

[0093] In the above formula, To estimate the error signal, 0<ξ<1, T o1 >0, α1>0, α2>0 and T o1 represents a predefined time coefficient that can be freely set, i.e. for any time constant t>T o1 Achievable and

[0094] In order to estimate δ2 in equation (1), a reference auxiliary variable is introduced, and its expression is

[0095]

[0096] In the above formula, γ2 is a known positive constant; the predefined time nonlinear disturbance observer for estimating the output power δ2 is designed as

[0097]

[0098] In the above formula, is the estimated value of the output power δ2; to ensure that the designed predefined time nonlinear disturbance observer satisfies Lyapunov stability, the function Designed to

[0099]

[0100] In the above formula, To estimate the error signal, 0<ξ<1, T o2 >0, α1>0, α2>0 and T o2 represents a predefined time coefficient that can be freely set, i.e. for any time constant t>T o2 Achievable and

[0101] It is worth noting that the construction of the predefined time nonlinear disturbance observer for output power estimation involves the input voltage information δ1. Therefore, T o1 The selection range should be (0, T o2 ).

[0102] In the passive control voltage loop design, in order to facilitate the design of the adaptive passive model predictive controller, the dynamic model (1) is expressed in the following matrix form:

[0103]

[0104] Where G = -G T , R = R T , Formula (12) satisfies the characteristics of the Euler-Lagrange model and provides a model basis for the design of the adaptive passive model predictive controller.

[0105] The design of the adaptive passive model predictive controller is divided into two stages:

[0106] (1) Energy reshaping stage: Adopt Transform and reshape formula (11) into

[0107]

[0108] In the above formula, X d and represent the reference value and interference value of X respectively;

[0109] (2) Damping injection stage: In this stage, the virtual damping matrix is ​​injected into both ends of equation (13): To modify the system dissipation function, expressed as

[0110]

[0111] To ensure that the system injected with the virtual damping matrix satisfies Lyapunov stability, there exists a global stable equilibrium point of the system The energy function of the system is selected as follows:

[0112]

[0113] The time derivative of the energy function (16) is expressed as

[0114]

[0115] In order to make the system globally stable and convergent, the injected virtual damping coefficient R should be ensured to be v Let the matrix R p All elements are positive. Under this premise, the control law of the passive control voltage loop is designed as follows:

[0116]

[0117] Although the passive control voltage loop control law obtained by equation (18) can be operated with multiple rated parameters, the randomness and time-varying nature of the source-load side parameters will lead to steady-state deviations of the control target. Therefore, substituting the estimated values ​​of equations (6) and (9) into (18) gives:

[0118]

[0119] In addition, by applying a current limit after the current reference, peak currents during startup and disturbances can be suppressed, reducing current stress on the device.

[0120] In the process of designing the model predictive control current loop, in order to avoid the high computational burden and low robustness of the multi-step prediction of the model predictive control, the present invention proposes a single-step offline model predictive control current loop with adaptive capability. Figure 2 The inductor current waveform when the duty cycle μ is 1 or 0, the slope of the current is calculated to be

[0121]

[0122] like Figure 2 As shown, according to the inductor current i at the current sampling time L (k), the inductor current i at the next moment L The predicted value of (k+1) is expressed as

[0123]

[0124] In the above formula, T s is the switching cycle; the cost function constructed based on the single-cycle predicted current and reference current is expressed as

[0125] J=(i L,ref -i L (k+1)) 2 (twenty two)

[0126] To obtain the optimal variable, the partial derivative of μ in equation (22) is obtained:

[0127]

[0128] The calculation formula of duty cycle μ can be derived as

[0129]

[0130] Substituting the estimated value of formula (6) into (24), we can obtain

[0131]

[0132] By combining the passive control voltage loop (19) and the model predictive control current loop (25), a nonlinear controller containing only one degree of freedom parameter can be formed, and with the assistance of a predefined time nonlinear disturbance observer, a relatively simple and practical control structure and excellent robustness are provided. Combining the above controller design, the adaptive passive model predictive controller of the present invention is finally as follows: Figure 3 shown.

[0133] In order to verify the effectiveness of the method of the present invention, the following Figure 1 The DC microgrid simulation model is shown in Figure 2. Figure 3 The method of the present invention is simulated and verified. Table 1 shows the converter and controller parameters of the system. The simulation case includes verification of the predefined time convergence characteristics and stability of the method of the present invention.

[0134] Table 1 System parameter configuration

[0135]

[0136] Case 1: The parameter selection and performance of the predefined time nonlinear disturbance observer proposed in this invention are simulated and verified. The initial startup parameters of the system are input voltage 100V and constant power load 600W. The simulation waveform is as follows: Figure 4 As shown. Figure 4 It can be seen that the relevant error signal, output power and input voltage all achieve accurate convergence. In addition, the upper limit of the convergence time of the predefined time nonlinear disturbance observer satisfies the freely set gain coefficient (T o ). It is worth noting that despite the smaller T o The convergence speed is very fast, but the inevitable noise and overshoot will lead to the degradation of system performance. Therefore, it is necessary to make a trade-off between dynamic response and ripple. The final selection coefficient T o1 = 0.01 and T o2 =0.02, the simulation results show that the proposed predefined time nonlinear disturbance observer can achieve accurate estimation of state variables within the predetermined time.

[0137] Case 2: The control method proposed in the present invention is simulated to verify its ability to suppress the limit cycle oscillation caused by the constant power load. The initial startup parameter of the system is a constant power load of 600W. Figure 5The simulation waveforms of the system switching from open-loop startup to the proposed controller under 600W constant power load conditions are shown. When the system operates in an open loop with a fixed control signal μ=0.5, the system is in a limit cycle oscillation state. When switching to the control method proposed in the present invention at t=0.6s, the virtual damping injected in the passive control voltage loop ensures energy dissipation, suppresses the limit cycle oscillation state, and the inductor current and output voltage stabilize to the equilibrium point within 5ms. It is worth noting that at the switching moment, the inductor current is limited to less than 10A. The results show that the control method proposed in the present invention has the ability to stabilize constant power loads and limit inductor current overshoot.

[0138] An experimental prototype was built in the laboratory to evaluate the effectiveness and superiority of the control algorithm proposed in the present invention. The device consists of a Chroma DC power supply, a Chroma electronic load, a DC / DC boost converter, a sampling circuit and an RT-Box. The RT-Box provides a 20kHz PWM signal to the DC / DC boost converter through an embedded control algorithm and a sampling signal. The programmable DC power supply and the electronic load realize the change of the source-load side working conditions. The effectiveness of the control method proposed in the present invention, as well as the dynamic response waveforms under the cases of power step change, load step change and output voltage reference value change are verified on this system.

[0139] Firstly, a comparative study is conducted between the control method proposed in the present invention and the traditional PI linear control under various load disturbances to evaluate the robustness and stability of the proposed method against load-side disturbances.

[0140] In the first case, the system's resistance load changes from 160Ω → 80Ω → 160Ω. Figure 6 The comparative experimental waveforms of the resistance load change are described. It can be seen that the control method proposed in the present invention can quickly stabilize the bus voltage, with a voltage overshoot of 0.7V and a stabilization time of 3ms. However, when using traditional PI linear control, the voltage overshoot increases to 5V and the stabilization time increases to 50ms. In the latter case, the CPL range of the system is: 200W→600W→200W. Figure 7 The comparative experimental waveforms of constant power load changes are described. It can be seen that the voltage overshoot is 0.9V and the stabilization time is 4ms, and the proposed method still shows ideal control performance. However, when using traditional PI linear control, the voltage overshoot increases to 8V and the stabilization time increases to 60ms. The experimental results show that the control method proposed in the present invention has excellent robustness to load disturbances.

[0141] Then, a comparative study is conducted between the control method proposed in the present invention and the traditional PI linear control under input voltage disturbance to evaluate the robustness and stability of the proposed method against source-side disturbance.

[0142] In the first case, the input voltage of the system changes from 100V to 125V to 100V. Figure 8 The comparative experimental waveform of the input voltage change (100V→125V→100V) is described. It can be seen that the control method proposed in the present invention can quickly stabilize the bus voltage, with a voltage overshoot of 0.4V and a stabilization time of 3ms. However, when using traditional PI linear control, the voltage overshoot increases to 2V and the stabilization time increases to 30ms. In the latter case, the input voltage of the system changes in the range of: 100V→75V→100V. Fig. 9 The comparative experimental waveform of the input voltage change (100V→75V→100V) is described. It can be seen that the voltage overshoot is 0.4V and the stabilization time is 3ms. The proposed method still shows ideal control performance. However, when the traditional PI linear control is adopted, the voltage overshoot increases to 2.5V and the stabilization time increases to 30ms. It is worth noting that the PI controller will produce a reverse overshoot of the inductor current at the moment of disturbance, while the inductor current of the control method proposed in the present invention tends to stabilize in a relatively stable state. The experimental results show that the control method proposed in the present invention has excellent robustness to input voltage disturbances.

[0143] Finally, the variation of the output voltage reference value of the control method proposed in the present invention is studied to evaluate the dynamic tracking performance of the proposed method.

[0144] In this experiment, the converter output voltage reference value changes from 200V to 210V and 230V respectively, and the initial constant power load is fixed at 500W. Fig.10 It can be seen that the output voltage of the converter quickly tracks the new reference value within 6ms and 10ms respectively, and there is no voltage overshoot. In addition, the control method proposed in the present invention can also limit the peak value of the inductor current to a preset value i L,max =10A. In addition, since the load end is pure CPL, the steady-state inductor current remains unchanged, while the output current decreases accordingly. The experimental results show that the control method proposed by the present invention has good voltage regulation capability.

[0145] The overall structure diagram of the controller with a DC microgrid structure of the present invention is as follows Fig.11As shown, it consists of a converter topology and a corresponding controller. The control method proposed in the present invention mainly combines passive control, model predictive control and observer technology. First, a predefined time nonlinear disturbance observer is designed to accurately estimate the time-varying and unknown load power and input voltage within a predetermined time. Then, an adaptive passive model prediction technology is proposed based on the estimated state variables to control the boost converter with a constant power load, thereby significantly improving the robustness and dynamic response characteristics of the controller. In addition, the proposed nonlinear control feedback loop avoids the complexity of obtaining nominal parameters and multi-degree-of-freedom parameter adjustment, and therefore has a relatively simple and practical control structure. In order to further evaluate the superiority of the proposed control method, the present invention conducts simulation and experimental comparative studies. The results show that the control method proposed in the present invention achieves the control target under various complex working conditions and has excellent robustness.

[0146] The present invention proposes an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a DC microgrid passive model predictive control method based on predefined time feedforward compensation are implemented.

[0147] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of a DC microgrid passive model predictive control method based on predefined time feedforward compensation.

[0148] The memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DRRAM). It should be noted that the memory of the method described in the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0149] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disc (SSD)).

[0150] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in a processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0151] It should be noted that the processor in the embodiment of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor can be combined and performed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0152] The above is a detailed introduction to the DC microgrid passive model predictive control method based on predefined time feedforward compensation proposed in the present invention. This article uses specific examples to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A DC microgrid passive model predictive control method based on predefined time feedforward compensation, characterized in that: The control method firstly designs a predefined time nonlinear disturbance observer to perform online estimation and feedforward compensation for external disturbances, which is specifically used to accurately estimate time-varying and unknown load power and input voltage within a predetermined time; then, an adaptive passive model predictive controller APMPC consisting of a passive control voltage loop and a model predictive control current loop is designed to control a DC / DC boost converter with a constant power load; finally, current limiting is achieved through a saturation current limiting link before the model predictive control current loop.

2. The method according to claim 1, characterized in that The DC / DC boost converter is mathematically modeled. Without loss of generality, the average dynamic model of the DC / DC boost converter is expressed as: Among them, L, C, E, v o ,i L , μ, R and P CPL are the inductance value, output side capacitance value, input voltage, output voltage, inductor current, switch duty cycle, resistive load and constant power load power of the DC / DC boost converter, δ1=E and δ2=-Po represent the state variables that need to be estimated; In order to take the output power in the dynamic model (1) as a disturbance form and satisfy the Brunovsky standard form, the exact feedback linearization technique is used for coordinate transformation, which is expressed as In formula (3), E0 is the rated input voltage, ψ1 and ψ2 represent the disturbance forms under different modeling respectively; The input voltage in the dynamic model (1) is regarded as a disturbance and converted into a form that complies with the Brunovsky criterion, expressed as 3. The method according to claim 2, characterized in that The design of the predefined time nonlinear disturbance observer is as follows: In order to estimate δ1 in equation (1) within the predetermined time, a reference auxiliary variable is introduced, whose expression is: In the above formula, γ1 is a known positive constant; the predefined time nonlinear disturbance observer for estimating the input voltage δ1 is designed as In the above formula, is the estimated value of the input voltage δ1; to ensure that the designed predefined time nonlinear disturbance observer satisfies Lyapunov stability, the function Designed to In the above formula, To estimate the error signal, 0<ξ<1, T o1 >0, α1>0, α2>0 and T o1 represents a predefined time coefficient that can be freely set, i.e. for any time constant t>T o1 Achievable and In order to estimate δ2 in equation (1), a reference auxiliary variable is introduced, and its expression is In the above formula, γ2 is a known positive constant; the predefined time nonlinear disturbance observer for estimating the output power δ2 is designed as In the above formula, is the estimated value of the output power δ2; to ensure that the designed predefined time nonlinear disturbance observer satisfies Lyapunov stability, the function Designed to In the above formula, To estimate the error signal, 0<ξ<1, T o2 >0, α1>0, α2>0 and T o2 represents a predefined time coefficient that can be freely set, i.e. for any time constant t>T o2 Achievable and 4. The method according to claim 3, characterized in that In the passive control voltage loop design, in order to facilitate the design of the adaptive passive model predictive controller, the dynamic model (1) is expressed in the following matrix form: Where G = -G T , R = R T , Formula (12) satisfies the characteristics of the Euler-Lagrange model and provides a model basis for the design of the adaptive passive model predictive controller.

5. The method according to claim 4, characterized in that The design of the adaptive passive model predictive controller is divided into two stages: (1) Energy reshaping stage: adopt Transform and reshape formula (11) into In the above formula, X d and represent the reference value and interference value of X respectively; (2) Damping injection stage: In this stage, the virtual damping matrix is ​​injected into both ends of equation (13): To modify the system dissipation function, expressed as 6. The method according to claim 5, characterized in that To ensure that the system injected with the virtual damping matrix satisfies Lyapunov stability, there exists a global stable equilibrium point of the system The energy function of the system is selected as follows: The time derivative of the energy function (16) is expressed as 7. The method according to claim 6, characterized in that The control law of the passive control voltage loop is designed as follows: Substituting the estimated values ​​of equations (6) and (9) into (18) we can express it as 8. The method according to claim 7, characterized in that In the process of model predictive control current loop design, the average state model (1) and the inductor current waveform when the duty cycle μ = 1 or 0 are combined to calculate the current slope as According to the inductor current i at the current sampling time L (k), the inductor current i at the next moment L The predicted value of (k+1) is expressed as In the above formula, T s is the switching cycle; the cost function constructed based on the single-cycle predicted current and reference current is expressed as J=(i L,ref -i L (k+1)) 2 (22) To obtain the optimal variable, the partial derivative of μ in equation (22) is obtained: The calculation formula of duty cycle μ can be derived as Substituting the estimated value of formula (6) into (24), we can obtain 9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium for storing computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images