Active-disturbance-rejection control method for energy storage converter based on observer bandwidth adjustment

Abstract

The invention provides an energy storage converter active-disturbance-rejection control method based on observer bandwidth adjustment, and belongs to the technical field of micro-grid control, and the method comprises the steps: obtaining a transfer function of a second-order linear active-disturbance-rejection controller from input to output; determining a bandwidth range of a linear expansion state observer in a second-order linear active disturbance rejection controller in a system stable working domain based on the transfer function; obtaining an estimation error based on the estimated and actual voltage values, and dynamically adjusting the bandwidth parameter of the observer according to the estimation error and the bandwidth range; setting the high-frequency gain of a second-order linear active-disturbance-rejection controller according to the operating parameters of the converter; and the controller for dynamically adjusting the bandwidth is used for controlling the energy storage converter. The method has the advantages that the bandwidth of the observer is dynamically increased to be close to the maximum safety value in the photovoltaic power fluctuation or load sudden change stage through a self-adaptive observer bandwidth adjusting mechanism, and the dynamic convergence process of the system is accelerated; and the bandwidth is automatically reduced to the minimum safety value in the steady-state operation stage, and the steady-state ripple amplitude is greatly reduced.

Description

Technical Field

[0001] This invention belongs to the field of microgrid control technology, specifically relating to an active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment. Background Technology

[0002] Currently, microgrid systems, composed of distributed renewable energy generation, dynamic loads, and energy storage units, are rapidly developing and gradually becoming an important component of modern power systems. Especially in remote agricultural and pastoral areas and islands far from the power grid, isolated solar-storage microgrids provide a stable and uninterrupted 24 / 7 green power supply, ensuring reliable power for local residents' lives and production. However, isolated solar-storage microgrids have limited capacity, and disturbances such as photovoltaic power fluctuations and load fluctuations can significantly impact the system's safe and stable power supply, affecting power quality and, in severe cases, even causing the entire system to malfunction. Since the maximum photovoltaic power generation and load power are uncontrollable, energy storage units become the most flexible adjustment resource in the microgrid. Therefore, in-depth research is needed on the control of energy storage converters, a crucial component of isolated microgrids, to enhance their anti-disruption capabilities and ensure the safe and stable operation of the entire isolated microgrid system.

[0003] Energy storage converters often employ traditional PID control, which is simple in structure and easy to tune parameters. However, it uses a passive control strategy for error feedback, making it impossible to predict or estimate the error between the actual and setpoint values ​​in advance. In isolated microgrid systems, disturbance signals can cause significant voltage fluctuations, severely impacting power quality. To enhance disturbance immunity, some energy storage converters currently utilize Active Disturbance Rejection Control (ADRC) technology to assess and compensate for uncertain operating conditions and disturbances in real time, significantly improving the system's anti-interference capability under specific conditions. However, traditional ADRC uses fixed bandwidth parameters, making it difficult to balance the different needs of dynamic and steady-state operation, and failing to achieve a balance between fast convergence and steady-state accuracy. To achieve fast convergence in the dynamic characteristics of the energy storage converter, traditional ADRC requires fixed high bandwidth parameters. However, high bandwidth amplifies system measurement noise, leading to significant fluctuations in the estimation results, especially in steady-state conditions where noise has a more pronounced impact. On the other hand, to improve steady-state accuracy, traditional ADRC requires lower bandwidth parameters, but low bandwidth parameters reduce the dynamic response speed of the energy storage converter, slowing down the system error adjustment speed. Therefore, improvements to traditional ADRC control are essential to achieve both fast convergence and steady-state accuracy. Summary of the Invention

[0004] In view of the above, this invention addresses the shortcomings of existing technologies by providing an active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment. To solve the aforementioned technical problems, the technical solution adopted by this invention includes: obtaining the transfer function of a second-order linear active disturbance rejection controller from input to output; determining the bandwidth range of the linear extended state observer in the system's stable operating domain based on the transfer function; obtaining the estimation error based on the estimated voltage value and the actual voltage value; dynamically adjusting the bandwidth parameter of the linear extended state observer according to the estimation error and the bandwidth range; tuning the high-frequency gain of the second-order linear active disturbance rejection controller according to the operating parameters of the energy storage converter; and using the dynamically adjusted bandwidth second-order linear active disturbance rejection controller to control the energy storage converter and compensate for system disturbances.

[0005] Furthermore, the transfer function of the second-order linear active disturbance rejection controller from input to output is expressed as:

[0006]

[0007] Where b0 is the high-frequency gain of the second-order linear active disturbance rejection controller; τ1 and τ2 are damping coefficients; u(s) is the input of the second-order linear active disturbance rejection controller; y(s) is the output of the second-order linear active disturbance rejection controller; s is the differential operator introduced by the Laplace transform; ω o is the bandwidth of the linearly extended state observer.

[0008] Furthermore, the method for determining the bandwidth range of the linear extended state observer in the stable operating region of a second-order linear active disturbance rejection controller based on the transfer function includes: deriving the stable operating region of the system based on the obtained transfer function using the Routh-Hurwitz criterion and the Nyquist criterion, thereby obtaining the minimum bandwidth ω of the stable operating region of the system. omin and maximum bandwidth ω omax .

[0009] Furthermore, the method for obtaining the estimation error based on the estimated voltage value and the actual voltage value, and dynamically adjusting the bandwidth parameter of the linearly extended state observer according to the estimation error and the bandwidth range, includes: an adaptive observer bandwidth adjustment strategy as follows: Where k is the gain constant; tanh(x) is the hyperbolic tangent function. The voltage value is estimated, and e is the actual voltage value.

[0010] Furthermore, the method for tuning the high-frequency gain of the second-order linear active disturbance rejection controller based on the operating parameters of the energy storage converter includes: the high-frequency gain is the ratio of the duty cycle of the controllable switching device of the energy storage converter to the bus capacitance on the energy storage side.

[0011] Furthermore, the duty cycle of the controllable switching device is the ratio of the low-voltage side voltage of the energy storage converter to the DC bus voltage.

[0012] Furthermore, a second-order linear active disturbance rejection controller with dynamically adjustable bandwidth is used for energy storage converter control. The methods for compensating for system disturbances include: when the photovoltaic power of the energy storage converter fluctuates or the load changes, the low-voltage side voltage and DC bus voltage of the energy storage converter are remeasured to update the high-frequency gain, and the dynamically adjusted observer bandwidth parameter ω is used. o The updated high-frequency gain input second-order linear active disturbance rejection controller compensates for system disturbances.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0014] 1. Through an adaptive observer bandwidth adjustment mechanism, the observer bandwidth is dynamically increased to near the maximum safe value during photovoltaic power fluctuations or load abrupt changes, significantly enhancing the observer's real-time tracking capability of disturbances in the extended state and accelerating the system's dynamic convergence process. During steady-state operation, the bandwidth is automatically reduced to the minimum safe value, effectively suppressing the impact of high-frequency measurement noise on the bus voltage and significantly reducing the steady-state ripple amplitude. This mechanism overcomes the technical contradiction between the incompatibility of dynamic response speed and steady-state accuracy inherent in traditional fixed bandwidth parameters, achieving synergistic optimization of both.

[0015] 2. The high-frequency gain parameters are automatically calculated by measuring the physical quantities of the circuit in real time, without the need for manual debugging; the observer bandwidth is dynamically adjusted according to the estimation error, and only three boundary parameters, namely the bandwidth range of the stable operating domain of the system and the gain constant, need to be preset; this design greatly reduces the complexity of on-site debugging and solves the engineering pain point of traditional active disturbance rejection control parameters relying on experience tuning, and is especially suitable for unmanned operation and maintenance scenarios of photovoltaic-storage microgrids in remote areas.

[0016] 3. Suppression of bus voltage ripple directly reduces the switching losses and thermal stress of power electronic devices; accelerated dynamic response reduces energy loss during photovoltaic power fluctuations; at the same time, stable bus voltage provides high-quality power supply for sensitive loads, improving energy utilization and power supply reliability. Attached Figure Description

[0017] The present invention will now be described in further detail with reference to the accompanying drawings.

[0018] Figure 1 : A schematic diagram of the second-order linear active disturbance rejection control in this invention;

[0019] Figure 2 The root locus diagram of the traditional LADRC control transfer function in this invention;

[0020] Figure 3 Bode plot of traditional LADRC control transfer function in this invention;

[0021] Figure 4 : A schematic diagram of the bidirectional DC-DC converter in this invention;

[0022] Figure 5 The structure and control diagram of the isolated photovoltaic-storage microgrid in this invention;

[0023] Figure 6 : Simulation model diagram of the isolated photovoltaic-storage microgrid in this invention;

[0024] Figure 7 The DC power supply output current curve in this invention;

[0025] Figure 8 The experimental voltage waveforms of groups A, B, and C in this invention are shown. Detailed Implementation

[0026] To better understand the present invention, the content of the invention is further clearly illustrated below with reference to embodiments and accompanying drawings. However, the scope of protection of the present invention is not limited to the embodiments described below. Numerous specific details are set forth in the following description to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the present invention can be practiced without one or more of these details.

[0027] Example 1: See Figure 1-8 This embodiment provides an active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment, which includes:

[0028] S1, see reference Figure 1 A second-order linear active disturbance rejection control (LADRC) framework is constructed, which includes a linear extended state observer (LESO) and a linear state error feedback control (LSEF).

[0029] S2. Construct the state-space equations of LADRC and LESO, construct the state feedback controller and update the state equations; obtain the transfer function of the LADRC controller from input u(t) to output y(t) based on the updated state equations, which includes:

[0030] For second-order linear active disturbance rejection control, the controlled object model is:

[0031]

[0032] Among them, u, f(y,ω,t) represents the input, output, and external disturbance signals, respectively. The parameter b0 > 0 is the high-frequency gain and is a known quantity, implying that the system has no internal disturbances. Design an output feedback controller to ensure that the output signal y tracks the input signal reference value ref. Introduce state variables x1 = y, ... And the extended state x3=f, transforming equation (1) into a state-space equation:

[0033]

[0034] in, This is the system state transition matrix, which reflects the coupling relationship between state variables x1, x2, and x3. The position of the zero element indicates that the system has no internal coupling. To control the input matrix, the control input u is mapped to the state differential equation, reflecting the influence of the control input u on the state variables. It only acts on state x2, reflecting the effect of the control quantity u on the output acceleration. The influence of control input u is that it only affects state x2 through matrix B; C = [1 0 0] is the disturbance input matrix; Represented as x 1-n The differential form of D; D = [0 0 1] T , where is the perturbation input matrix, and the perturbation rate of change f˙ is mapped to the state differential equation.

[0035] Based on equation (2), a linear extended state observer is constructed, the expression of which is:

[0036]

[0037] Where z1, z2, and z3 represent the estimated values ​​of x1, x2, and x3, respectively, and represent the perturbation value of y, which can be written as:

[0038]

[0039] L is the gain of the linearly extended state observer, which is:

[0040] L = [β1 β2 β3] T (5)

[0041] • The state-space equation of the LADRC state observer is derived from equations (3) and (4):

[0042] Where β1, β2, and β3 are the parameters to be adjusted, β1 represents the convergence rate of the control state estimate z1 (bus voltage y), and β2 represents the convergence rate of the control state estimate z2 (voltage change rate). The convergence speed of ) ; the convergence speed of the extended state estimator z3 (total perturbation f) controlled by β3; this application adopts a single-parameter structure, introduces a positive parameter ω0, i.e., the observer bandwidth, so that the observer error dynamics are completely dominated by ω0, realizing the cascaded pole placement of the integrator, and setting the gain as:

[0043] β1=3ω0, β2=3ω0 2 ,β3=ω0 3 (7)

[0044] The state feedback controller is designed as follows:

[0045]

[0046] Where, k p and k d Two parameters to be adjusted are the feedback gains of the controller and the observer, respectively, satisfying k p =2ω0,k d =ω0 2 ;

[0047] Substituting equation (7) into equation (6), we get

[0048]

[0049] Based on equation (9), a Laplace transform is performed on the state-space equation, and the operator s is introduced to obtain:

[0050]

[0051] Let u(s), r(s), and y(s) correspond to the Laplace transforms of signals u, r, and y, respectively. Then we have:

[0052]

[0053] Further simplification yields:

[0054]

[0055] Equation (12) is the transfer function of the LADRC controller from input u(t) to output y(t).

[0056] S3, see reference Figure 2-3 The dynamic and steady-state characteristics of the LADRC transfer function model are analyzed through transfer function analysis, and the fixed bandwidth parameter ω of the traditional LADRC is experimentally verified. o The dynamic-steady-state contradiction.

[0057] Based on the single-parameter relationship β1=3ω0,β2=3ω0 2 ,β3=ω0 3 In the transfer function In the equation, ω0 simultaneously governs the dynamic term s. 2 coefficients and steady-state terms The molecule constant term; when ω0 is fixed, the dynamic and steady-state characteristics of the transfer function model can be described as follows:

[0058] ω0 increases → molecule s 2 As the term coefficient increases, the dynamic response speed improves, the resonance peak increases, which in turn reduces noise immunity and increases steady-state ripple.

[0059] Decreasing ω0 leads to improved steady-state accuracy and a lower resonance peak value; the molecule constant term ω o 3 Decreasing the step response slope reduces the rise time slope and delays the dynamic response.

[0060] This creates a performance conflict where dynamic response and steady-state accuracy cannot be simultaneously achieved.

[0061] Draw the root locus and Bode plot of the system transfer function according to equation (12), as follows: Figure 2 and Figure 3 As shown. In equation (12), b0 is 5, τ1 is 10, and τ2 is 25.

[0062] As the low-frequency gain increases with ω0, the steady-state error under a unit step input decreases, indicating that the system's ability to track low-frequency signals is enhanced. In terms of dynamic response, as ω0 increases, the root locus poles separate from the real axis into conjugate complex roots, and the damping ratio transitions from critical damping to underdamping. Correspondingly, the overshoot of the step response increases and the settling time is shortened. At the same time, the Bode plot shows that as the observer bandwidth increases, the resonant peak value increases synchronously. The system's dynamic tracking performance improves, but its noise immunity decreases and the steady-state ripple increases.

[0063] Based on the above analysis, it can be concluded that as the observer bandwidth ω0 increases within a reasonable range, the dynamic response speed of the system is significantly improved, but the sensitivity to high-frequency noise increases simultaneously, resulting in an increase in ripple amplitude during steady-state operation. Specifically, during the dynamic adjustment phase, the linear expansion state observer needs to rely on high observation bandwidth parameters to achieve accurate estimation of disturbances, and the increase of such parameters will directly exacerbate the increase in steady-state ripple.

[0064] S4. Determine the stable operating region of the observer system using the transfer function.

[0065] Based on the transfer function model of the linear active disturbance rejection controller, the root distribution stability of the closed-loop system characteristic equation is analyzed by the Routh-Hurwitz criterion to ensure that all poles are located in the left half of the complex plane. At the same time, the stability of the open-loop frequency response is verified by combining the Nyquist criterion to determine the frequency domain stability boundary where the system phase margin is greater than 45° and the gain margin approaches positive infinity.

[0066] Combining root locus analysis and Bode plot frequency domain analysis, it can be seen that the observer bandwidth ω0 has a significant impact on the dynamic and steady-state characteristics of the transfer function. The root locus shows that under system parameters of b0 = 5, τ1 = 10, and τ2 = 25, in terms of steady-state characteristics, all closed-loop poles are located in the left half-plane of the complex plane, satisfying the Hurwitz criterion, indicating system stability. Furthermore, the phase margin in the Bode plot is greater than 45°, and the gain margin approaches positive infinity, both indicating that the system has strong anti-interference capability.

[0067] By combining the Routh-Hurwitz criterion with the Nyquist criterion and integrating the constraints of both criters, the parameter domain for stable system operation is derived, thereby quantifying the minimum safe value ω of the observer bandwidth. omin and maximum safety value ω omax ;

[0068] S5. Based on the observer bandwidth parameter ω o The system's stable operating domain is used to obtain an adaptive observer bandwidth adjustment strategy, which dynamically adjusts the observer bandwidth parameter ω based on the estimation error. o It includes:

[0069] The adaptive bandwidth of the second-order adaptive extended state observer can be designed as follows:

[0070]

[0071] Where, ω omin ω omax Let and k be the minimum value of ω0, the maximum value of ω0, and the gain constant, respectively; tanh(x) is the hyperbolic tangent function. To estimate the voltage value, e is the actual voltage value. As shown in equation (13), the adaptive bandwidth of the second-order linear extended state observer is adjusted in real time according to the estimation error. During the dynamic transition process, when the estimation error is large, the bandwidth of the adaptive extended state observer is high to ensure fast convergence; in steady state, when the estimation error is small, the bandwidth of the adaptive extended state observer is low, thereby reducing the ripple in steady state.

[0072] S6. The updated high-frequency gain and dynamically adjusted bandwidth are input to the LADRC controller to compensate for system disturbances and maintain bus voltage stability. This includes:

[0073] The structure of a bidirectional DC-DC energy storage converter is as follows: Figure 4 As shown, take the bus voltage u dc Inductor current i L When the bidirectional converter losses and DC-side inductance losses are ignored as state variables, the system model is as follows:

[0074]

[0075] Among them, iL i o These are the energy storage inductor current and the equivalent load current of the DC system, respectively; u b and u dc These are the low-voltage side voltage and the DC bus voltage, respectively; resistor R L The inductor's equivalent resistance; L and C are the energy storage side inductance and bus capacitance, respectively; α = u b / u dc dt represents the duty cycle of the controllable switching device, and di represents the derivative with respect to time. L This represents the differentiation of the inductor current.

[0076] Due to the DC bus voltage u dc Specific inductor current i L The change is much slower; during the control process, the DC bus voltage u dc It remains almost unchanged and can be considered a constant value.

[0077] Take i L u dc , α, i O u b The steady-state component is i Le u dce α e i oe u be Based on equations (14) and (15), the steady-state operating point of the bidirectional buck-boost converter is obtained as follows:

[0078] u be -α e u dce -R L i Le =0 (16)

[0079] α e i Le -i oe =0 (17)

[0080] The duty cycle α at the static operating point is obtained from equation (16). e for:

[0081] α e =u be / u dce (18)

[0082] to i L u dc , α, i O u b Introducing a perturbation, we get:

[0083]

[0084] Substituting equation (19) into equation (15), we obtain the bus voltage u. dc With inductor current i L The equation of change is

[0085]

[0086] Multiply both sides of equation (20) by u dc ,get:

[0087]

[0088] Near the steady-state equilibrium point, linearizing equation (21) yields:

[0089]

[0090] According to equation (22), we get:

[0091]

[0092] Let the outer loop input Output If h is the disturbance, then we get:

[0093]

[0094] in, For load current disturbance;

[0095] According to equation (18), we get:

[0096]

[0097] When photovoltaic power fluctuates or load changes, u is remeasured. b, u dc Update b0 to dynamically adjust ω o The updated b0 input LADRC controller compensates for system disturbances.

[0098] S7. Verify the proposed active disturbance rejection control strategy for energy storage converters based on adaptive observer bandwidth adjustment:

[0099] according to Figure 5 The isolated photovoltaic-storage microgrid structure and control diagram shown are illustrated. A system was built using Matlab software, as follows: Figure 6The simulation model of the isolated photovoltaic-storage microgrid system shown is composed of a DC current source, an energy storage DC-DC converter, batteries, a DC bus voltage regulator, and loads. For simplicity, the DC current source is used to replace the photovoltaic modules. The specific simulation parameters are shown in Table 1. To verify that the proposed strategy can ensure the stable operation of the isolated microgrid system under different operating conditions, a sudden drop in DC source output power is set at 5 seconds to simulate the reduction in photovoltaic array power output caused by environmental factors, and simulation analysis is performed.

[0100] Table 1 Simulation Model Parameters for Photovoltaic Energy Storage Islanded Microgrid

[0101] parameter numerical values <![CDATA[DC bus voltage V s > 400V <![CDATA[DC bus voltage stabilizing capacitor C dc > 0.008F <![CDATA[DC bus voltage stabilizing inductor L dc > 0.005F <![CDATA[DC / DC converter DC side voltage stabilizing capacitor C dc > 0.002F <![CDATA[DC / DC converter DC-side voltage stabilizing inductor L dc > 0.01F <![CDATA[LADRC observer gain coefficient k d > 5 <![CDATA[LADRC observer bandwidth ω0]]> 50 <![CDATA[DC / DC converter DC proportional control parameter k dip > 1 <![CDATA[Integral control parameter k in the current loop of a DC / DC converter dii > 0.08

[0102] At 5 seconds, the DC current source output is controlled to suddenly drop from 6A to 5.5A to simulate a sudden drop in output of the photovoltaic energy storage microgrid due to external factors. The DC current source output is as follows: Figure 7 As shown in the figure. Three sets of experiments were conducted. Group A used traditional LADRC control with an observer bandwidth ω0 of 5; Group B used traditional LADRC control with an observer bandwidth ω0 of 20; and Group C used the adaptive observer bandwidth adjustment LADRC control strategy proposed in this application, where the observer bandwidth ω0 adaptively adjusts according to the system operating state. The simulation results for Group A are shown in the figure. Figure 8 As shown by the green curve, the simulation results for group B are as follows: Figure 8 As shown by the red curve, the simulation results for group C are as follows: Figure 8 As shown by the blue curve.

[0103] See Figure 8 Observations revealed that, under stable operating conditions, Group B, due to its larger observer bandwidth, exhibited a larger steady-state ripple than Groups A and C, nearly twice that of the other two groups. Group A, with its smaller observer bandwidth parameter, showed a significantly slower dynamic response and adjustment speed compared to the other two groups during the 5-second dynamic response. While Group B had a faster adjustment speed, its overshoot was significantly higher than that of Groups A and C. Group C, employing the adaptive observer bandwidth LADRC control strategy, showed a smaller steady-state ripple than Group B during stable operation, comparable to Group A with its smaller observer bandwidth. During the 5-second dynamic response, due to the real-time adjustment of the adaptive observer bandwidth, Group C's overshoot was lower than both Groups A and B with fixed observer bandwidth, and its adjustment speed was similar to Group B with its high observer bandwidth, but faster than Group A with its low observer bandwidth.

[0104] Technical effects of this embodiment:

[0105] 1. When the energy storage converter adopts the adaptive observer bandwidth LADRC control strategy proposed in this application, the adaptive observer bandwidth adjustment mechanism dynamically increases the observer bandwidth to near the maximum safe value ω during photovoltaic power fluctuations or load abrupt changes. omaxThis significantly enhances the extended state observer's real-time tracking capability of disturbances, accelerating the system's dynamic convergence process; and automatically reduces the bandwidth to the minimum safe value ω during steady-state operation. omin This mechanism can reduce steady-state ripple by using low observer bandwidth under stable operating conditions, and accelerate response speed by using high observer bandwidth during dynamic response. It effectively suppresses the impact of high-frequency measurement noise on bus voltage and significantly reduces steady-state ripple amplitude. This mechanism breaks through the technical contradiction that traditional fixed bandwidth parameters cannot be compatible with dynamic response speed and steady-state accuracy, and achieves synergistic optimization of the two, providing a reliable guarantee for the safe and stable power supply of isolated photovoltaic-storage microgrid systems.

[0106] 2. The high-frequency gain parameter b0 is automatically calculated through real-time measurement circuit physical quantities, requiring no manual adjustment; the observer bandwidth ω0 is dynamically adjusted according to the estimation error, requiring only ω to be preset. omax ω omin The design incorporates three boundary parameters: k, t, and k. This design significantly reduces the complexity of on-site commissioning and addresses the engineering pain point of traditional active disturbance rejection control parameters relying on experience-based tuning. It is particularly suitable for unmanned operation and maintenance scenarios of photovoltaic-storage microgrids in remote areas.

[0107] 3. Suppression of bus voltage ripple directly reduces the switching losses and thermal stress of power electronic devices; accelerated dynamic response reduces energy loss during photovoltaic power fluctuations; at the same time, stable bus voltage provides high-quality power supply for sensitive loads, improving energy utilization and power supply reliability.

[0108] 4. Bandwidth security boundary ω based on Routh-Hurwitz and Nyquist dual criteria tuning omax -ω omin This ensures that all poles of the closed-loop system are strictly located in the left half of the complex plane, with a phase margin always greater than 45° and a gain margin approaching infinity, thus enhancing system stability in both the time and frequency domains. Under strong disturbance conditions such as sudden drops in photovoltaic power or step changes in load, the maximum deviation of the bus voltage is limited to within a safe threshold, significantly improving the anti-interference capability and long-term operational robustness of the islanded microgrid in complex operating environments.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solutions of the present invention, as long as they do not depart from the spirit and scope of the technical solutions of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A method for active disturbance rejection control of an energy storage converter based on observer bandwidth adjustment, characterized in that, include: Obtain the transfer function of the second-order linear active disturbance rejection controller from input to output; The bandwidth range of the linear extended state observer in the stable operating domain of the second-order linear active disturbance rejection controller is determined based on the transfer function. The estimation error is obtained based on the estimated voltage value and the actual voltage value, and the bandwidth parameter of the linear extended state observer is dynamically adjusted according to the estimation error and the bandwidth range. The high-frequency gain of the second-order linear active disturbance rejection controller is tuned according to the operating parameters of the energy storage converter; A second-order linear active disturbance rejection controller with dynamically adjustable bandwidth is used to control the energy storage converter and compensate for system disturbances.

2. The active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment as described in claim 1, characterized in that, The transfer function of a second-order linear active disturbance rejection controller from input to output is expressed as: Where b0 is the high-frequency gain of the second-order linear active disturbance rejection controller; τ1 and τ2 are damping coefficients; u(s) is the input of the second-order linear active disturbance rejection controller; y(s) is the output of the second-order linear active disturbance rejection controller; s is the differential operator introduced by the Laplace transform; ω o is the bandwidth of the linearly extended state observer.

3. The active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment as described in claim 2, characterized in that, Methods for determining the bandwidth range of a linear extended state observer in the stable operating domain of a second-order linear active disturbance rejection controller based on the transfer function include: Based on the obtained transfer function, the stable operating region of the system is derived using the Routh-Hurwitz criterion and the Nyquist criterion, thus obtaining the minimum bandwidth ω of the stable operating region of the system. omin and maximum bandwidth ω omax .

4. The active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment as described in claim 3, characterized in that, A method for obtaining an estimation error based on an estimated voltage value and an actual voltage value, and dynamically adjusting the bandwidth parameter of a linearly extended state observer based on the estimation error and the bandwidth range, includes: The adaptive observer bandwidth adjustment strategy is as follows: Where k is the gain constant; tanh(x) is the hyperbolic tangent function. The voltage value is estimated, and e is the actual voltage value.

5. The active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment as described in claim 4, characterized in that, Methods for tuning the high-frequency gain of a second-order linear active disturbance rejection controller based on the operating parameters of the energy storage converter include: The high-frequency gain is the ratio of the duty cycle of the controllable switching device in the energy storage converter to the capacitance of the bus on the energy storage side.

6. The active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment as described in claim 5, characterized in that, The duty cycle of the controllable switching device is the ratio of the low-voltage side voltage of the energy storage converter to the DC bus voltage.

7. The active disturbance rejection control method for energy storage converters based on observer bandwidth adjustment as described in claim 6, characterized in that, Methods for compensating for system disturbances by using a second-order linear active disturbance rejection controller with dynamically adjustable bandwidth for energy storage converter control include: When the photovoltaic power of the energy storage converter fluctuates or the load changes, the low-voltage side voltage and DC bus voltage of the energy storage converter are remeasured to update the high-frequency gain, and the observer bandwidth parameter ω is dynamically adjusted. o The updated high-frequency gain input second-order linear active disturbance rejection controller compensates for system disturbances.

Images