Voltage control method of photovoltaic grid-connected system

Abstract

The invention belongs to the technical field of power grid voltage control, and particularly discloses a voltage control method of a photovoltaic grid-connected system, which comprises the following steps of: setting an optimization process comprising a duty ratio control scheme; all duty ratio control schemes and corresponding input power are optimized through an optimization process; outputting and controlling a direct current side according to the duty ratio control scheme with the highest input power; active power and reactive power are determined in real time through a duty ratio control scheme; resolving a coupling relationship between the active power and the reactive power in the two-axis coordinate system to obtain a voltage state variable and a current state variable on each coordinate axis; calculating a voltage estimation value and a current estimation value on each coordinate axis through a voltage state variable and a current state variable under the condition of considering a disturbance error, and controlling an alternating current side according to the voltage estimation value and the current estimation value; the method has the advantages that the energy efficiency and stability of the direct current side are enhanced through multi-scheme optimization and dynamic selection, transient events are quickly responded, and high-quality electric energy supply and power grid stability are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of grid voltage control, and more particularly, to a voltage control method for a photovoltaic grid-connected system. Background Art

[0002] Photovoltaic systems often face problems of voltage imbalance and sudden drops during grid connection operations, which can affect the stability and power generation efficiency of the system. Voltage imbalance may lead to equipment damage and power loss, while voltage sudden drops will affect the reliability of the grid and the photovoltaic system.

[0003] Therefore, a voltage control method for a photovoltaic grid-connected system is proposed to solve the above-mentioned problems. Summary of the Invention

[0004] The present invention aims to provide a voltage control method for a photovoltaic grid-connected system to solve or improve the problems that the stability and efficiency of the photovoltaic system are often affected by voltage imbalance and sudden drops during grid connection in the above technical problems.

[0005] In view of this, a first aspect of the present invention is to provide a voltage control method for a photovoltaic grid-connected system.

[0006] The first aspect of the present invention provides a voltage control method for a photovoltaic grid-connected system, including: setting an optimization process including multiple duty cycle control schemes on the DC side of the photovoltaic grid-connected system; at each moment, optimizing all the duty cycle control schemes and corresponding input powers through the optimization process, and making each duty cycle control scheme and corresponding input power meet preset conditions; at each moment, outputting and controlling the DC side according to the duty cycle control scheme with the highest input power; on the AC side of the photovoltaic grid-connected system, determining the active power and reactive power in real time through the duty cycle control scheme output at each moment; respectively resolving the coupling relationship between the active power and the reactive power in a two-axis coordinate system to obtain the voltage state variables and current state variables on each coordinate axis; calculating the voltage estimation value and current estimation value on each coordinate axis through the voltage state variables and the current state variables considering the disturbance error, and controlling the AC side according to the voltage estimation value and the current estimation value at each moment.

[0007] In any of the above technical solutions, at each moment, the DC side and the AC side are controlled in real time according to the same duty cycle control scheme, and the duty cycle control scheme includes the three-phase switch states of the three-phase inverter between the DC side and the AC side within a unit time.

[0008] In any of the above technical solutions, the active power and the reactive power are determined through the following steps: calculating the voltage and current of the three-phase inverter in a two-axis coordinate system based on the three-phase switch states; corresponding the two coordinate axes to the active power and the reactive power respectively; and calculating the mean active power and reactive power per unit time according to the voltage and current of each coordinate axis.

[0009] In any of the above technical solutions, the step of separately resolving the coupling relationship of the active power and the reactive power in the two-axis coordinate system includes: constructing a virtual synchronous machine model for simulating a synchronous generator by the three-phase inverter; and determining the voltage loop part and the current loop part in the two-axis coordinate system according to the active power loop part and the reactive power loop part in the virtual synchronous machine model.

[0010] In any of the above technical solutions, the voltage loop part is determined through the following steps: obtaining the rate-of-change equation of the voltage with respect to time on each coordinate axis according to the active power loop part and the reactive power loop part; and constructing the first voltage state equation of the voltage loop part through the rate-of-change equation.

[0011] Separating the disturbance quantity of the photovoltaic grid-connected system in the first voltage state equation and updating it to the second voltage state equation.

[0012] In any of the above technical solutions, the rate-of-change equation includes the following formula: where, the is the rate of change of v d with respect to time t, and the v d is the component of the photovoltaic grid-connected voltage on the d-axis in the two-axis coordinate system; the is the rate of change of v q with respect to time t, and the v q is the component of the photovoltaic grid-connected voltage on the q-axis in the two-axis coordinate system; the i d is the component of the current on the d-axis; the i q is the component of the current on the q-axis; the ω is the actual angular frequency that changes due to the dynamic process under grid disturbances; the C is the capacitance value; and the ΔC is the parameter error on the capacitor.

[0013] In any of the above technical solutions, the estimated value is obtained by constructing the LESO control equation of the voltage loop part, and the construction steps of the LESO control equation of the voltage loop part include: obtaining the disturbance quantity on each coordinate axis through the rate-of-change equation; and introducing the disturbance quantity on each coordinate axis into the voltage parameter of the second voltage state equation and using it as the LESO control equation of the voltage loop part.

[0014] In any of the above technical solutions, the LESO control equation of the voltage loop part includes the following formula: Among them, the is the estimated value of the state variable of the voltage of the photovoltaic grid-connected system disturbance on the d-axis in the two-axis coordinate system; the is the estimated value of the state variable of the voltage of the photovoltaic grid-connected system disturbance on the q-axis in the two-axis coordinate system; the is the estimated value of the total voltage disturbance variable of the photovoltaic grid-connected system on the d-axis; the is the estimated value of the total voltage disturbance variable of the photovoltaic grid-connected system on the q-axis; the z vd2 is the estimated value of x vd2 , and x vd2 is the total disturbance including internal and external disturbances on the d-axis; the β v1 is a LESO gain coefficient of the voltage; the z vd1 is the estimated value of the x vd1 , and the x vd1 is the d-axis voltage; the b v is the system parameter of the photovoltaic grid-connected system; the y vd1 is the component of the output voltage in the d-axis of the input power; the z vq2 is the estimated value of x vq2 , and the x vq2 is the total disturbance including internal and external disturbances on the q-axis; the z vq1 is the estimated value of the x vq1 , and the x vq1 is the q-axis voltage; the y vq1 is the component of the output voltage in the q-axis of the input power; the β v2 is another LESO gain coefficient of the voltage.

[0015] In any of the above technical solutions, the photovoltaic grid-connected system further includes a PD controller and a dynamic voltage restorer located on the AC side, and the step of controlling the AC side according to the voltage estimated value and the current estimated value at each moment includes: calculating the output of the PD controller through the numerical values of the voltage estimated value or the current estimated value on each coordinate axis; controlling the dynamic voltage restorer through the output of the PD controller to compensate for the voltage dips or voltage surges that randomly occur on the AC side at each moment.

[0016] In any of the above technical solutions, the output of the PD controller is calculated by the following formula: Among them, the u vd1 is the output of the PD controller on the d-axis; the u vq1 is the output of the PD controller on the q-axis; the k vpis the proportionality coefficient in the PD controller; the vd is the given value of the d-axis component of the controlled voltage; the vq is the given value of the q-axis component of the controlled voltage.

[0017] Advantages of the present invention compared with the prior art:

[0018] Through the dynamic selection mode of parallel optimization of multiple schemes, the system can continuously operate close to the maximum power point under rapidly changing environmental conditions (such as light changes, temperature fluctuations), maximizing the power generation efficiency of the photovoltaic panel. Avoid the phenomenon of long-term deviation from the optimal point that may occur in traditional single-scheme control, and effectively improve the energy utilization rate and economic benefits on the DC side.

[0019] Simplify the traditional complex three-phase control problem into an independent problem in the two-axis coordinate system, providing a mathematical basis for precise control of active and reactive power. Effectively separate the active and reactive regulation paths, enhance the flexibility and accuracy of grid regulation, and contribute to ensuring power quality and grid stability during grid connection.

[0020] Maintain accurate perception of the system state under complex working conditions, enable the controller to quickly adjust strategies in the face of transient events, and improve the dynamic response ability of the system. Enhance the anti-interference ability and robustness of the system, and provide a solid data basis for voltage control, frequency control, and power quality maintenance.

[0021] The photovoltaic grid-connected system can maintain high efficiency and stability in both short-term disturbances and long-term changes, reduce power losses, ensure the safety of electrical equipment and the high-quality supply of electric energy. Provide more stable voltage and frequency support for the grid side, and contribute to the safe and reliable operation of the overall power system.

[0022] The additional aspects and advantages according to the embodiments of the present invention will become apparent in the following description part, or be learned through the practice according to the embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and be readily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0024] Figure 1 is the flow chart of the method steps of the present invention;

[0025] Figure 2 is the schematic diagram of the optimization process steps of the present invention;

[0026] Figure 3 is the topological diagram of the photovoltaic inverter grid-connected system of the present invention;

[0027] Figure 4 is the structure diagram of the VSI with an LC filter of the present invention;

[0028] Figure 5 This is the main circuit and control structure diagram of the three-phase grid-connected inverter of the present invention;

[0029] Figure 6 This is the Simulink simulation diagram of the dynamic voltage restorer of the present invention;

[0030] Figure 7 This is the schematic diagram of the grid-connected voltage response under normal operating conditions of the present invention;

[0031] Figure 8 This is the schematic diagram of the total harmonic distortion rate of the grid-connected voltage under normal operating conditions of the present invention;

[0032] Figure 9 This is the schematic diagram of the grid-connected current response under normal operating conditions of the present invention;

[0033] Figure 10 This is the schematic diagram of the total harmonic distortion rate of the grid-connected current under normal operating conditions of the present invention;

[0034] Figure 11 This is the schematic diagram of the grid-connected voltage response under voltage unbalance / sag conditions of the present invention;

[0035] Figure 12 This is the schematic diagram of the total harmonic distortion rate of the grid-connected voltage under voltage unbalance / sag conditions of the present invention;

[0036] Figure 13 This is the schematic diagram of the grid-connected current response under voltage unbalance / sag conditions of the present invention;

[0037] Figure 14 This is the schematic diagram of the total harmonic distortion rate of the grid-connected current under voltage unbalance / sag conditions of the present invention. Detailed implementation manners

[0038] In order to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described in detail below with reference to the drawings and specific implementation manners. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.

[0039] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention may be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0040] Please refer to Figures 1 - 14 , and the following describes a voltage control method for a photovoltaic grid-connected system according to some embodiments of the present invention.

[0041] As described in the background art, energy is essential for economic, social, and industrial development. It is expected that the energy consumption in the coming years will be many times higher than the current level. New energy sources have proven to be viable solutions, and grid-connected photovoltaic power plants have become the future development direction and research focus. As the interface between renewable energy and the power grid, photovoltaic grid-connected inverters play a crucial role in achieving high-efficiency power conversion and high-quality power supply. The development of photovoltaic technology and the rapid reduction of photovoltaic costs have demonstrated the potential for large-scale installation of photovoltaics globally. Therefore, photovoltaic power generation units are the main research objects at present.

[0042] Voltage imbalance / sag is one of the most common mismatch and performance loss situations in photovoltaic systems. The output will change significantly. Affected by external conditions, the output power of the photovoltaic power generation system is unstable, which in turn leads to voltage sags. The output performance of a three-phase inverter mainly depends on the modulation algorithm. Under voltage imbalance / sag conditions, an appropriate inverter control strategy determines the effectiveness of the photovoltaic system.

[0043] For complex systems and complex operating conditions, the organic combination of DC-side control and AC-side control can improve the efficiency of the photovoltaic grid-connected system. To obtain a more efficient grid-connected voltage waveform, the DC-side and AC-side controls need to work together, that is, the DC-side boost maximum power point tracking control and the AC-side VSG control strategy based on LADRC work together, and the response speeds of each link cooperate with each other to achieve the best effect. Structurally, the DC-side control and the AC-side control are executed separately. The main objectives of the DC-side controller and the AC-side controller are maximum power point tracking and controlling the grid injection power with the required power factor, respectively.

[0044] The DC-side voltage Vdc and the AC-side voltage and current are monitored in real time, and the system starts vector control to adjust the active and reactive powers. The inverter control diagram of the photovoltaic grid-connected system used in this paper is as Figure 5 shown. To achieve efficient and stable inverter voltage on the AC side, the direct-through duty ratio can be adjusted by the MPPT algorithm to adjust the VSI input voltage Vdc. The direct-through duty ratio D is obtained by the MPPT algorithm based on the detected voltage and current at the input end of the photovoltaic grid-connected system.

[0045] The maximum power point tracking MPPT technology is one of the important methods to improve the efficiency of photovoltaic power generation. Given the particularity of the mutual restriction between direct-through boost and inverter modulation, how to set the direct-through duty ratio and the inverter modulation coefficient to maintain the maximum DC-side voltage utilization rate is very crucial in the entire control system.

[0046] The virtual synchronous generator (VSG) technology introduces virtual inertia and virtual damping, which plays a supporting role in stabilizing the grid frequency. Due to the vector control technology, the VSG provides decoupled control of active and reactive power and fast dynamic characteristics, and two PI control loops, namely the external control loop and the internal current control, are required.

[0047] A control strategy for a virtual synchronous generator (VSG) based on linear active disturbance rejection control (LADRC) and an active disturbance rejection control (ADRC) dynamic voltage restorer (DVR). Based on the voltage-current double closed-loop control strategy of the new LADRC, pre-synchronization control is adopted to effectively protect the stable operation of inverters and load devices in the power grid. To improve the overshoot phenomenon that occurs in traditional PI control under voltage imbalance / sag conditions, a dynamic voltage restorer (DVR) based on active disturbance rejection control (ADRC) is used to compensate for the voltage fluctuations of the photovoltaic system, and rapid compensation of distorted voltages is achieved to ensure the safety of the grid-connected system.

[0048] There are certain limitations in traditional DC-side boost control, AC-side inverter control, and the coordinated control of DC-side boost and AC-side inverter. Therefore, this paper explores the influence of the control strategy of a virtual synchronous generator (VSG) based on linear active disturbance rejection control (LADRC) and an active disturbance rejection control (ADRC) dynamic voltage restorer (DVR) under the condition of voltage imbalance / sag in photovoltaic grid connection.

[0049] Figure 1 The circuit in the middle is mainly divided into three parts: the main photovoltaic power generation circuit module, the inverter circuit controller part, and the dynamic voltage restorer (DVR). The main circuit module mainly consists of a photovoltaic array, a voltage source inverter with an LC filter, a three-phase circuit breaker, a dynamic voltage restorer, and an AC power grid. Based on the input of the photovoltaic array (PV), a three-phase sine wave is output through a voltage source inverter with an LC filter to obtain a voltage that meets the grid connection requirements and is connected to the AC power grid. The function of the three-phase circuit breaker is to control the on-off of the circuit and protect power equipment from faults such as overload and short circuit, so as to improve the reliability and safety of the power system. The output characteristic curve of the photovoltaic input due to partial shading will affect the three-phase sine wave output by the inverter. The simultaneous operation of the dynamic restorer and the three-phase circuit breaker improves the quality of the three-phase sine wave.

[0050] Figure 2 The voltage source inverter (VSI) with an LC filter in the middle effectively filters out high-frequency switching ripple and harmonic components, significantly improves the quality of the output voltage and current, reduces electromagnetic interference and improves grid friendliness, while reducing current ripple to protect load devices and enhancing the dynamic response and stability of the system.

[0051] Figure 3 The middle shows the working principle of a three-phase grid-connected inverter.

[0052] Figure 4The Simulink simulation diagram of the dynamic voltage restorer is given. The dynamic voltage restorer (DVR) system is an advanced flexible AC transmission device that can monitor the load voltage waveform in the power system in real time. When abnormal phenomena such as voltage sags, swells, or imbalances are detected, the system quickly activates the compensation mechanism. The active disturbance rejection controller (ADRC) processes the voltage error signal, calculates the required compensation voltage, and converts the DC power supply into an appropriate AC compensation voltage through a secondary converter. Subsequently, the generated compensation voltage is injected into the distribution system to correct voltage anomalies and restore the stability of the load voltage. Throughout the process, the DVR system continuously monitors voltage fluctuations and makes dynamic adjustments through a real-time feedback mechanism to ensure the stable operation of the power grid. The grid-connected load in the system reference distribution is a sensitive load, so the DVR with a controller is placed between the circuit breaker and the load. The input signals Vload and Iload first pass through an abc-to-dq coordinate transformation module to convert the three-phase abc signals into two-phase dq signals, i.e., the reference voltage on the dq axis, to simplify subsequent control design and analysis. Then, the reference voltage on the dq axis passes through the ADRC module to convert the voltage signal into a current signal, and the converted dq signal passes through a PI controller to adjust the dynamic response of the voltage and current. The ADRC controller realizes precise regulation of the system performance by real-time estimating and compensating the internal and external disturbances of the system. The ADRC module provides a reference voltage signal as the output for the PWM generator by collecting the input signals and combining the measured output voltage amplitude, thereby further improving the accuracy and response speed of voltage control. The input signals of the ADRC controller include the voltage feedback value, and the signals are sent to the controller after corresponding regulation and transformation to generate a control output for regulating the system switching signals.

[0053] Figures 7 - 10 The grid-connected voltage, current, and total harmonic distortion rate responses under normal operating conditions are given. Figure 7 is the grid-connected voltage response under normal operating conditions, Figure 8 is the grid-connected voltage THD under normal operating conditions, Figure 9 The grid-connected current response under normal operating conditions, Figure 10 is the grid-connected voltage THD under normal operating conditions.

[0054] Figures 11 - 14 The grid-connected voltage, current, and total harmonic distortion rate responses under voltage unbalance / sag conditions are given. Figure 11 is the grid-connected voltage response under voltage unbalance / sag conditions, Figure 12 is the grid-connected voltage THD under voltage unbalance / sag conditions, Figure 13 The grid-connected current response under voltage unbalance / sag conditions, Figure 14 is the grid-connected voltage THD under voltage unbalance / sag conditions.

[0055] Numerical simulation of the photovoltaic panel of SPR-305E-WHT-D, and its data are shown in Table 1. A voltage source inverter with a control strategy of a virtual synchronous generator based on linear active disturbance rejection control and an active disturbance rejection control (ADRC) dynamic voltage restorer is used to achieve the tracking of the maximum power point of the system. The present invention is applied to the simulation of a 380V 50HZ grid-connected system. The results are shown in Table 2 and Table 3.

[0056] Table 1. Characteristics of Photovoltaic Generator Sets

[0057]

[0058] Table 2. Normal Conditions

[0059]

[0060] Table 3. Voltage Unbalance / Sag Conditions

[0061]

[0062] VSI based on ADRC VSG and DVR is an aspect of the present invention. From the experimental diagrams and tables, it can be obtained that the control method of the present invention, which is based on the linear active disturbance rejection control (LADRC) virtual synchronous generator (VSG) and the control strategy of the active disturbance rejection control (ADRC) dynamic voltage restorer, has good applicability in the photovoltaic grid-connected system, with a fast response speed, good harmonic suppression effect, and guarantee for the safety of the grid-connected system.

[0063] As described in the above background art, an embodiment of the first aspect of the present invention proposes a voltage control method for a photovoltaic grid-connected system. In some embodiments of the present invention, as Figures 1 - 14 shown, the voltage control method includes the following steps:

[0064] S101, on the DC side of the photovoltaic grid-connected system, set an optimization process including multiple duty ratio control schemes; at each moment, optimize all duty ratio control schemes and the corresponding input powers through the optimization process, and make each duty ratio control scheme and the corresponding input power meet the preset conditions; at each moment, output and control the DC side according to the duty ratio control scheme with the highest input power.

[0065] Here, by pre-preparing multiple duty cycle control schemes, when the system faces voltage imbalance and voltage sag problems, it is no longer limited to a single strategy, but has multiple alternative schemes for comparison. In this way, the controller has the ability to flexibly allocate resources and can quickly switch to a more suitable scheme according to the current operating conditions, improving the adaptability and reliability. Conduct real-time review and optimization of the existing multiple duty cycle control schemes, that is, calculate the input power performance of each scheme based on the actual operation data at the current moment (such as voltage, current, load conditions, external environment, and grid status), and check whether these schemes meet the pre-set standards and limiting conditions. Through this real-time dynamic optimization and review, the system can ensure that unqualified schemes are excluded in a timely manner, and only those schemes that are safe, stable, and meet the electrical index requirements can enter the next selection process. Ensuring that only qualified schemes that meet the conditions participate in the comparison and obtaining their corresponding input power performance from these qualified schemes provide a reliable basis for the final decision-making in this link. The finally selected high-power scheme is not randomly determined, but is accurately obtained based on the data support, condition constraints, and optimization results of all previous related steps. Through this final selection action, the entire optimization process forms a closed loop: the reserve and review of multiple schemes in the early stage provide decision-making space for this link, and the decision-making result of this link will act on the DC side to improve voltage regulation and power generation efficiency; at the next moment, the execution effect of this decision-making result will feedback and affect the optimization and selection process of the next cycle, thus forming a virtuous associated cycle of continuous iteration and continuous improvement of stability and efficiency.

[0066] As described above, it provides a rich set of candidate solutions for the subsequent optimization evaluation process, ensuring that in the next step (optimizing and selecting multiple solutions), the system has sufficient solutions for evaluation and selection, making the subsequent decision-making more flexible and adaptable. Conduct an actual performance and compliance screening of the solutions within these sets. Through this associative effect, the system realizes the key transformation from "having alternatives" to "having viable and compliant alternatives", laying a solid foundation for the final solution selection. At the same time, the results of this optimization process are also associated with the subsequent solution selection step, providing the necessary information guarantee for selecting the optimal solution. Ensure that only qualified solutions that meet the conditions participate in the comparison, and obtain their corresponding input power performance from these qualified solutions - providing a reliable basis for the final decision-making in this step. The finally selected high-power solution is not randomly determined, but accurately obtained based on the data support, condition constraints, and optimization results of all previous associated steps. Through this final selection action, the entire optimization process forms a closed loop: the previous reserve and review of multiple solutions provide decision-making space for this step, and the decision-making results of this step will in turn act on the DC side to improve voltage regulation and power generation efficiency; at the next moment, the execution effect of this decision-making result will feedback and affect the optimization and selection process of the next cycle, thus forming a virtuous associative cycle of continuous iteration and continuous improvement of stability and efficiency.

[0067] In any of the above embodiments, the optimization process includes the following steps:

[0068] ACAC control uses maximum power point tracking based on the Gravitational Search Algorithm (GSA). This algorithm optimizes the search by utilizing the mutual gravitational force and the resulting acceleration among particles in the population, avoiding being trapped in local optima. The flowchart of the GSA-MPPT algorithm is as Figure 2 shown.

[0069] In this embodiment, MPPT is achieved through the GSA algorithm, that is, by simulating gravity and motion among the population particles to find the maximum power output point of the photovoltaic system under given conditions. The GSA algorithm simulates the interaction and gravitational force among particles, causing the particles to move towards positions with higher power, thereby searching for the optimal solution in the global search space and effectively avoiding the premature convergence of the algorithm to local optimal points. The core advantage of the GSA-MPPT algorithm is to dynamically adjust the position and velocity of particles in response to the gravitational force generated by the interaction among particles, continuously updating the search direction and step size. This method can greatly increase the possibility of finding the global optimal solution, especially suitable for dealing with multi-peak power curves, which are very common in photovoltaic power generation systems.

[0070] The GSA provides an efficient global optimization tool for MPPT. By simulating gravity through the interaction between particles, the algorithm can dynamically adjust the search strategy within a wider search space, making it more likely to find the true maximum power point. This algorithm not only improves the efficiency of MPPT but also enhances its adaptability in complex environments. For example, it can still maintain excellent performance in environments with rapid light changes and large temperature fluctuations. The particle dynamic adjustment mechanism in the gravitational search algorithm enables each particle to adjust its moving direction and speed based on its own and the group's best experience. This strategy effectively reduces the risk of the algorithm falling into local optima. When the algorithm avoids local optima, it can more effectively explore more possible power output points, thus significantly improving the power generation efficiency of the entire photovoltaic system and the power quality of the power grid.

[0071] Specifically, M i (s) is to calculate the mass of the i-th particle, and m i (s) is the proportion of the fitness of the i-th particle, and the calculation formula is:

[0072]

[0073] where N is the total number of particles; best(t) is the best fitness among all particles during iteration, worst(t) is the worst fitness among all particles during iteration, s is the number of iterations, j is the j-th particle, and fit i (s) is the fitness of the i-th particle at the s-th iteration. The fitness is used to compare the power output at the current working point, the output power of the photovoltaic array at a specific voltage and current, that is, the maximum power output; the maximum and minimum value problems are solved according to the optimization objective as follows:

[0074]

[0075] The expression for the magnitude of the gravitational force between particles is as follows:

[0076]

[0077] where is the magnitude of the gravitational force between particles, G(s) is the gravitational constant, is the position of the j-th particle in the m-dimensional space at the s-th iteration, the position of the i-th particle in the m-dimensional space at the s-th iteration, and R ij (s) is the Euclidean distance between particle i and particle j, ε is a small constant, and m is the dimension.

[0078] The expression for the gravitational constant is as follows:

[0079]

[0080] Among them, e is a tiny constant, G0 is the initial gravitational force, α is the attenuation coefficient, and T is the maximum number of iterations.

[0081] The resultant force of particle interaction is expressed as:

[0082]

[0083] Among them, F i d (s) is the resultant force generated by the gravitational force of the particle affected by other particles, and rand j is a random factor when moving in the search space.

[0084] The acceleration generated by the particle under the resultant force is expressed as:

[0085]

[0086] Among them, is the acceleration generated by the particle under the action of this resultant force.

[0087] Update the velocity and position of the particles participating in the next iteration, and the update expressions are as follows:

[0088]

[0089] Among them, V i d (s + 1) is the velocity of the particle in the next iteration.

[0090] In the algorithm, the duty cycle control is used as a method to achieve MPPT, representing the position of the particle. When the differences between the fitness of all particles and the maximum value and the differences in positions are both less than the set threshold, the algorithm stops iterating.

[0091] For the above specific description, the mass of each particle is calculated based on the difference between its fitness and the best and worst fitness values in the current population. This calculation ensures that particles with high fitness have greater mass and thus exert a greater gravitational force in the population dynamics. Fitness is an indicator for evaluating the performance of particles and is directly related to the output power of the photovoltaic array at specific voltages and currents, especially at the maximum power output point. The level of fitness reflects the quality of the particle's position (i.e., the operating point of the photovoltaic array). By calculating the mutual gravitational forces between each pair of particles, GSA drives the particles to move towards regions of better fitness. The calculation of the gravitational force depends on the relative positions and respective masses of the particles, enabling the entire search process to naturally explore directions approaching the global maximum. The acceleration of each particle is determined by the resultant force acting on it, which includes the gravitational forces from other particles and a random factor that helps the algorithm avoid local optima. The velocity of the particle is updated based on the acceleration, and then its position in the search space is updated. This process is dynamic to adapt to the continuously changing search environment. When the differences between the fitness of all particles and the maximum value and the differences in their positions are both less than a set threshold, the algorithm stops iterating. This means that the system has come close enough to the global optimal solution, or further search cannot significantly improve the result.

[0092] As can be seen from the above, the mass of a particle directly affects the magnitude of the gravitational force it exerts on other particles. High-mass particles play a more significant guiding role in the population, thus effectively leading the population towards the global optimal solution. This setting enhances the global search ability of the algorithm and reduces the risk of getting stuck at local optima. The fitness of a particle directly determines its behavior in subsequent iterations, including the gravitational force it experiences and the final direction of movement. The position of a particle with high fitness corresponds to a better photovoltaic output power. Therefore, fitness evaluation is the core driving force in the optimization process. The calculated gravitational force determines the acceleration of the particle, and the update of the acceleration directly affects the velocity and position of the particle. Through this physical simulation process, GSA can naturally adjust the search strategy and explore the solution space. The update of the velocity is the basis for the particle to explore new positions, and the fitness of the new position in turn verifies whether it is close to the maximum power point. When the improvement becomes insignificant, the iteration termination condition ensures that the algorithm stops at a reasonable solution, optimizing resource usage and preventing unnecessary calculations.

[0093] Furthermore, the control objective of the DCDC converter is to enable the output power of the generator on the DC side to track the maximum energy or stabilize the DC side voltage V dc . The slope at the maximum power point is 0. dI / dU is the change in conductance, and -I / U is the output conductance. When the two values are opposite, the photovoltaic array operates at the maximum power point.

[0094] Pmax = UI

[0095]

[0096] Among them, Pmax is the maximum power; U is the voltage at each moment; I is the current at each moment; dP is the derivative of power; dU is the derivative of voltage; dI is the derivative of current.

[0097] The control objective of the DC-AC (direct current to alternating current) converter is to solve the PI coupling problem on the AC side and improve the grid connection quality. By introducing virtual inertia and virtual damping, it provides support for the stability of the grid frequency, and a double closed-loop control virtual synchronous generator (VSG) control strategy based on LADRC (linear active disturbance rejection control) is used. For example, the voltage and current of phase a of a three-phase transformer:

[0098]

[0099] Among them, I Ca is the capacitance current of phase a. From the voltage and current of phase a, the corresponding voltage and current equations of phases b and c can be deduced. To obtain a better filtering waveform, a small resistor R is set for Lf; U a is the voltage of phase a; I a is the current of phase a; dU a is the derivative of the voltage of phase a; dI La is the derivative of the current of phase a; dt is the derivative with respect to time.

[0100] As can be seen from the above, the control objective of the DC-DC converter is to enable the output power of the DC-side generator to track the maximum energy point or maintain a stable and efficient DC-side voltage. In the maximum power point tracking (MPPT) control, the converter adjusts its output so that the operating point of the photovoltaic array can reach the optimal combination of voltage and current, thereby achieving the maximum energy output. By monitoring dI / dU (the change rate of current with respect to voltage) and -I / U (output conductance), the control system can accurately determine whether the maximum power point is reached. When the values of these two parameters are opposite, it means that the photovoltaic array is operating at the maximum power point because the slope (the derivative of the power curve) at this point is zero. This control method of the DC-DC converter directly affects the power generation efficiency and the stable operation of the system. Through effective MPPT control, not only is the energy utilization efficiency improved, but also the stable output of voltage and power is ensured, which is crucial for protecting the devices and loads connected to the system.

[0101] The control objective of the DC-AC inverter is to solve the PI coupling problem on the AC side, improve the grid connection quality, and introduce virtual inertia and virtual damping to support the stability of the grid frequency. This is achieved by using a double closed-loop control strategy based on linear active disturbance rejection control (LADRC), which is designed for a virtual synchronous generator (VSG). The VSG control strategy increases the dynamic stability of the grid-connected system by simulating the inertia and damping characteristics of a traditional synchronous generator. This control strategy can effectively regulate the active and reactive power to respond to grid demands and frequency changes. This control strategy of the DC-AC inverter interacts with the MPPT function of the DC-DC converter to jointly ensure the overall performance of the system. By stabilizing the output on the AC side, the inverter improves the overall quality of the grid and the grid connection efficiency of the photovoltaic system. At the same time, through precise frequency and power control, it helps the system better adapt to grid demands and environmental changes, reducing disturbances and instabilities in the grid.

[0102] S102. On the AC side of the photovoltaic grid-connected system, the active power and reactive power are determined in real time through the duty ratio control scheme output at each moment; the coupling relationship of the active power and reactive power in the two-axis coordinate system is solved respectively to obtain the voltage state variables and current state variables on each coordinate axis; considering the disturbance error, the voltage estimation value and current estimation value on each coordinate axis are calculated through the voltage state variables and current state variables, and the AC side is controlled according to the voltage estimation value and current estimation value at each moment.

[0103] Here, the duty ratio control strategy optimized from the DC side is converted into a power command executable on the AC side, that is, the corresponding active and reactive power is output according to the current control scheme. This can ensure that the power generated on the AC side not only meets the grid requirements but also maximally utilizes the photovoltaic power generation resources. The scalar active and reactive power are decomposed into components in the dq coordinate system to obtain more underlying state information, that is, the d-axis voltage, current state variables and q-axis voltage, current state variables. This decomposition enables the control system to no longer face complex three-phase alternating variables but to perform precise control based on the relatively simplified and stable d and q components like DC variables. Considering the grid disturbance, the function of compensating the error of the state variables and calculating the estimated values of the voltage and current is to ensure the control accuracy and the robustness of the system. No matter how the external conditions change, the electrical parameters closer to the actual working conditions can be obtained through correction to ensure the reliability of the control command. Finally, the estimated voltage and current parameters are used to control the AC side so that the waveforms and amplitudes of the output voltage and current can be maintained within the ideal range, thereby ensuring the grid voltage quality and power generation efficiency and reducing the impact of voltage imbalance and sudden drop on the system.

[0104] As described above, the duty cycle control scheme provides a decision basis for the AC-side power output. The process of determining active and reactive power is actually to transform the control strategy of the previous link into an executable power distribution scheme, laying a foundation for the subsequent dq coordinate decomposition and state variable acquisition. The dq decomposition and state variable acquisition enable the system to shift from macroscopic power control to fine regulation of microscopic electrical parameters (d-axis and q-axis voltage and current). Through this association, the power-level decision is effectively implemented as more underlying and specific voltage and current regulation means. After obtaining the state variables, if grid disturbances and measurement errors are not considered, the control may deviate from the actual operating conditions. By taking the disturbances into account and correcting the estimated values of voltage and current accordingly, the system ensures the adaptability of the control strategy in a real and complex environment. The state variables provide the original data for error compensation, and the error compensation makes the state variables more in line with the actual situation, providing a more reliable basis for the final control output. The estimated voltage and current provide a direct control basis for the final AC-side execution level, thereby achieving precise voltage regulation and current distribution. That is to say, all the previous steps (duty cycle scheme selection, power decomposition, state quantity acquisition, and error compensation) ultimately converge to the estimated values of voltage and current, and then the AC side is actually controlled through these estimated values to enable the power grid to obtain high-quality power output and ensure the stable and efficient operation of the system even in the situations of voltage imbalance and sudden drop.

[0105] In any of the above embodiments, at each moment, the DC side and the AC side are controlled in real time according to the same duty cycle control scheme, and the duty cycle control scheme includes the three-phase switch states of the three-phase inverter between the DC side and the AC side within a unit time.

[0106] In this embodiment, formulating a unified duty cycle control scheme involves determining the specific states of the switches of each phase of the three-phase inverter within a unit time. This scheme is based on the current DC-side input conditions and AC-side output requirements, optimizing the switching frequency and timing of the inverter to achieve the best energy conversion efficiency and grid matching. According to the formulated duty cycle control scheme, energy conversion control is carried out on the DC side and the output is adjusted on the AC side to meet the grid quality standards at the same time. The focus of the DC side is to adjust the inverter input to ensure the maximization of energy conversion; while the AC side focuses on the stability of the output and compatibility with the grid, especially in terms of voltage and frequency regulation.

[0107] The duty cycle control scheme directly affects the conversion efficiency of the inverter on the DC side. By precisely controlling the switching states of each phase, the available power of the solar panel can be maximally extracted and converted into the form of alternating current required by the power grid. This control not only optimizes the energy collection efficiency but also reduces the losses during the energy conversion process. On the AC side, the output of the inverter must strictly follow the operating standards of the power grid, including voltage waveform, frequency, and phase. The duty cycle control scheme of the inverter ensures the accuracy of the AC output, which is crucial for the stability of the power grid frequency and the ability to quickly respond to load changes. By adjusting the switching states of the three phases, the inverter can effectively respond to the changes in the power grid demand, improving the overall stability and efficiency of the system.

[0108] Specifically, the active power and reactive power are determined through the following steps:

[0109] Calculate the voltage and current of the three-phase inverter in the two-axis coordinate system through the three-phase switching states.

[0110] Correspond the two coordinate axes to the active power and reactive power respectively.

[0111] Calculate the time-averaged active power and reactive power per unit time based on the voltage and current of each coordinate axis.

[0112] Regarding the above specific description, by analyzing the switching states of the three-phase inverter, calculate the voltage and current output by the inverter in the two-axis coordinate system (usually the dq coordinate system, where the d-axis corresponds to the direct axis and the q-axis corresponds to the quadrature axis). This conversion allows the system to simplify the complex three-phase AC variables into two DC variables, facilitating power calculation and control. In the dq coordinate system, the d-axis voltage and current are usually associated with the active power, while the q-axis voltage and current are associated with the reactive power. This correspondence makes it possible to directly calculate the power from the voltage and current, simplifying the process of power control and regulation. Based on the voltage and current values of each coordinate axis, the time-averaged active power and reactive power per unit time are obtained through mathematical calculations. This calculation includes the product of the voltage and current and the consideration of their phase difference, providing accurate power output data for the system, which are the basis for power grid quality assessment and control.

[0113] As described above, the three-phase switching states of the inverter directly determine the waveforms and magnitudes of the output voltage and current. By precisely controlling these switching states, the output voltage and current can be accurately adjusted to meet the requirements of the power grid, which is a prerequisite for calculating active and reactive power. The voltage and current data obtained in the dq coordinate system are directly used for the calculation of active power and reactive power. This direct calculation relationship from voltage and current to power simplifies the control strategy and improves the speed and accuracy of system response, especially when the power grid load or external conditions change. The calculated active power and reactive power data are the basis for controlling the grid-connected behavior of the inverter. Active power directly affects the energy supply of the power grid, while reactive power is related to the voltage stability and load balance of the power grid. By accurately calculating these powers, the system can interact with the power grid more effectively, optimize energy distribution, and improve power grid stability.

[0114] Specifically, calculating the voltage and current of a three-phase inverter in a two-axis coordinate system through three-phase switching states includes the following steps:

[0115] The system can effectively filter out high-frequency switching ripples and harmonic components, significantly improve the quality of the output voltage and current, reduce electromagnetic interference, and improve grid friendliness. At the same time, it reduces current ripple to protect load equipment, enhances the dynamic response and stability of the system. For a typical three-phase inverter VSI with an LC filter, the output phase voltage can be derived as:

[0116]

[0117] where, V dc is the DC voltage of the photovoltaic array after passing through the Boost boost circuit; v an , v bn and v cn are the output voltages of phases a, b, and c respectively; S a , S b and S c are the three-phase switching states of phases a, b, and c respectively.

[0118] The three-phase switching states of the three-phase inverter are:

[0119]

[0120] Among them, the branch switch states S = {Sa, Sb, Sc} determine the conducting state "1" and the cut-off state "0". For the VSI, there are eight possible switch states and eight inverter voltage vectors vx = vα,x + jvβ,x, where α is the forward direction of the normal state voltage, β is the direction obtained by rotating the forward direction of the normal voltage counterclockwise by 90°, vx is the output voltage, j is the rotation factor for counterclockwise rotation by 90°, and x ∈ {0, 1, 7} (v0 = v7 = 0); S5 and S6 are the states of the fifth and sixth thyristors respectively; off and on are the off and on states of the switch respectively.

[0121] By transforming the three-phase coordinate system into a two-phase dq rotating coordinate system, the inverter voltage and current in the dq coordinate system can be obtained as follows:

[0122]

[0123] Among them, Li q is the inductance multiplied by the q-axis component of the current; Cv q is the capacitance multiplied by the q-axis component of the voltage; Cv d is the capacitance multiplied by the d-axis component of the voltage, based on Kirchhoff's law.

[0124] As described above, the role of the LC filter in the three-phase inverter is to reduce the high-frequency ripples and harmonics generated by the inverter switching operation. This not only improves the waveform quality of the output voltage and current, but also significantly reduces electromagnetic interference, making the inverter output smoother and more stable. By reducing the current ripple and optimizing the output waveform, the system can better be compatible with the power grid and reduce the interference to the power grid. At the same time, the smooth output also helps to protect the load devices connected to the inverter from damage caused by fluctuating or spike currents. The optimized output current and voltage quality directly enhance the system's response ability to load changes and operating stability. When the power grid or load conditions change, the system can adjust its output more quickly and effectively to maintain stable operation. Transforming the output of the three-phase inverter from the three-phase coordinate system to the two-phase dq coordinate system is to simplify the control strategy and enhance the feasibility of system analysis. In the dq coordinate system, the dynamic behavior of the inverter can be regarded as a DC system, making the control algorithm easier to implement and the analysis more intuitive.

[0125] The effective filtering function of the LC filter reduces the output ripple, thereby reducing the electromagnetic interference transmitted to the power grid. This interference reduction effect directly improves the overall power quality of the power grid, significantly enhancing the compatibility between the inverter and the power grid. By improving the quality of the output voltage and current, the system's resistance to external disturbances is enhanced, and its adaptability to internal load changes is improved. This high-quality power output not only protects the terminal equipment but also makes the overall operation of the system more stable and the response more sensitive. The dq coordinate system provides a more stable and continuous way to observe and control the performance of the inverter. By analyzing the voltage and current in the dq coordinate system, the control strategy can more precisely adjust the active and reactive power, thus better meeting the power grid's requirements and responding to the power grid's dynamic changes.

[0126] Specifically, the steps to separately solve the coupling relationship of the active power and reactive power in the two-axis coordinate system include:

[0127] Construct a virtual synchronous machine model for the three-phase inverter to simulate a synchronous generator.

[0128] Based on the active power loop part and reactive power loop part in the virtual synchronous machine model, determine the voltage loop part and current loop part in the two-axis coordinate system.

[0129] Regarding the above specific description, the virtual synchronous machine model allows the three-phase inverter to simulate the behavior of a traditional synchronous generator. This includes providing virtual inertia and damping to the power grid to help stabilize the grid frequency and respond to load changes. The construction of the virtual synchronous machine enables the inverter to operate in the power grid like a traditional generator, improving the system's grid friendliness. Within the framework of the virtual synchronous machine model, the active power loop and reactive power loop are respectively associated with the d-axis and q-axis in the two-axis coordinate system. The active power loop usually controls the voltage and current on the d-axis to adjust the active output of the inverter; the reactive power loop controls the voltage and current on the q-axis to manage the reactive power output. This separation ensures the independence of active and reactive control, enhancing the flexibility and precision of control.

[0130] As described above, by simulating the dynamic behavior of a synchronous machine, the virtual synchronous generator (VSG) model provides a framework for real-time dynamic adjustment of the voltage and current loops. The settings of the voltage loop and current loop depend on the active and reactive power control strategies in the model, ensuring that the output of the inverter can not only match the grid demand but also effectively respond to changes in the grid state. In the dq coordinate system, the optimization of the voltage loop and current loop directly affects the output of active and reactive power. By precisely adjusting the voltage and current on the d-axis and q-axis, the inverter can more accurately control the power transmitted to the grid, optimize energy distribution, and maintain the stability and efficiency of the grid simultaneously. The control of the voltage loop and current loop is not only related to the efficiency of power output but also directly affects the stability of the grid. The stable adjustment of the voltage loop helps to maintain the grid voltage level, while the control of the current loop helps to cope with the impact of grid load or supply changes, thereby enhancing the dynamic response ability and reliability of the entire system.

[0131] Furthermore, the determination steps of the active power loop part and reactive power loop part in the virtual synchronous generator model include:

[0132] Calculate the instantaneous active power and reactive power output by the grid-connected inverter;

[0133]

[0134] Among them, p is the instantaneous active power output by the three-phase inverter, q is the instantaneous reactive power output by the grid-connected inverter, v d is the component of the PV grid-connected voltage on the d-axis in the two-axis coordinate system, i d is the component of the current on the d-axis, v q is the component of the PV grid-connected voltage on the q-axis in the two-axis coordinate system, i q is the component of the current on the q-axis.

[0135] By passing through a low-pass filter, the average active power and reactive power output by the grid-connected inverter can be obtained respectively as:

[0136]

[0137] Among them, ω c is the cut-off frequency of the filter; s is a complex variable; P is the average active power; Q is the average reactive power.

[0138] Establish the VSG model, including the active power loop of the virtual synchronous generator and the reactive power loop of the virtual synchronous generator.

[0139] The active power loop of the virtual synchronous generator is:

[0140]

[0141] Among them, J is the moment of inertia of the synchronous generator, is the actual angular velocity of the rotor, ω is the actual angular velocity of the rotor, ω0 is the rated angular velocity of the rotor, D is the damping, D p is the damping coefficient, P m is the mechanical power, P e is the electromagnetic power, T m is the mechanical torque, T e is the electromagnetic torque.

[0142] Reactive power loop of the virtual synchronous generator:

[0143]

[0144] Among them, k is the droop coefficient; du / dt is the derivative of the output voltage with respect to time; Q set is the set reactive power value; Q e is the actual reactive power value and the reactive power output by the circuit; D q is the reactive power droop coefficient; u is the output voltage; u n is the rated voltage reactive power droop coefficient.

[0145] Introduce the active - frequency droop characteristic regulation equation:

[0146] P m = P ref + k ω (ω0 - ω)

[0147] Among them, P ref is the reference mechanical power; k ω is the power - frequency droop coefficient.

[0148] As can be seen from the above, the voltage and current of the grid - connected inverter are monitored in real time, and the active power and reactive power output are calculated instantaneously through these parameters. This real - time calculation is the basis for evaluating the performance of the inverter and the grid response, and provides initial data for further power regulation. The data of instantaneous active and reactive power are processed using a low - pass filter to obtain the average active power and reactive power. This processing step removes high - frequency fluctuations, provides a smoother power reading, helps to control the power output more accurately, and reduces the over - response of the system to instantaneous changes. The virtual synchronous generator model adds an active power loop and a reactive power loop on the basis of simulating a traditional generator. These control links are responsible for regulating the frequency and voltage of the inverter to adapt to the grid demand and maintain system stability. The active power loop is responsible for regulating the frequency response, while the reactive power loop regulates the voltage. The active - frequency droop characteristic regulation is a control method commonly used in synchronous generators to regulate the relationship between the output power and the frequency, so as to respond to the grid load changes. In the VSG model, by simulating this characteristic, it helps the inverter to integrate into the grid more naturally and improves the adaptability to load changes.

[0149] The active and reactive power data calculated in real time are processed by a low-pass filter to obtain more stable power output data. This smoothing process ensures that the control strategy is formulated based on reliable and representative data, avoiding overreactions or inappropriate responses caused by data fluctuations. The averaged active and reactive power data are used in the VSG model, directly affecting the control strategies of the active and reactive loops. These data provide the necessary inputs for the model to regulate the grid frequency and voltage, ensuring harmonious operation between the inverter and the grid. The active loop adjusts the frequency response and active output to adapt to grid load changes, while the reactive loop regulates the voltage to maintain grid voltage stability. The droop characteristic regulation is particularly important in the active loop, enabling the active output of the inverter to flexibly adapt to frequency changes, thereby providing more stable grid support.

[0150] Specifically, the voltage loop part is determined through the following steps:

[0151] Obtain the equations for the rate of change of voltage with respect to time on each axis of the voltage loop part based on the active loop part and the reactive loop part.

[0152] Construct the first voltage state equation of the voltage loop part through the rate-of-change equations.

[0153] Separate the disturbance quantities of the photovoltaic grid-connected system in the first voltage state equation and update it to the second voltage state equation.

[0154] For the above specific description, the data obtained from the active loop part and the reactive loop part are used to determine the rate of change of voltage with respect to time on each axis (d-axis and q-axis) of the voltage loop part. This step is based on the current operating state of the inverter and grid conditions, calculating the voltage change trend and providing the necessary dynamic information for constructing the voltage state equation. Using the obtained voltage rate-of-change equations, the first voltage state equation of the voltage loop part is constructed. This equation describes how the voltage changes over time and is the core part of the control algorithm design, used to predict and adjust the voltage output of the inverter to match the stable operation requirements of the grid. Based on the first voltage state equation, the disturbance quantities of the photovoltaic grid-connected system are further identified and separated, and then the equation is updated to the second voltage state equation. This step aims to optimize the voltage control strategy, enhancing the robustness and adaptability of the control scheme by considering system disturbances.

[0155] As described above, the rate-of-change equation of voltage directly affects the construction of the first voltage state equation. The rate-of-change equation provides a specific mathematical description of voltage change, which is the basis for constructing the state equation. This association ensures that the voltage control strategy can reflect the latest changes in the grid and inverter states in real time. The disturbance quantity information included in the first voltage state equation is the key to identifying external and internal disturbances of the system. By analyzing and separating these disturbance quantities, the voltage control strategy can be adjusted more accurately to form the second voltage state equation. This separation and update process improves the system's ability to handle uncertainties and enhances the accuracy and stability of voltage regulation. Identifying and handling disturbance quantities is crucial for optimizing the entire voltage control link. Through this optimization, the inverter can respond more flexibly and effectively to fluctuations and changes in the grid, ensuring the stability of the output voltage, thereby supporting the reliable operation of the grid and the efficient grid connection of the photovoltaic system.

[0156] In any of the above embodiments, there is a coupling relationship between the d-axis and q-axis of the grid-connected converter, and the active power and reactive power cannot be controlled separately. To better control the inverter voltage and current of the VSG in the dq coordinate system, the voltage loop part is rewritten, and the rate-of-change equation includes the following formula:

[0157]

[0158] Where, is the rate of change of v q with respect to time t; is the rate of change of v q with respect to time t; ω is the actual angular frequency that changes due to the dynamic process under grid disturbances; C is the capacitance value; ΔC is the parameter error on the capacitor.

[0159] In this embodiment, the interdependent relationship between the voltages and currents on the d-axis and q-axis of the grid-connected converter is analyzed. This coupling relationship usually stems from the physical and electrical characteristics of the inverter, such as electromagnetic induction and capacitance effects, which affect the inverter's response ability to grid dynamics. To mitigate the coupling effect between the d-axis and q-axis, the voltage loop control strategy needs to be rewritten. The rewritten control strategy will adopt more advanced algorithms or control logics, such as decoupling control strategies, which can regulate the voltages on the d-axis and q-axis more independently, making the active and reactive power control more accurate and independent.

[0160] The identified coupling relationship between the d-axis and the q-axis directly guides the rewriting of the voltage loop control strategy. After clarifying the specific form and impact of the coupling, a more suitable control strategy can be designed to specifically address this issue, such as reducing the impact of the coupling by adding control links or adjusting feedback parameters. The rewritten voltage loop control strategy will directly affect the performance of the grid-connected converter, especially its ability to accurately control active and reactive power. By more effectively controlling the voltages on the d-axis and the q-axis, the system can more flexibly respond to the demands of the power grid, improving the efficiency of energy conversion and the stability of the power grid. The new voltage loop control strategy can more effectively handle changes in the grid load and other dynamic conditions, thereby enhancing the overall stability and responsiveness of the system. By decoupling the control of the d-axis and the q-axis, the inverter can more independently regulate the active and reactive outputs, thus more accurately meeting the requirements of grid operation.

[0161] Specifically, the disturbance quantity of the photovoltaic grid-connected system is separated in the first voltage state equation and updated to the second voltage state equation;

[0162] According to It can be seen that the order of the voltage loop system is first-order. The system disturbance quantity is separated and based on the following relationship: The second voltage state equation can be expressed as:

[0163]

[0164] Among them, the x represents voltage; the is the voltage state variable of the d-axis in the two-axis coordinate system; the is the voltage state variable of the q-axis in the two-axis coordinate system; the f vd is the sum of the internal disturbance quantity on the d-axis and the voltage disturbance quantity on the AC side; the f vq is the sum of the internal disturbance quantity on the q-axis and the voltage disturbance quantity on the AC side; the internal disturbance quantity includes the coupling between the d-axis and the q-axis, as well as the parameter error on the capacitor; the b v is the system parameter of the photovoltaic grid-connected system; the y val is the component of the output voltage on the d-axis in the input power; the y vq1 is the component of the output voltage on the q-axis in the input power.

[0165] Specifically, the estimated value is obtained by constructing the LESO control equation of the voltage loop part, and the steps for constructing the LESO control equation of the voltage loop part include:

[0166] The disturbance quantity on each coordinate axis is obtained through the rate-of-change equation.

[0167] The disturbance quantity on each coordinate axis is introduced into the voltage parameter of the second voltage state equation and used as the LESO control equation of the voltage loop part.

[0168] For the above specific description, initially, the disturbance quantities on each coordinate axis are calculated through the rate-of-change equation. This equation takes into account the rates of change of voltage and current, and through real-time analysis, it can identify disturbances caused by external factors such as grid load variations or internal factors such as the dynamic changes of the inverter itself. This step is the basis for estimating the unmeasurable part of the system state. The disturbance quantities obtained from the rate-of-change equation are integrated into the second voltage state equation of the voltage loop part. This step allows the control system to consider the actual impact of these disturbances on voltage control by adding disturbance terms to the state equation, thereby enabling more precise voltage regulation. The construction of the LESO control equation for the voltage loop part is completed. This control equation uses the technology of the extended state observer to estimate not only the measurable system states such as voltage and current but also those disturbance states that are not directly measurable. The introduction of LESO greatly enhances the robustness and adaptability of the control system, enabling the system to more effectively cope with external and internal disturbances.

[0169] As can be seen from the above, the rate-of-change equation provides the basis for real-time data analysis and directly affects the identification accuracy of the disturbance quantities. Identifying accurate disturbance quantities is a prerequisite for effective voltage regulation because these disturbances directly affect the stability and efficiency of the system. Integrating the identified disturbance quantities into the voltage state equation enables the control strategy to be adjusted based on the complete system dynamic information, including those non-intuitive internal variables and external influencing factors. This integration method makes voltage control more precise and faster in response. The implementation of the LESO control equation directly improves the performance of the entire voltage control loop, enabling the system to maintain stability within a wider operating range. By accurately estimating and regulating the voltage, LESO helps the system maintain the continuity and efficiency of operation in the face of grid or internal disturbances.

[0170] Specifically, the step of introducing the disturbance quantities on each coordinate axis into the voltage parameters of the second voltage state equation includes:

[0171] When the disturbance is introduced into the voltage parameters of the second voltage state equation, the voltage state equation can be written as:

[0172]

[0173] Among them, is the derivative of the total disturbance including internal and external disturbances on the d-axis; is the derivative of the total disturbance including internal and external disturbances on the q-axis.

[0174] As described above, the disturbance quantity identified from the system operation is introduced into the second voltage state equation. This includes adding the disturbance quantity obtained from the rate of change equation to the control logic of the voltage loop to ensure that the voltage control can reflect the actual grid and internal system states in real time. The updated voltage state equation now incorporates the influence of the disturbance quantity, which allows the control system to more precisely adjust the output voltage of the inverter to match the actual grid conditions and counteract possible disturbances.

[0175] After the integration of the disturbance quantity and the update of the state equation, the LESO control equation is finally formed. This control equation uses the extended state observer technique to estimate and compensate for those states that are not directly measured, such as the disturbances caused by factors such as grid load changes.

[0176] By identifying and integrating the disturbance quantity, the second voltage state equation is updated, which enables the voltage control system to operate based on more comprehensive system dynamic information. This update directly improves the adaptability of the control strategy, enabling it to effectively respond to grid changes and other external disturbances. The updated voltage state equation provides the basis for the formation of the LESO control equation, enabling LESO to accurately estimate the system state, including those disturbance quantities that are not easily directly measured. The implementation of this control equation greatly enhances the robustness of the system and improves the accuracy and stability of voltage control. Through the LESO control equation, the system can more accurately predict and compensate for voltage changes and respond to external and internal disturbances. This not only improves the system's dynamic response ability but also ensures the continuity and stability of the voltage output, which is crucial for improving the overall quality of grid connection operation.

[0177] In any of the above embodiments, the LESO control equation of the voltage loop includes the following formula:

[0178]

[0179] Wherein, the is the estimated value of the state variable of the voltage of the PV grid-connected system disturbance on the d-axis in the two-axis coordinate system; the is the estimated value of the state variable of the voltage of the PV grid-connected system disturbance on the q-axis in the two-axis coordinate system; the is the estimated value of the total disturbance variable of the PV grid-connected system voltage on the d-axis; the is the estimated value of the total disturbance variable of the PV grid-connected system voltage on the q-axis; the z vd2 is the estimated value of x vd2 and x vd2 is the total disturbance including internal and external disturbances on the d-axis; the β v1 is a LESO gain coefficient of the voltage; the z vd1 is the estimated value of the said x vd1 and the said xvd1 is the d-axis voltage; the b v are the system parameters of the photovoltaic grid-connected system; the y vd1 is the component of the output voltage on the d-axis in the input power; the z vq2 is the estimated value of x vq2 and the x vq2 is the total disturbance including internal and external disturbances on the q-axis; the z vq1 is the estimated value of the x vq1 and the x vq1 is the q-axis voltage; the y vq1 is the component of the output voltage on the q-axis in the input power; the β v2 is another LESO gain coefficient of the voltage.

[0180] Specifically, the photovoltaic grid-connected system further includes a PD controller and a dynamic voltage restorer on the AC side, and steps of controlling the AC side according to the voltage estimation value and current estimation value at each moment, including:

[0181] Calculating the output of the PD controller through the values of the voltage estimation value or current estimation value on each coordinate axis.

[0182] Controlling the dynamic voltage restorer to compensate for the voltage sags or voltage surges that randomly occur on the AC side at each moment through the output of the PD controller.

[0183] For the above specific description, the PD controller uses the estimated values of voltage or current to calculate its control output. These estimated values reflect the voltage or current states on the AC side on each coordinate axis. The PD controller calculates the necessary adjustment signals based on these data to respond to the upcoming voltage changes. The output of the PD controller is based on proportional-derivative logic and can quickly and effectively compensate for the sudden changes in voltage or current. The dynamic voltage restorer (DVR) receives the output signal from the PD controller and adjusts its intervention in the power grid accordingly. The main task of the DVR is to compensate for the short-term voltage sags or surges on the AC side and maintain the stable operation of the power grid by dynamically adjusting the voltage level.

[0184] As can be seen above, the estimated values of voltage and current provide the basic data for the operation of the PD controller. The accuracy of these data directly affects the effectiveness of the PD controller's output because the controller's response is based on these real-time estimates to adjust the intensity and direction of the output signal in order to quickly respond to changes in voltage or current. The output of the PD controller directly drives the operation of the DVR. This output usually contains the necessary compensation instructions to guide the DVR on how to adjust its output to counteract the effects of voltage sags or surges. The response speed and accuracy of the DVR determine the quality of the compensation effect, thus maintaining the stability of the power grid. The effective operation of the DVR can significantly improve the overall stability of the power grid. By compensating for voltage sags and surges in a timely manner, the DVR helps prevent these short-term voltage anomalies from causing extensive interference or damage to the power grid and the loads connected to it, which is particularly crucial in industrial and commercial applications that highly rely on stable power supplies.

[0185] In any of the above embodiments, the output of the PD controller is calculated by the following formula:

[0186]

[0187] where, the u vd1 is the output of the PD controller on the d-axis; the u vq1 is the output of the PD controller on the q-axis; the k vp is the proportionality coefficient in the PD controller; the v vd is the given value of the d-axis component of the controlled voltage; the v vq is the given value of the q-axis component of the controlled voltage.

[0188] A voltage control method for a photovoltaic grid-connected system provided by the present invention optimizes the process including multiple duty ratio control schemes on the DC side, and optimizes and compares all schemes and their corresponding input powers at each moment to ensure that each duty ratio control scheme and its input power meet the preset conditions. Finally, the system will select the duty ratio control scheme with the highest input power from them to control the DC side. This multi-scheme parallel optimization and screening mode enables the system to dynamically adapt to environmental and load changes, and still maintain a working state close to the maximum power point under complex conditions such as light fluctuations and voltage dips, thus significantly improving the power generation efficiency and reducing the power loss caused by voltage instability.

[0189] On the AC side, through real-time decision-making corresponding to the duty cycle control scheme on the DC side, active and reactive powers are obtained, and their coupling relationships are respectively resolved in the dq two-axis coordinate system. This process can clearly distinguish the active and reactive power control objectives and accurately control them, enabling the system to quickly adjust the output power distribution at the moment when voltage imbalance or sudden drop occurs, thereby reducing the impact of grid fluctuations on the system. Considering the disturbance error, the present invention calculates the voltage and current estimated values on each axis through voltage and current state variables to obtain more accurate state information. With these accurate estimated values, the system can respond and adjust the AC side with high flexibility, helping to respond faster to voltage mutations and sudden drops and maintaining the stability of the power grid and power quality.

[0190] Through the fast response characteristic of the PD controller and the instant compensation ability of the DVR for voltage transient problems, short-term abnormalities of the grid voltage can be quickly identified and corrected. This enables the system to inject or absorb reactive power in a timely manner for voltage compensation when voltage sag or sudden drop occurs, maintaining power quality and grid stability, thus effectively solving the adverse effects caused by grid voltage abnormalities mentioned in the background art.

[0191] In summary, the following technical advantages are respectively obtained:

[0192] The parallel optimization and real-time selection of multiple duty cycle control schemes enable the system to operate near the maximum power point under different environmental conditions, thereby improving the energy utilization rate of photovoltaic power generation. This can not only obtain higher power generation and benefits in long-term operation but also effectively reduce the efficiency loss of equipment caused by operation deviating from the optimal working conditions.

[0193] With the decomposition and decoupling control of active and reactive powers in the dq coordinate system, the present invention enables the system to quickly adjust the power output strategy in the face of load fluctuations, grid voltage abnormalities, and other disturbances. Coupled with the introduction of advanced control methods such as LESO (Linear Extended State Observer), the system has higher anti-disturbance ability and operates more smoothly.

[0194] The collaborative effect of the DVR and the PD controller significantly improves the quality of the grid voltage. When the grid voltage appears abnormal, the DVR can quickly compensate, and the PD controller helps maintain good voltage conditions with its proportional-derivative fast response characteristic. This process can reduce the risk of equipment damage and power outages in the power system and improve grid reliability and power consumption safety.

[0195] The entire technical solution realizes multi-level closed-loop control from the DC side to the AC side through state variable estimation, accurate calculation of voltage and current estimation values, and real-time compensation of disturbances. The system can quickly detect and adapt to changes, and implement control strategies with high precision, thereby reducing overshoot, oscillation, and control delay, and ensuring the smoothness and efficiency of the grid connection process.

[0196] In any of the above embodiments, the current loop is designed by analogy with the current loop, and the current loop LESO is designed as:

[0197]

[0198] Wherein, is the estimated value of the state variable of the current of the photovoltaic grid-connected system disturbance on the d-axis in the two-axis coordinate system; is the estimated value of the state variable of the current of the photovoltaic grid-connected system disturbance on the q-axis in the two-axis coordinate system; is the estimated value of the total current disturbance variable of the photovoltaic grid-connected system on the d-axis; is the estimated value of the total current disturbance variable of the photovoltaic grid-connected system on the q-axis; z cd1 is the estimated value of the said x cd1 and the said x cd1 is the d-axis current; z cd2 is the estimated value of X cd2 and the said X cd2 is the sum of the inner and outer disturbances on the d-axis; z cq1 is the estimated value of x cq1 and the said x cq1 is the q-axis voltage; z cq2 is the estimated value of X cq2 and the said X cq2 is the sum of the inner and outer disturbances on the q-axis; β c1 ; β c2 is the voltage LESO gain coefficient; b c is the system parameter with a magnitude of -1 / L; y cd1 is, y cq1 are the d- and q-axis components of the voltage.

[0199] In any of the above embodiments, the current loop system compensation is set as:

[0200]

[0201] Wherein, u cd1 is the d-axis output of the current loop PD controller; u cq1 is the q-axis output of the current loop PD controller; k cp is the proportional coefficient in the PD controller; v cd is the given value of the d-axis component of the controlled current; v cdis the given value of the q - axis component of the controlled current; z cd1 is the estimation of the d - axis component of the current; z cd2 is the estimation of the disturbance on the d - axis; z cq1 is the estimation of the d - axis component of the current; z cq2 is the estimation of the disturbance on the q - axis; b c = - 1 / L.

[0202] In any of the above - mentioned embodiments, the dynamic voltage restorer DVR compensates the grid - connected voltage based on ADRC, and establishes the DVR state equation according to the filter inductor L and the filter capacitor C:

[0203]

[0204] wherein, R is the protection resistor; is the derivative of the inductor current of each phase on the output side of the inverter, and i fm is the current value flowing through the filter inductor; u m is the output phase voltage of the inverter; u cm is the voltage value across the capacitor; i lm is the inductor current of each phase on the output side of the inverter; m = 1, 2, 3 represent three phases.

[0205] In any of the above - mentioned embodiments, although the dynamic voltage restorer DVR system is a complex non - linear system, in order to perform closed - loop control on the second - order DVR system, a second - order ADRC controller is adopted. This controller mainly includes three parts:

[0206] Tracking differentiator TD, which extracts high - frequency dynamic information and differential data in the system to improve the system response.

[0207] Extended state observer ESO, which estimates the unmeasured state and external disturbance of the system and provides real - time estimation for the system state.

[0208] Non - linear state error feedback control law NLSEF, which calculates the control signal based on the state information provided by the observer, reduces the system error, and achieves the control goal.

[0209] In this embodiment, the working process of the second - order ADRC controller includes that the input signal extracts differential information after being processed by TD, performs state estimation through ESO, and finally calculates the control signal by ISEF, which acts on the DVR system to achieve closed - loop control, thereby improving the dynamic performance and robustness of the DVR system.

[0210] As described above, the drive system of a photovoltaic generator set consists of a series of high-efficiency components that are responsible for converting solar energy into electrical energy. It is a key part of the photovoltaic power generation system and, due to its complexity and operating environment, often becomes the main factor affecting the stability and reliability of the system. Ensuring the stable operation of the drive system of a photovoltaic generator set is crucial for avoiding system downtime and ensuring continuous power supply. With the continuous progress of photovoltaic technology, the capacity and scale of photovoltaic generator sets are increasing day by day, and the requirements for the working performance of photovoltaic systems are also increasing. In the process of converting high-energy-density sunlight into electrical energy, the photovoltaic system needs to bear greater load changes, which may lead to more complex dynamic behaviors, including potential grid connection safety problems. Currently, although there are various dynamic models of photovoltaic systems, most still focus on the analysis of simplified single-mass systems and fail to fully capture the voltage and current distortions caused by voltage imbalance / sag and their profound impact on the overall dynamic response of the system. To solve this problem, the purpose of the present invention is to use the control strategies of a virtual synchronous generator (VSG) with linear active disturbance rejection control (LADRC) and an active disturbance rejection control (ADRC) dynamic voltage restorer (DVR) for a photovoltaic grid-connected system with voltage imbalance / sag, considering the quality of its output voltage and current, to achieve high-quality control of the system.

[0211] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0212] The embodiments described above are only for describing the preferred mode of the present invention and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A voltage control method for a photovoltaic grid-connected system, characterized in that, Comprising: On the DC side of the photovoltaic grid-connected system, an optimization process including multiple duty cycle control schemes is set up; At each moment, all the duty cycle control schemes and the corresponding input powers are optimized through the optimization process, and each duty cycle control scheme and the corresponding input power meet the preset conditions; at each moment, the duty cycle control scheme with the highest input power is output and used to control the DC side; On the AC side of the photovoltaic grid-connected system, the active power and the reactive power are determined in real time through the duty cycle control scheme output at each moment; The coupling relationships of the active power and the reactive power in the two-axis coordinate system are respectively solved to obtain the voltage state variables and the current state variables on each coordinate axis; under the consideration of disturbance errors, the voltage estimated values and the current estimated values on each coordinate axis are calculated through the voltage state variables and the current state variables, and the AC side is controlled according to the voltage estimated values and the current estimated values at each moment.

2. The voltage control method according to claim 1, wherein At each moment, the DC side and the AC side are controlled in real time according to the same duty cycle control scheme, and the duty cycle control scheme includes the three-phase switch states of the three-phase inverter between the DC side and the AC side within a unit time.

3. The voltage control method according to claim 2, characterized in that The active power and the reactive power are determined through the following steps: Calculate the voltage and current of the three-phase inverter in the two-axis coordinate system through the three-phase switch states; Correspond the two coordinate axes to the active power and the reactive power respectively; Calculate the averaged active power and reactive power within a unit time according to the voltage and current of each coordinate axis.

4. The voltage control method according to claim 2, characterized in that, The step of respectively solving the coupling relationships of the active power and the reactive power in the two-axis coordinate system includes: Construct a virtual synchronous machine model for simulating a synchronous generator by the three-phase inverter; Determine the voltage loop part and the current loop part in the two-axis coordinate system according to the active power loop part and the reactive power loop part in the virtual synchronous machine model.

5. The voltage control method according to claim 4, characterized in that The voltage loop part is determined through the following steps: Obtain the rate-of-change equations of the voltage with respect to time on each coordinate axis of the voltage loop part according to the active power loop part and the reactive power loop part; Construct the first voltage state equation of the voltage loop part through the rate-of-change equations; Separate the disturbance quantity of the photovoltaic grid-connected system in the first voltage state equation and update it to the second voltage state equation.

6. The voltage control method according to claim 5, characterized in that, The rate-of-change equations include the following formulas: Among them, the is the rate of change of the d with respect to time t, and the d is the d-axis component of the grid-connected voltage of the photovoltaic system in the two-axis coordinate system; the is the rate of change of the q with respect to time t, and the q is the q-axis component of the grid-connected voltage of the photovoltaic system in the two-axis coordinate system; the d is the d-axis component of the current; the q is the q-axis component of the current; the ω changes due to the dynamic process under grid disturbances; the C is the capacitance value; the ΔC is the parameter error of the capacitor.

7. The voltage control method according to claim 5, wherein The estimated values are obtained by constructing the LESO control equations of the voltage loop part, and the steps for constructing the LESO control equations of the voltage loop part include: Obtain the disturbance quantities on each coordinate axis through the rate-of-change equations; Introduce the disturbance quantities on each coordinate axis into the voltage parameters of the second voltage state equation and use them as the LESO control equations of the voltage loop part.

8. The voltage control method according to claim 7, wherein The LESO control equations of the voltage loop part include the following formulas: Among them, the is the estimated value of the state variable of the voltage of the photovoltaic grid-connected system disturbance on the d-axis in the two-axis coordinate system; the is the estimated value of the state variable of the voltage of the photovoltaic grid-connected system disturbance on the q-axis in the two-axis coordinate system; the is the estimated value of the total voltage disturbance variable of the photovoltaic grid-connected system on the d-axis; the is the estimated value of the total voltage disturbance variable of the photovoltaic grid-connected system on the q-axis; the z vd2 is the estimated value of x vd2 , and x vd2 is the total disturbance including internal and external disturbances on the d-axis; the β v1 is a LESO gain coefficient of the voltage; the z vd1 is the estimated value of the vd1 x, and the vd1 x is the d-axis voltage; the b v is the system parameter of the photovoltaic grid-connected system; the y vd1 is the component of the output voltage on the d-axis in the input power; the z vq2 is the estimated value of x vq2 , and the vq2 x is the total disturbance including internal and external disturbances on the q-axis; the z vq1 is the estimated value of the vq1 x, and the vq1 x is the q-axis voltage; the y vq1 is the component of the output voltage on the q-axis in the input power; the β v2 is another LESO gain coefficient of the voltage.

9. The voltage control method according to claim 8, wherein The photovoltaic grid-connected system further includes a PD controller and a dynamic voltage restorer located on the AC side, and the step of controlling the AC side according to the voltage estimated values and the current estimated values at each moment includes: Calculate the output of the PD controller based on the values of the voltage estimation or current estimation on each coordinate axis; Control the dynamic voltage restorer by the output of the PD controller to compensate for the voltage sags or voltage surges that randomly occur on the AC side at each moment.

10. The voltage control method according to claim 9, wherein The output of the PD controller is calculated by the following formula: wherein, the u vd1 is the output of the PD controller on the d-axis; the u vq1 is the output of the PD controller on the q-axis; the k vp is the proportional coefficient in the PD controller; the v vd is the given value of the d-axis component of the controlled voltage; the v vq is the given value of the q-axis component of the controlled voltage.

Images