Power adjusting method, device and equipment of photovoltaic energy storage system and storage medium

Abstract

The invention relates to the technical field of photovoltaic energy storage systems, and discloses a power regulation method, device and equipment for a photovoltaic energy storage system and a storage medium, and the method comprises the steps: carrying out the real-time collection and wavelet decomposition of voltage and current parameters of a photovoltaic array, and obtaining power fluctuation components and power trigger signals of different frequencies; executing power vector analysis to obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy; constructing a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel, and obtaining a coordination trigger instruction of the photovoltaic side and the energy storage side; and performing multi-objective optimization decomposition and slope limitation processing on the coordination triggering instruction to obtain a voltage reference value of the photovoltaic MPPT controller and a current reference value of the battery charging and discharging controller, thereby realizing accurate identification and prediction of different frequency power fluctuation components, realizing collaborative cooperation of photovoltaic side feed-forward and energy storage side feedback, and improving the reliability of the system. And the power oscillation risk caused by control mode switching is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic energy storage systems, and in particular to a power regulation method, device, equipment and storage medium for a photovoltaic energy storage system. Background Art

[0002] With the rapid deployment and application of distributed photovoltaic energy storage systems, power stability issues have become a severe challenge facing the power system. Due to factors such as light intensity fluctuations and load changes, the output power of photovoltaic systems often shows instability. This power fluctuation will have an impact on the power grid, reduce the quality of power, and affect the stable operation of the power grid. Especially in grid-connected applications, unstable power output may cause grid frequency fluctuations, trigger the malfunction of protection devices, and even cause local grid collapse.

[0003] Traditional photovoltaic energy storage systems generally use a power balance control method based on DC bus voltage feedback, which indirectly achieves power balance by stabilizing the DC bus voltage. However, this method often exhibits power oscillation problems under conditions of rapid changes in light intensity or sudden changes in load, and it is difficult to quickly and effectively suppress power fluctuations. In particular, when the system encounters sudden changes in light intensity caused by extreme weather conditions or sudden changes in load caused by grid failures, the lag and singleness of traditional control methods will lead to untimely power regulation and even cause system instability. Summary of the invention

[0004] The present invention provides a power regulation method, device, equipment and storage medium for a photovoltaic energy storage system. The present invention realizes accurate identification and prediction of power fluctuation components of different frequencies, realizes the coordinated cooperation of photovoltaic side feedforward and energy storage side feedback, and avoids the risk of power oscillation caused by control mode switching.

[0005] In a first aspect, the present invention provides a power regulation method for a photovoltaic energy storage system, the power regulation method for a photovoltaic energy storage system comprising: The voltage and current parameters of the photovoltaic array are collected in real time and decomposed by wavelet to obtain power fluctuation components of different frequencies and power trigger signals; Performing power vector analysis according to the power fluctuation components of different frequencies and the power trigger signal to obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy; Based on the oscillation mode characteristic parameter set and the oscillation suppression strategy, a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel are constructed to obtain coordinated trigger instructions for the photovoltaic side and the energy storage side; The coordinated trigger instruction is subjected to multi-objective optimization decomposition and slope limitation processing to obtain a voltage reference value of a photovoltaic MPPT controller and a current reference value of a battery charge and discharge controller.

[0006] In a second aspect, the present invention provides a power regulating device for a photovoltaic energy storage system, the power regulating device for the photovoltaic energy storage system comprising: Real-time acquisition module, used for real-time acquisition and wavelet decomposition of the voltage and current parameters of the photovoltaic array to obtain power fluctuation components of different frequencies and power trigger signals; A power vector analysis module, used to perform power vector analysis according to the power fluctuation components of different frequencies and the power trigger signal to obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy; A feedback trigger module, used to construct a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel based on the oscillation mode characteristic parameter set and the oscillation suppression strategy, and obtain coordinated trigger instructions for the photovoltaic side and the energy storage side; The processing module is used to perform multi-objective optimization decomposition and slope limiting processing on the coordinated trigger instruction to obtain a voltage reference value of the photovoltaic MPPT controller and a current reference value of the battery charge and discharge controller.

[0007] The third aspect of the present invention provides a power regulation device for a photovoltaic energy storage system, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the power regulation device of the photovoltaic energy storage system executes the above-mentioned power regulation method of the photovoltaic energy storage system.

[0008] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned power regulation method for the photovoltaic energy storage system.

[0009] In the technical solution provided by the present invention, accurate identification and prediction of power fluctuation components of different frequencies are achieved through multi-order polynomial fitting and wavelet decomposition technology. The quantitative analysis method based on the power vector analysis model clarifies the influence mechanism of each frequency component on the system stability. The design of the dual-channel power trigger mechanism realizes the coordinated cooperation of photovoltaic side feedforward and energy storage side feedback, and the power fluctuation can be effectively suppressed without calculating complex power compensation values ​​on the photovoltaic side. The multi-objective optimized power allocation strategy takes into account both response speed and energy utilization efficiency, and realizes the optimal allocation of system resources. Differentiated processing strategies are adopted for power fluctuations with different frequency characteristics to avoid the risk of power oscillation caused by control mode switching. The introduction of system operation status evaluation and control strategy optimization mechanism ensures that the system maintains the best performance state during long-term operation, and significantly improves the power stability and energy utilization efficiency of the photovoltaic energy storage system under extreme conditions such as drastic changes in light or sudden changes in load.

[0010] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0011] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of an embodiment of a power regulation method for a photovoltaic energy storage system in an embodiment of the present invention; Figure 2 A schematic diagram of an embodiment of a power regulating device for a photovoltaic energy storage system in an embodiment of the present invention; Figure 3 Schematic diagram of an embodiment of a power regulation device of a photovoltaic energy storage system in an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0015] To facilitate understanding of this embodiment, a power regulation method for a photovoltaic energy storage system disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps: 101. Real-time acquisition and wavelet decomposition of the voltage and current parameters of the photovoltaic array are performed to obtain power fluctuation components of different frequencies and power trigger signals; It is understandable that the execution subject of the present invention may be a power regulating device of a photovoltaic energy storage system, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0016] Specifically, a voltage sensor and a current sensor are installed at the output end of the photovoltaic array to collect the output voltage and current of the photovoltaic array. After the collected voltage and current data are processed by signal, a product operation is performed to calculate the output power of the photovoltaic array. At the same time, a voltage sensor is installed at the DC bus to measure the voltage value of the DC bus in real time. The voltage value is calculated by difference with the preset DC bus voltage reference value to obtain the voltage deviation of the DC bus, reflecting the stability of the DC bus voltage. If the voltage deviation is large, it means that there is a large mismatch between the output power of the photovoltaic array and the load power, and the power allocation strategy needs to be adjusted to maintain the stable operation of the system. The DC bus voltage deviation is processed by adaptive bandpass filtering to extract the power fluctuation characteristic signal. The center frequency of the bandpass filter is dynamically adjusted according to the real-time operating state of the system to adapt to different power fluctuation frequencies. Through this filter, power fluctuations in different frequency ranges can be effectively identified, and short-term fluctuations and long-term trend changes can be distinguished. Adaptive filtering automatically adjusts the filtering parameters according to the dynamic changes of the photovoltaic system, so that components of different power fluctuation frequencies can be accurately extracted. After adaptive bandpass filtering, a signal characterizing the power fluctuation characteristics is obtained, which is used to determine whether there is obvious power fluctuation in the system and its influence range. Multi-order polynomial fitting is performed on the photovoltaic output power and power fluctuation characteristic signals. The polynomial fitting method is used to predict short-term power trends. By using the historical data of photovoltaic output power, the power change trend in the short term in the future is fitted to improve the responsiveness of power regulation. At the same time, the fitting method can analyze the instantaneous trend of power changes to determine whether there are sudden power changes and optimize the dynamic response strategy of the energy storage system. After completing the multi-order polynomial fitting, the wavelet decomposition method is used to decompose the photovoltaic output power and power fluctuation characteristic signals to extract power fluctuation components of different frequencies. The wavelet decomposition method decomposes the signal into multiple components of different frequencies to distinguish short-term fluctuations from long-term trend changes. Through this method, the power fluctuation signal is decomposed into high-frequency components, medium-frequency components and low-frequency components. The high-frequency components mainly reflect short-term power fluctuations, such as changes caused by sudden load changes or environmental disturbances; the medium-frequency components correspond to periodic fluctuations, such as short-term power fluctuations caused by changes in light; and the low-frequency components characterize the long-term power change trend of the system, such as the overall impact of weather changes on photovoltaic power generation. Based on the different frequency power fluctuation components obtained by wavelet decomposition, it is judged whether there is abnormal power fluctuation and the corresponding power regulation strategy is triggered. A power trigger signal is set to identify significant power fluctuations. When the fluctuation amplitude of the high-frequency component, the medium-frequency component or the low-frequency component exceeds the set threshold, the corresponding power trigger signal is triggered to further adjust the power allocation strategy.

[0017] The fifth-order polynomial fitting process is performed on the historical data sequence of photovoltaic output power to obtain the photovoltaic output power forecast value for the next sampling period. In order to ensure the accuracy of the fitting results, the coefficients of the polynomial are dynamically updated by the recursive least squares method, so that it is continuously optimized according to the latest data, thereby improving the reliability and real-time performance of the prediction. Since the output power of the photovoltaic system is affected by the light intensity, ambient temperature and load changes, the real-time adjustment of the recursive least squares method can more accurately track the change trend of the photovoltaic output power and provide a reasonable estimate of the future power level. The weighted moving average and exponential smoothing process are performed on the photovoltaic output power forecast value for the next sampling period to reduce the interference of random fluctuations on the power regulation strategy. Among them, the weighted moving average method effectively filters out the influence of short-term power fluctuations and makes the prediction results more stable, while the exponential smoothing method dynamically adjusts the weights according to the historical trends, so that the latest data contributes more to the predicted value, thereby improving the sensitivity of short-term prediction. In order to ensure the smoothness and stability of the prediction, the results of the weighted moving average and exponential smoothing are comprehensively weighted, and the weight coefficients are reasonably allocated to meet the requirements of balancing short-term changes and long-term trends to obtain the medium-term power forecast value. According to the medium-term power forecast value, the power fluctuation characteristic signal is subjected to wavelet decomposition to extract the power fluctuation components in different frequency ranges. The wavelet decomposition method splits the original signal into multiple components of different scales to distinguish short-term fluctuations from long-term trend changes. In this process, the wavelet transform method is used to decompose the collected power fluctuation characteristic signal, and the high-frequency component, medium-frequency component and low-frequency component are extracted respectively. Among them, the high-frequency component reflects the sudden power fluctuation in a short time, such as the influence of rapidly changing load demand or environmental disturbance; the medium-frequency component corresponds to the periodic power fluctuation, such as the short-term change of sunshine intensity; and the low-frequency component is used to characterize the long-term power change trend of the system, such as climate change or long-term load demand fluctuation. Through wavelet decomposition, the power fluctuation in different frequency ranges is identified, and support is provided for the formulation of power regulation strategies. Based on the amplitude of the power fluctuation components of different frequencies, adaptive weight coefficients are assigned to the power fluctuation components of each frequency and weighted summed to determine the compensation demand. The compensation weight of the high-frequency component is higher to ensure a rapid response to sudden power fluctuations, while the compensation weight of the low-frequency component is relatively low to avoid affecting the long-term stability of the system due to excessive adjustment. The adaptive weight coefficient is dynamically adjusted according to the current power fluctuation situation. For example, when the power fluctuation characteristic signal indicates that the high-frequency component is dominant, the system will increase the weight of high-frequency compensation accordingly to enhance the suppression effect of short-cycle power fluctuations; when the influence of the intermediate frequency or low-frequency component is more significant, the system will adjust the weight distribution accordingly to optimize the overall power regulation strategy. In this way, the power compensation demand can be accurately calculated according to the power fluctuation characteristics of different frequencies.The power stability index is calculated based on the ratio of the power compensation demand to the system rated power. This index is the ratio of the power compensation demand to the system rated power and is used to measure the impact of the current power fluctuation on the system stability. When the power compensation demand is small, the power stability index is high, indicating that the system is in a relatively stable operating state; when the power compensation demand is large, the power stability index is low, indicating that the system has a large power fluctuation and appropriate adjustment measures need to be taken to restore the stability of the system. The power stability index is compared with the preset threshold to determine whether power adjustment measures need to be taken. When the power stability index is lower than the preset threshold, it means that the power fluctuation of the system exceeds the allowable range, affecting the safe operation of the system, so a power trigger signal is generated to start the power adjustment process.

[0018] 102. Perform power vector analysis according to power fluctuation components of different frequencies and power trigger signals to obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy; Specifically, the power of the system is vectorized to more intuitively reflect the dynamic characteristics of the power on the photovoltaic side, battery side and load side. The system power is expressed in complex form, where the real part corresponds to the active power and the imaginary part corresponds to the reactive power. In order to obtain accurate power vector distribution, the photovoltaic side power vector, battery side power vector and load side power vector are calculated based on the phase difference information of the photovoltaic output power, battery power and load power. The photovoltaic side power vector is calculated by the relationship between the photovoltaic output power and the phase angle, while the battery side power vector is vectorized according to the current, voltage and charge and discharge state of the battery, and the calculation of the load side power vector is based on the power demand characteristics of the load. Through this step, the power vector distribution is obtained to describe the power flow relationship between different power sources and loads. In order to evaluate the impact of system power oscillation, the system power angle and its derivative are calculated, and the damping power component and the synchronization power component are extracted based on this information. The damping power component is used to measure the system's ability to suppress power disturbances, while the synchronization power component is used to describe the synchronization characteristics of the system under different power states. The damping power is calculated by the time derivative of the power angle, and the synchronous power is calculated by combining the power angle deviation to quantify the impact of dynamic power changes. At the same time, in order to evaluate the power oscillation risk of the system, the ratio of the damping power component to the synchronous power component is analyzed, and the system is judged whether it is in a potential power oscillation state based on this ratio. When the damping power is small, it means that the power disturbance of the system cannot be effectively suppressed, resulting in the occurrence of power oscillation. When the synchronous power is too high, the system will produce large-scale power oscillations, affecting its stability. Through this evaluation method, the risk level of power oscillation is determined based on real-time monitoring data, and corresponding adjustment strategies are formulated accordingly. After completing the evaluation of the power oscillation risk, in order to quantify the power dynamic characteristics of the system, a state space model is constructed. The model contains state variables such as DC bus voltage, battery current, photovoltaic power, and power angle, and uses photovoltaic reference voltage and battery reference current as input variables to form a complete mathematical description of the system. The DC bus voltage is used to reflect the voltage stability of the entire system, while the battery current is directly related to the charging and discharging state of the energy storage system. At the same time, photovoltaic power is the main energy input source of the system, and the power angle is used to describe the interaction between different power components. On this basis, these state variables are constructed into a vector and combined with the input variable vector to establish a state space model to obtain the system matrix and input matrix. The system matrix is ​​used to describe the dynamic relationship between the state variables, while the input matrix is ​​used to characterize the influence of the photovoltaic reference voltage and the battery reference current on the system state. The eigenvalues ​​of the system matrix are calculated, and the stability of the system is judged based on the eigenvalues. The calculation process of the eigenvalue involves decomposing the system matrix and extracting the real and imaginary parts to analyze the dynamic characteristics of the system under different operating conditions.The real part of the eigenvalue determines the stability of the system. When the real part is positive, the system is in an unstable state, and when the real part is negative, the system tends to be stable. At the same time, the imaginary part of the eigenvalue reflects the oscillation characteristics of the system. The larger its absolute value, the higher the oscillation frequency of the system. By calculating the real and imaginary parts of the eigenvalue, it is judged whether the current power state is stable and the power oscillation that occurs is predicted. On the basis of obtaining the eigenvalue, the power oscillation evaluation index is calculated. The index is calculated by the ratio of the maximum real part of the eigenvalue to the minimum absolute value of the imaginary part, thereby quantifying the intensity of the power oscillation. The power oscillation evaluation index can provide a measurement standard to indicate the current oscillation severity of the system. At the same time, in order to analyze the impact of power fluctuations of different frequencies on system oscillation, the contribution values ​​of high-frequency, medium-frequency and low-frequency power fluctuation components to the power oscillation evaluation index are calculated respectively to obtain the oscillation contribution value set. The high-frequency component corresponds to the rapidly changing power disturbance, while the medium-frequency component mainly reflects the periodic power oscillation, and the low-frequency component reflects the long-term stability of the system. Through this analysis method, the main frequency components that cause power oscillation are accurately located, and targeted oscillation suppression strategies are formulated accordingly. The dominant frequency component of the oscillation is determined based on the set of high-frequency, medium-frequency and low-frequency oscillation contribution values, and an appropriate oscillation suppression strategy is selected based on the dominant frequency. When the high-frequency component is dominant, a fast-response power compensation strategy is adopted to suppress power disturbances in a short period of time, and when the medium-frequency component is dominant, a smooth transition power allocation strategy is adopted to reduce the impact of periodic power fluctuations. When the low-frequency component dominates the oscillation, a long-time constant power filtering strategy is adopted to optimize the long-term stability of the system. Output oscillation mode characteristic parameter set, including oscillation mode, oscillation frequency and oscillation phase.

[0019] The high-frequency, medium-frequency and low-frequency oscillation contribution value sets are normalized to eliminate the influence of data of different scales and make them comparable within the same numerical range. After normalization, the oscillation contribution values ​​of different frequencies are compared in size, and the frequency component corresponding to the maximum contribution value is determined as the current oscillation dominant frequency component. If the high-frequency oscillation contribution value is the largest, it means that the power fluctuation of the system is mainly caused by sudden disturbances in a short period of time, while if the medium-frequency oscillation contribution value is the largest, it means that the power fluctuation of the system presents periodic oscillation characteristics, which is affected by external periodic loads or changes in light, and if the low-frequency oscillation contribution value is the largest, it means that the main source of the system power fluctuation is long-term power imbalance or the slow regulation response of the energy storage system. Through this process, the frequency component of the dominant oscillation is determined. Strategy matching is performed based on the dominant frequency component of the oscillation to select the most suitable oscillation suppression method. If the high-frequency component is dominant, it means that the system needs to respond quickly to quickly suppress short-term power fluctuations. Therefore, a fast-response power compensation strategy is selected to quickly adjust the charging and discharging power of the energy storage system to weaken the impact of high-frequency power disturbances. When the intermediate frequency component is dominant, it indicates that the power fluctuation of the system mainly presents periodic changes. A smooth transition power allocation strategy is selected to make the power regulation process smoother to avoid the impact of drastic adjustments on the stability of the system. When the low-frequency component is dominant, it indicates that the power fluctuation of the system is long-term trending. At this time, a long-time constant power filtering strategy is adopted to slowly adjust the power allocation of the system, thereby optimizing the power balance and ensuring the long-term stability of the system. Through the strategy matching method based on the dominant frequency component, the most suitable oscillation suppression strategy is selected for different types of oscillation characteristics to maximize the effect of power regulation. The power fluctuation time domain signal corresponding to the dominant frequency component is subjected to singular value decomposition to obtain the oscillation modal eigenvalue. The singular value decomposition method can effectively extract the main modal information in the power fluctuation signal and remove irrelevant noise, so that the oscillation modal eigenvalue can accurately characterize the main power oscillation mode of the system. In order to analyze the frequency characteristics of the oscillation, the power fluctuation signal of the dominant frequency component is subjected to fast Fourier transform to obtain the oscillation frequency eigenvalue. The time domain signal is converted to the frequency domain, and the main frequency component of the power fluctuation is analyzed to quantify the oscillation frequency characteristics of the system. Calculate the phase difference between the dominant frequency component and the adjacent frequency component to obtain the oscillation phase characteristic value. The calculation of the phase difference can reflect the relative timing relationship between different frequency components, thereby revealing the propagation characteristics of power oscillation in the time domain. If the phase difference between adjacent frequency components is large, it means that the power oscillation in the system presents a strong phase change characteristic, and the power regulation strategy needs to be further optimized to reduce the impact of phase offset on system stability. On the contrary, if the phase difference is small, it means that the power fluctuation of the system is relatively synchronized, and the overall stability of the system can be enhanced through appropriate compensation strategies.The oscillation mode eigenvalue, the oscillation frequency eigenvalue and the oscillation phase eigenvalue are combined into an oscillation mode characteristic parameter set.

[0020] 103. Based on the oscillation mode characteristic parameter set and the oscillation suppression strategy, a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel are constructed to obtain coordinated trigger instructions for the photovoltaic side and the energy storage side; Specifically, the difference between the predicted value of the photovoltaic output power and the current photovoltaic output power is calculated. The difference reflects the deviation between the current photovoltaic power generation and the predicted power, which is used to calculate the feedforward compensation. The calculation of the feedforward compensation needs to take into account the dynamic change characteristics of the system, so the feedforward gain is set to ensure the rapidity and accuracy of the compensation effect. Since the power oscillation characteristics of the system will affect the effectiveness of the feedforward compensation, the feedforward gain is dynamically adjusted according to the high-frequency oscillation contribution value in the oscillation mode characteristic parameter set. If the high-frequency oscillation contribution value is large, it means that the power fluctuation of the system is mainly affected by short-term disturbances. In this case, the feedforward gain needs to be increased accordingly to enhance the system's response ability to sudden power changes and ensure that the power regulation on the photovoltaic side can quickly follow the power demand changes. On the contrary, if the high-frequency oscillation contribution value is low, it means that the power fluctuation of the system is mainly dominated by medium and low frequency components. At this time, the feedforward gain should be appropriately reduced to prevent over-regulation from causing system instability. Through the dynamic adjustment mechanism, the photovoltaic side feedforward trigger channel can effectively adapt to different types of power fluctuations and improve the dynamic response capability of the system. At the same time, the construction of the energy storage side feedback trigger channel is based on the DC bus voltage deviation and its change rate to calculate the feedback compensation amount. The DC bus voltage deviation reflects the power balance state of the current system, while its rate of change can describe the dynamic characteristics of power fluctuations. The feedback compensation is calculated based on these two parameters to more accurately characterize the power demand of the system. In the process of calculating the feedback compensation, the proportional gain and differential gain are set to ensure that the calculation of the compensation can accurately reflect the dynamic changes of the system. Since the intermediate frequency and low frequency components of power oscillation usually affect the long-term stability of the system, the proportional gain and differential gain need to be dynamically adjusted according to the contribution values ​​of the intermediate frequency and low frequency oscillations in the oscillation mode characteristic parameter set. When the intermediate frequency oscillation contribution value is large, it means that the power fluctuation of the system is mainly affected by periodic factors. At this time, the proportional gain is enhanced to improve the ability to suppress medium-cycle power fluctuations, while the differential gain should be appropriately reduced to avoid short-term rapid changes affecting the stability of the system. When the low-frequency oscillation contribution value is large, it indicates that the power fluctuation of the system is mainly caused by long-term power imbalance. At this time, the role of the proportional gain should be appropriately reduced, and the differential gain should be enhanced to ensure that the system can adapt to long-term power trend changes more smoothly, thereby optimizing the charging and discharging strategy of the energy storage system. Through the dynamic adjustment mechanism, the energy storage side feedback trigger channel can effectively compensate for the long-term power fluctuations of the system and maintain the stability of the system operation. After completing the construction of the photovoltaic side feedforward trigger channel and the energy storage side feedback trigger channel, the trigger instructions of the photovoltaic side and the energy storage side are generated according to the changes in the photovoltaic side feedforward compensation and the energy storage side feedback compensation. For the generation of the photovoltaic side trigger instruction, it is judged whether the absolute value of the feedforward compensation exceeds the set feedforward threshold, or whether its change rate exceeds the change rate threshold.If the absolute value of the feedforward compensation exceeds the feedforward threshold, it means that the current photovoltaic power regulation demand is large, and the power adjustment on the photovoltaic side is triggered at this time to quickly respond to the power fluctuation demand. Similarly, if the change rate of the feedforward compensation exceeds the change rate threshold, it means that the photovoltaic power changes rapidly, and the power adjustment instruction on the photovoltaic side is triggered at this time to prevent the power change from affecting the stability of the system too quickly. Through this judgment mechanism, the photovoltaic side trigger channel can quickly respond to different types of power fluctuations and provide corresponding compensation strategies. In the process of generating the trigger instruction on the energy storage side, the judgment is made based on the comparison result of the absolute value of the feedback compensation and the feedback threshold, and whether the DC bus voltage deviation exceeds the safety range. If the absolute value of the feedback compensation exceeds the set feedback threshold, it means that the current energy storage system needs a large charge and discharge adjustment to maintain the stability of the DC bus voltage, so the power adjustment instruction on the energy storage side is triggered to increase the responsiveness of the energy storage system. If the DC bus voltage deviation exceeds the safety range, it indicates that there is a large imbalance in the power distribution of the current system. At this time, the energy storage side needs to be adjusted immediately to avoid voltage fluctuations affecting the normal operation of the system. Through this trigger mechanism, the trigger channel on the energy storage side can provide precise adjustment instructions under different power fluctuation conditions to maintain the balance of system power. After the trigger instructions on the photovoltaic side and the energy storage side are generated, the two trigger instructions are combined, and the coordination coefficient is dynamically adjusted according to the oscillation suppression strategy to obtain the final coordinated trigger instruction. The adjustment of the coordination coefficient needs to take into account the oscillation characteristics of the current system. When the contribution value of high-frequency oscillation is large, the weight of the photovoltaic side is increased to enhance the response capability to high-frequency power fluctuations. When the contribution value of medium-frequency and low-frequency oscillations is large, the weight of the energy storage side is increased to enhance the system's compensation capability for long-term power fluctuations. Through reasonable coordinated trigger instructions, the system can dynamically allocate power regulation tasks between the photovoltaic side and the energy storage side to optimize the stability and operation efficiency of the system.

[0021] 104. Perform multi-objective optimization decomposition and slope limitation processing on the coordinated trigger instruction to obtain a voltage reference value of the photovoltaic MPPT controller and a current reference value of the battery charge and discharge controller.

[0022] Specifically, the coordination trigger instruction is multiplied by the power proportional factor to complete the conversion of the power demand and obtain the total power regulation demand. The setting of the power proportional factor depends on the rated power and current operating state of the system, ensuring that the calculated total power regulation demand can accurately reflect the current power allocation demand. On this basis, an optimization objective function is constructed to optimize the power regulation process while meeting the system operation constraints. In order to ensure that the power regulation can both quickly respond to the power change demand of the system and minimize energy loss, the optimization objective function considers the response speed target and the energy loss target at the same time. In terms of response speed, a new power balance state is reached as soon as possible to avoid the impact of power fluctuations on system stability, so the objective function contains a term that can minimize the regulation time. In terms of energy loss, the conversion loss of photovoltaic power generation and energy storage system is reduced, so the objective function contains a term that can minimize energy loss. In this way, the optimization objective function can comprehensively consider the dynamic performance and energy efficiency level of the system, thereby realizing an efficient power regulation strategy. On the basis of the optimization objective function, the optimization solution is performed under certain constraints to ensure that the calculated power regulation scheme meets the actual operating constraints of the system. Among them, the photovoltaic output power is constrained by the photovoltaic power generation capacity, so the photovoltaic side power regulation instruction must be within the maximum output power range allowed by the photovoltaic system. The charging and discharging capacity of the energy storage system is limited by the battery state of charge, so the energy storage side power regulation instruction needs to ensure that it will not cause the battery state of charge to exceed the safe range. Based on these constraints, the optimal photovoltaic side power regulation instruction and energy storage side power regulation instruction are calculated using the optimization solution algorithm. According to the photovoltaic output characteristic curve, the voltage reference value of the photovoltaic MPPT controller is calculated. In order to ensure that the photovoltaic system always operates near the optimal power point, the photovoltaic output characteristic curve is found to determine the maximum power point voltage value under the current conditions. Since the power regulation instruction causes the adjustment of the photovoltaic power, a voltage offset is calculated to ensure that the photovoltaic output power can meet the new power regulation requirements. The maximum power point voltage value is added to the calculated offset to obtain the first reference value of the photovoltaic MPPT controller, thereby ensuring that the photovoltaic system can operate stably under the new power demand. At the same time, according to the energy storage side power regulation instruction, the current reference value of the battery charge and discharge controller is calculated. Since the charging and discharging power of the battery is related to the battery voltage, the power regulation instruction on the energy storage side is divided by the current battery voltage value to calculate the actual charging and discharging current demand. Through this calculation method, it is ensured that the charging and discharging process of the battery matches the power regulation instruction, thereby optimizing the operating state of the energy storage system and improving the overall energy utilization efficiency of the system. In order to prevent excessively fast voltage or current changes during the power regulation process, which affects the stability of the system, the calculated photovoltaic MPPT controller voltage reference value and battery charge and discharge controller current reference value are slope limited.The maximum voltage change rate is set to ensure that the voltage reference value of the photovoltaic MPPT controller does not change suddenly, thereby avoiding unstable power output of the photovoltaic system due to excessively fast voltage changes. Similarly, the maximum current change rate is set to ensure that the current reference value of the battery charge and discharge controller is adjusted smoothly within a reasonable range, thereby avoiding the energy storage system from causing battery life reduction or system oscillation due to excessively fast current changes. Through slope limiting processing, the stability of the power regulation process is guaranteed, and the operating stability and safety of the entire photovoltaic energy storage system are improved.

[0023] In the embodiment of the present invention, accurate identification and prediction of power fluctuation components of different frequencies are achieved through multi-order polynomial fitting and wavelet decomposition technology. The quantitative analysis method based on the power vector analysis model clarifies the influence mechanism of each frequency component on the system stability. The design of the dual-channel power trigger mechanism realizes the coordinated cooperation of photovoltaic side feedforward and energy storage side feedback, and the power fluctuation can be effectively suppressed without calculating complex power compensation values ​​on the photovoltaic side. The multi-objective optimized power allocation strategy takes into account both response speed and energy utilization efficiency, and realizes the optimal allocation of system resources. Differentiated processing strategies are adopted for power fluctuations with different frequency characteristics to avoid the risk of power oscillation caused by control mode switching. The introduction of system operation status evaluation and control strategy optimization mechanism ensures that the system maintains the best performance state during long-term operation, and significantly improves the power stability and energy utilization efficiency of the photovoltaic energy storage system under extreme conditions such as drastic changes in light or sudden changes in load.

[0024] In a specific embodiment, the process of executing step 101 may specifically include the following steps: The output voltage and current of the photovoltaic array are collected by a voltage sensor and a current sensor arranged at the output end of the photovoltaic array, and the output voltage and the output current are multiplied to obtain the photovoltaic output power; The DC bus voltage value is collected by a voltage sensor arranged on the DC bus, and the difference between the DC bus voltage value and a preset DC bus voltage reference value is calculated to obtain a DC bus voltage deviation; Adaptively bandpass filter the DC bus voltage deviation to obtain a power fluctuation characteristic signal; Based on the photovoltaic output power and power fluctuation characteristic signals, multi-order polynomial fitting and wavelet decomposition are performed to obtain power fluctuation components with different frequencies and power trigger signals.

[0025] Specifically, voltage sensors and current sensors are installed at the output end of the photovoltaic array to collect the output voltage and current values ​​of the photovoltaic array in real time. These sensors convert the voltage and current information of the photovoltaic power generation unit into digital signals through a high-precision data acquisition module and transmit them to the control system. After receiving the data, the system calculates and obtains the photovoltaic output power by multiplying the collected voltage value and current value. This power value can accurately characterize the power generation capacity of the photovoltaic system under the current environmental conditions. At the same time, a voltage sensor is installed at the DC bus to obtain the real-time DC bus voltage. The DC bus voltage is a key parameter of the entire system and directly affects the power balance between the photovoltaic power generation unit, the energy storage unit and the load. By measuring the voltage value of the DC bus and comparing it with the preset DC bus voltage reference value, the DC bus voltage deviation is calculated to reflect the current power balance of the system. If the deviation is too large, it indicates that there is a large mismatch between photovoltaic power generation, energy storage charging and discharging, and load power, and a reasonable power regulation strategy is needed to optimize the system operation. Adaptive bandpass filtering is performed on the DC bus voltage deviation to remove irrelevant low-frequency drift and high-frequency noise, and at the same time, the power fluctuation characteristic signal that has a significant impact on the system operation state is extracted. Adaptive bandpass filtering can dynamically adjust the filtering parameters according to the real-time operation state of the system to adapt to different power fluctuation characteristics. When the system is in a stable state, the center frequency of the filter is relatively low to capture the slowly changing power fluctuation trend, and when the system detects a rapid power change, the center frequency of the filter is automatically adjusted up to improve the responsiveness to high-frequency power fluctuations. Multi-order polynomial fitting is performed on the photovoltaic output power and power fluctuation characteristic signals to predict the power change trend in a short period of time. Multi-order polynomial fitting establishes a mathematical model based on historical data to predict future power level changes. By continuously updating the fitting parameters, it adapts to different weather conditions and load changes, thereby improving the accuracy of power prediction. In order to decompose the different frequency components of power fluctuations and extract the key power disturbance signals, the fitted power signal is subjected to wavelet decomposition. The original power signal is split into multiple components of different scales to distinguish short-term fluctuations from long-term trend changes, and power fluctuation components of different frequencies and power trigger signals are obtained.

[0026] In a specific embodiment, the execution step performs multi-order polynomial fitting and wavelet decomposition based on the photovoltaic output power and power fluctuation characteristic signal to obtain power fluctuation components of different frequencies and a power trigger signal, which may specifically include the following steps: The historical data series of photovoltaic output power is processed by fifth-order polynomial fitting, and the polynomial coefficients are updated by recursive least squares method to obtain the predicted value of photovoltaic output power in the next sampling period; The photovoltaic output power forecast value of the next sampling period is processed by weighted moving average and exponential smoothing, and weight coefficients are assigned for weighted calculation to obtain the medium-term power forecast value; According to the mid-term power prediction value, the power fluctuation characteristic signal is processed by wavelet decomposition to extract the high-frequency component, the medium-frequency component and the low-frequency component respectively, and the power fluctuation components of different frequencies are obtained; Based on the amplitude of the power fluctuation components of different frequencies, an adaptive weight coefficient is assigned to the power fluctuation components of each frequency and weighted sum is performed to obtain the power compensation demand; The power stability index is calculated based on the ratio of the power compensation demand to the system rated power; The power stability index is compared with a preset threshold value, and a power trigger signal is generated when the power stability index is lower than the preset threshold value.

[0027] Specifically, the historical data sequence of photovoltaic output power is analyzed, and a mathematical model is established using the fifth-order polynomial fitting method to predict the photovoltaic output power in the next sampling period. In this process, the output power data of the photovoltaic array is continuously collected, and the historical power data sequence within a period of time is stored as the basis for fitting. The fifth-order polynomial fitting can better describe the trend of photovoltaic power changes over time and can adapt to complex nonlinear fluctuations, thereby providing a more accurate power prediction value. In order to improve the real-time performance and accuracy of the fitting, the recursive least squares method is used to update the polynomial coefficients. The recursive least squares method is a dynamic update algorithm that adjusts the existing polynomial fitting results according to the new measurement data, so as to ensure that the fitting model can respond quickly when the photovoltaic output power changes. The photovoltaic output power prediction value of the next sampling period is smoothed to eliminate the impact of random fluctuations in a short period of time on power regulation. The weighted moving average and exponential smoothing method are combined to optimize the predicted data. In the weighted moving average processing process, different weights are assigned to multiple historical sampling points, so that the newer data has a greater impact on the final prediction result, while the older data has a relatively small impact. At the same time, the exponential smoothing method increases the system's sensitivity to recent data changes and reduces the cumulative error of historical data by assigning exponential decay weights to new data. This weighted calculation method enables the power forecast results to reflect short-term trend changes and maintain a certain stability when the data fluctuates greatly, thereby improving the reliability of power regulation. After calculating the medium-term power forecast value, the forecast value is used to perform wavelet decomposition on the power fluctuation characteristic signal to extract power fluctuation components in different frequency ranges. The wavelet decomposition method can decompose the power fluctuation signal into components of multiple scales, thereby distinguishing short-term violent fluctuations from long-term trend changes. The high-frequency component, the medium-frequency component and the low-frequency component are extracted respectively, among which the high-frequency component mainly reflects the power mutation in a short period of time, such as the power change caused by load mutation or environmental disturbance; the medium-frequency component corresponds to periodic fluctuations, such as power oscillation caused by weather changes or power grid fluctuations; and the low-frequency component mainly reflects the long-term power change trend of the system, such as the impact of seasonal sunshine changes on photovoltaic power generation. Based on the amplitude of the power fluctuation components of different frequencies, they are weighted and summed to calculate the overall power compensation demand. In order to ensure that the power compensation strategy accurately adapts to different types of power fluctuations, adaptive weight coefficients are assigned to the power fluctuation components of each frequency, and these weights are dynamically adjusted according to the power fluctuation characteristics. For example, when the system detects that the amplitude of the high-frequency component is large, it means that the power change in a short period of time is more drastic. At this time, the weight of the high-frequency power fluctuation component is increased, thereby enhancing the compensation ability for short-term power fluctuations; when the amplitude of the intermediate frequency or low frequency component is large, it indicates that the main source of power fluctuations is periodic oscillation or long-term trend changes. At this time, the weight of the intermediate frequency or low frequency compensation is correspondingly increased to optimize the power balance strategy.After calculating the power compensation demand, evaluate the current power stability and calculate the power stability index, which represents the ratio of the power compensation demand to the system rated power, thereby quantifying the impact of the current power fluctuation on the system stability. When the power stability index is high, it means that the system power fluctuation is small and the overall operation is relatively stable. When the power stability index is low, it means that the system power fluctuation is large and corresponding adjustment measures need to be taken to restore the power balance. Compare the power stability index with the preset threshold to determine whether power adjustment measures need to be taken. If the power stability index is lower than the preset threshold, it means that the current system power fluctuation has exceeded the safety range, affecting the normal operation of the system, and a power trigger signal is generated to start the power adjustment process.

[0028] In a specific embodiment, the process of executing step 102 may specifically include the following steps: The system power is expressed in complex form and the photovoltaic side power vector, the battery side power vector and the load side power vector are calculated according to the phase difference information to obtain the system power vector distribution; The damping power component and the synchronous power component are calculated according to the system power angle and its derivative, and the ratio of the damping power component to the synchronous power component is evaluated to obtain the system power oscillation risk determination result; According to the system power vector distribution and the system power oscillation risk determination results, a system state space model including a state variable vector of DC bus voltage, battery current, photovoltaic power and power angle and an input variable vector including photovoltaic reference voltage and battery reference current is established to obtain a system matrix and an input matrix; Calculate the eigenvalue of the system matrix, extract the real and imaginary parts from the eigenvalue, and obtain the system stability criterion; The power oscillation evaluation index is calculated based on the ratio of the maximum real part of the eigenvalue to the minimum absolute imaginary part, and the contribution value of each frequency power fluctuation component to the power oscillation evaluation index is calculated respectively to obtain a set of high-frequency, medium-frequency and low-frequency oscillation contribution values; The dominant frequency component of the oscillation is determined according to the set of high-frequency, medium-frequency and low-frequency oscillation contribution values, and the corresponding oscillation suppression strategy is selected according to the dominant frequency component of the oscillation, and the oscillation mode characteristic parameter set is output at the same time.

[0029] Specifically, the system power is expressed in complex form to more intuitively describe the power distribution on the photovoltaic side, battery side and load side. The complex representation method of active power and reactive power is adopted so that the power analysis takes into account both amplitude and phase information. The calculation of the power vector on the photovoltaic side is based on the phase angle between the photovoltaic output power and the voltage and current, while the power vector on the battery side is calculated in combination with the battery charging and discharging power and the dynamic characteristics of the energy storage system. At the same time, the calculation of the power vector on the load side depends on the power demand of the load and the dynamic change of the load. Through this method, the power flow relationship between photovoltaic power generation, energy storage unit and load is obtained, so as to construct a power vector distribution diagram. After obtaining the power vector distribution of the system, the power angle and its derivative are calculated to analyze the power oscillation characteristics of the system. The power angle reflects the relative phase between different power components, and the rate of change of the power angle is used to measure the intensity of power oscillation. In order to evaluate the power oscillation risk of the system, the damping power component and the synchronous power component are calculated. The damping power component describes how the system dissipates power disturbances, while the synchronous power component reflects the response characteristics of the system under power fluctuations. In order to determine the risk of power oscillation, the ratio of the damping power component to the synchronous power component is calculated, and the current power stability is judged based on this. When the damping power component is large, the system can effectively suppress power oscillation, while when the synchronous power component is large, the system is easily affected by external disturbances and produces unstable power oscillation. Through this analysis method, the severity of power oscillation is monitored in real time, and it is determined whether further adjustment measures are needed. Based on the above power vector distribution and power oscillation risk determination results, a state space model is established to describe the dynamic characteristics of the entire photovoltaic energy storage system. In this model, the state variable vector of the system contains key parameters such as DC bus voltage, battery current, photovoltaic power and power angle, while the input variable vector includes photovoltaic reference voltage and battery reference current. The state variable vector is used to describe the dynamic state of the system, while the input variable vector is used to control the power output of the system and the charging and discharging strategy of the energy storage unit. In order to construct a complete state space model, a state equation is defined to describe the relationship between the state variables and time, and an input equation is defined to characterize the influence of the input variables on the system state. After constructing the state space model, the eigenvalues ​​of the system matrix are calculated, and the real and imaginary parts are extracted from the eigenvalues ​​to determine the stability of the system. The real part of the eigenvalue determines the stability of the system. When the real part is positive, the system is in an unstable state, and when the real part is negative, the system tends to be stable. At the same time, the imaginary part of the eigenvalue reflects the oscillation characteristics of the system. The larger its absolute value, the higher the oscillation frequency of the system. In order to quantify the intensity of power oscillation, the power oscillation evaluation index is calculated based on the ratio of the maximum real part of the eigenvalue to the minimum absolute value of the imaginary part. This index can provide an intuitive measurement standard to indicate the current severity of the system's oscillation. Calculate the contribution values ​​of the high-frequency, medium-frequency and low-frequency power fluctuation components respectively, and construct an oscillation contribution value set.The contribution value of the high-frequency component reflects the power disturbance in a short period of time, the contribution value of the medium-frequency component reflects the periodic oscillation of the system, and the contribution value of the low-frequency component reflects the impact of long-term power imbalance. Through this step, the power oscillation components in different frequency ranges are accurately identified to determine the dominant frequency component of the oscillation. Based on the set of oscillation contribution values, the current dominant frequency component of the oscillation is determined, and a suitable oscillation suppression strategy is selected based on the component. When the high-frequency component is dominant, a fast-response power compensation strategy is adopted to suppress power disturbances in a short period of time, and when the medium-frequency component is dominant, a smooth transition power allocation strategy is adopted to reduce the impact of periodic power fluctuations. When the low-frequency component dominates the oscillation, a long-time constant power filtering strategy is adopted to optimize the long-term stability of the system. Output oscillation mode characteristic parameter set, including oscillation mode characteristics, oscillation frequency characteristics and oscillation phase characteristics.

[0030] In a specific embodiment, the execution step determines the dominant frequency component of the oscillation according to the high-frequency, medium-frequency and low-frequency oscillation contribution value sets, and selects the corresponding oscillation suppression strategy according to the dominant frequency component of the oscillation, and the process of outputting the oscillation mode characteristic parameter set can specifically include the following steps: Compare and normalize the high-frequency, medium-frequency and low-frequency oscillation contribution value sets, and determine the frequency component corresponding to the maximum contribution value as the oscillation dominant frequency component; The strategy matching is performed based on the dominant frequency component of the oscillation. When the high-frequency component is dominant, a fast-response power compensation strategy is selected. When the intermediate-frequency component is dominant, a smooth-transition power allocation strategy is selected. When the low-frequency component is dominant, a long-time constant power filtering strategy is selected to obtain an oscillation suppression strategy. Perform singular value decomposition on the power fluctuation time domain signal corresponding to the dominant frequency component to obtain the oscillation modal eigenvalue; perform fast Fourier transform on the power fluctuation signal of the dominant frequency component to obtain the oscillation frequency eigenvalue; calculate the phase difference between the dominant frequency component and the adjacent frequency component to obtain the oscillation phase eigenvalue; The oscillation mode eigenvalue, the oscillation frequency eigenvalue and the oscillation phase eigenvalue are combined into an oscillation mode characteristic parameter set.

[0031] Specifically, the high-frequency, medium-frequency and low-frequency oscillation contribution value sets are analyzed, and the proportion of each frequency component in the overall oscillation is determined. Since power fluctuations of different frequencies have different effects on system stability, the high-frequency, medium-frequency and low-frequency oscillation contribution values ​​are normalized during oscillation analysis to eliminate the differences in numerical scales and ensure that each frequency component can be compared under the same standard. The normalized data more accurately reflects the relative influence of each oscillation component and avoids misjudgment due to different data magnitudes. After the normalization process is completed, the contribution values ​​of the high-frequency, medium-frequency and low-frequency components are compared, and the frequency component corresponding to the maximum contribution value is determined as the current dominant frequency component of the oscillation. If the high-frequency component has the largest contribution value, it means that the power fluctuation of the current system is mainly caused by rapid disturbances in a short period of time, and if the medium-frequency component has the largest contribution value, it means that the power fluctuation of the system has a certain periodicity and is affected by external periodic loads or weather changes, and if the low-frequency component has the largest contribution value, it means that the power fluctuation of the system mainly comes from long-term power imbalance or slow regulation response of the energy storage system. Strategy matching is performed based on the dominant frequency component to select the most suitable oscillation suppression method. If the high-frequency component is dominant, it means that the system needs to respond quickly to quickly suppress short-term power fluctuations. A fast-response power compensation strategy is selected to quickly adjust the charging and discharging power of the energy storage system to weaken the impact of high-frequency power disturbances. For example, the short-term discharge power of the energy storage system is increased to compensate for the power gap and reduce the fluctuation amplitude of the DC bus voltage. When the intermediate-frequency component is dominant, it indicates that the power fluctuation of the system mainly presents periodic changes. At this time, direct rapid compensation will lead to system instability. Therefore, a smooth transition power allocation strategy is selected to make the power regulation process smoother. For example, a slow power adjustment method is adopted to reduce the system's sensitivity to periodic disturbances and reduce the power oscillation caused during the regulation process. When the low-frequency component is dominant, it means that the power fluctuation of the system is long-term trending. A long-time constant power filtering strategy is adopted to slowly adjust the system's power allocation, thereby optimizing the power balance and ensuring the long-term stability of the system. For example, a long-term power balance strategy is adopted to gradually increase the discharge of the energy storage system, while optimizing the energy allocation strategy on the load side to adapt to the long-term change trend of photovoltaic output. Analyze the power fluctuation characteristics corresponding to the dominant frequency component to extract the characteristic parameter set of the oscillation mode. In order to obtain the modal characteristics of the oscillation, the power fluctuation time domain signal corresponding to the dominant frequency component is subjected to singular value decomposition to obtain the oscillation modal eigenvalue. The singular value decomposition method can effectively extract the main modal information in the power fluctuation signal and remove irrelevant noise, so that the oscillation modal eigenvalue can accurately characterize the main power oscillation mode of the system. In order to analyze the frequency characteristics of the oscillation, the power fluctuation signal of the dominant frequency component is subjected to fast Fourier transform to obtain the oscillation frequency eigenvalue.Fast Fourier transform converts the time domain signal to the frequency domain and analyzes the main frequency component of the power fluctuation, thereby quantifying the oscillation frequency characteristics of the system. In order to analyze the phase characteristics of the power fluctuation, the phase difference between the dominant frequency component and the adjacent frequency component is calculated to obtain the oscillation phase eigenvalue. The calculation of the phase difference can reflect the relative timing relationship between different frequency components, thereby revealing the propagation characteristics of power oscillation in the time domain. For example, in a photovoltaic off-grid system, the power fluctuations of different frequency components are detected to have a large phase offset, which indicates that the power regulation strategy of the system needs to be optimized to reduce the delayed response of the energy storage system to the photovoltaic power change, thereby improving the stability of the overall system. If the phase difference between adjacent frequency components is large, it means that the power oscillation presents a strong phase change characteristic in the system, and the power regulation strategy needs to be further optimized to reduce the impact of the phase offset on the system stability. If the phase difference is small, it means that the power fluctuation of the system is relatively synchronized, and the overall stability of the system can be enhanced through appropriate compensation strategies. The oscillation mode eigenvalue, oscillation frequency eigenvalue and oscillation phase eigenvalue are combined into a complete oscillation mode characteristic parameter set.

[0032] In a specific embodiment, the process of executing step 103 may specifically include the following steps: The feedforward compensation amount is calculated according to the difference between the predicted photovoltaic output power and the current photovoltaic output power, and the feedforward gain is set for the feedforward compensation amount. The feedforward gain is dynamically adjusted according to the high-frequency oscillation contribution value in the oscillation mode characteristic parameter set to construct a photovoltaic side feedforward trigger channel. The feedback compensation amount is calculated based on the DC bus voltage deviation and its change rate, and the proportional gain and differential gain are set for the feedback compensation amount. The proportional gain and differential gain are dynamically adjusted according to the contribution values ​​of the medium-frequency and low-frequency oscillations in the oscillation mode characteristic parameter set to construct the energy storage side feedback trigger channel; The photovoltaic side feedforward trigger channel is used to generate a photovoltaic side trigger instruction according to the comparison result between the absolute value of the feedforward compensation amount and the feedforward threshold or its change rate and the change rate threshold; The energy storage side feedback trigger channel is used to generate an energy storage side trigger instruction according to the comparison result of the absolute value of the feedback compensation amount and the feedback threshold or the judgment result that the DC bus voltage deviation exceeds the safety range; The photovoltaic side trigger command and the energy storage side trigger command are combined, and the coordination coefficient is dynamically adjusted according to the oscillation suppression strategy to obtain the coordinated trigger command.

[0033] Specifically, a feedforward trigger channel is constructed on the photovoltaic side so that the feedforward compensation can be calculated based on the difference between the predicted photovoltaic output power and the current photovoltaic output power. The predicted photovoltaic output power is obtained by fitting the historical power data with a multi-order polynomial and dynamically updating it in combination with the recursive least squares method, which can reflect the power trend in the short term in the future. When there is a significant difference between the predicted value and the current photovoltaic output power, it means that the power balance of the system has changed and the power output on the photovoltaic side needs to be adjusted. The feedforward compensation is determined by this power deviation and directly affects the regulation ability of the photovoltaic system. Therefore, the feedforward gain is set to adjust the compensation intensity according to different power fluctuations. The size of the feedforward gain is not fixed, but is dynamically adjusted according to the high-frequency oscillation contribution value in the oscillation mode characteristic parameter set. If the high-frequency oscillation contribution value is large, it means that the power fluctuation of the system is mainly caused by short-term disturbances. At this time, the feedforward gain should be appropriately increased to enhance the response capability to high-frequency power fluctuations, so that the photovoltaic power can be adjusted quickly, thereby suppressing the impact of short-term power disturbances. If the high-frequency oscillation contribution value is low, it indicates that the power fluctuation of the system is mainly affected by the medium and low frequency components. At this time, the feedforward gain should be appropriately reduced to avoid over-compensation and system instability. At the same time, in order to optimize power regulation, a feedback trigger channel on the energy storage side is constructed, which calculates the feedback compensation based on the DC bus voltage deviation and its change rate. The DC bus voltage deviation can reflect the power balance of the current system, and the change rate of the voltage deviation further describes the dynamic characteristics of power regulation. When the bus voltage deviation is large, it means that there is a large deviation in the power matching between photovoltaic power generation, energy storage unit and load, which is compensated by charging and discharging the energy storage system. In order to ensure that the compensation amount can adapt to different power fluctuations, the proportional gain and differential gain are set for the feedback compensation amount. Among them, the role of the proportional gain is to directly adjust the power output of the energy storage system according to the size of the voltage deviation, while the differential gain can predict the future power trend according to the change rate of the voltage deviation and respond in advance. In order to optimize the regulation effect of the feedback channel, the size of the proportional gain and differential gain is dynamically adjusted according to the contribution values ​​of the medium and low frequency oscillations in the oscillation mode characteristic parameter set. When the contribution value of the intermediate frequency oscillation is large, it indicates that the power fluctuation of the system is mainly manifested as periodic oscillation. At this time, the proportional gain should be appropriately increased to enhance the regulation ability of periodic power fluctuations, while the differential gain should be appropriately reduced to avoid the energy storage system from excessively following the power fluctuations, resulting in oscillations during the regulation process. When the contribution value of the low-frequency oscillation is large, it indicates that the power fluctuation of the system is mainly caused by long-term power imbalance. At this time, the differential gain should be appropriately increased to enhance the system's adaptability to long-term power changes, while the proportional gain should be relatively reduced to ensure a smoother power regulation process. The photovoltaic side feedforward trigger channel and the energy storage side feedback trigger channel are used to generate trigger instructions for the photovoltaic side and the energy storage side respectively.For the generation of the trigger instruction on the photovoltaic side, it is determined whether the absolute value of the feedforward compensation exceeds the set feedforward threshold, or whether its rate of change exceeds the rate of change threshold. If the absolute value of the feedforward compensation exceeds the feedforward threshold, it means that the current photovoltaic power regulation demand is large. At this time, the power adjustment instruction on the photovoltaic side is triggered to make the photovoltaic output power respond quickly to the change in power demand. Similarly, if the rate of change of the feedforward compensation exceeds the rate of change threshold, it means that the photovoltaic power changes rapidly. At this time, the power adjustment instruction on the photovoltaic side is triggered to prevent the power change from affecting the stability of the system too fast. For example, in a photovoltaic energy storage system, when sudden cloud cover causes the photovoltaic output power to drop in a short period of time, and the system's power forecast value still maintains a high level, the absolute value of the feedforward compensation will increase rapidly and exceed the set feedforward threshold. At this time, the photovoltaic side trigger channel will immediately issue an instruction to adjust the reference voltage of the photovoltaic MPPT controller to adjust the photovoltaic output power to a new steady state. In the process of generating the trigger instruction on the energy storage side, the judgment is made based on the comparison result of the absolute value of the feedback compensation and the feedback threshold, and whether the DC bus voltage deviation exceeds the safety range. If the absolute value of the feedback compensation exceeds the set feedback threshold, it means that the current energy storage system needs a large charge and discharge adjustment to maintain the stability of the DC bus voltage, so the power adjustment command on the energy storage side is triggered to increase the responsiveness of the energy storage system. If the DC bus voltage deviation exceeds the safe range, it means that there is a large imbalance in the power distribution of the current system. At this time, the energy storage side needs to be adjusted immediately to avoid voltage fluctuations affecting the normal operation of the system. For example, when the DC bus voltage changes beyond the safe range due to a sudden load change, the energy storage side feedback channel will immediately detect this change and trigger the discharge command of the energy storage system to compensate for the bus voltage fluctuations and restore the system to stability. After the trigger instructions on the photovoltaic side and the energy storage side are generated, the two trigger instructions are combined, and the coordination coefficient is dynamically adjusted according to the oscillation suppression strategy to obtain the final coordinated trigger instruction. The adjustment of the coordination coefficient needs to take into account the oscillation characteristics of the current system. When the contribution value of high-frequency oscillation is large, the system increases the weight of the photovoltaic side to enhance the responsiveness to high-frequency power fluctuations. When the contribution value of intermediate-frequency and low-frequency oscillations is large, the system increases the weight of the energy storage side to enhance the system's compensation ability for long-term power fluctuations.

[0034] In a specific embodiment, the process of executing step 104 may specifically include the following steps: The coordination trigger instruction is multiplied by the power proportional factor to obtain the total power regulation demand; Establishing an optimization objective function taking into account a response speed objective function and an energy loss objective function based on the total power regulation demand; Under the constraints of photovoltaic output power and battery state of charge, the optimization objective function is optimized to obtain photovoltaic power regulation instructions and energy storage power regulation instructions; According to the photovoltaic side power regulation instruction and the photovoltaic output characteristic curve, a table lookup mapping is performed to calculate the maximum power point voltage and the offset, and the maximum power point voltage and the offset are added to obtain the first reference value of the photovoltaic MPPT controller; According to the power regulation instruction on the energy storage side divided by the battery voltage value, a second reference value of the battery charge and discharge controller is calculated; The first reference value and the second reference value are slope limited, the change rate of the first reference value is limited within the maximum voltage change rate range, and the change rate of the second reference value is limited within the maximum current change rate range, so as to obtain the voltage reference value of the photovoltaic MPPT controller and the current reference value of the battery charge and discharge controller.

[0035] Specifically, the coordination trigger instruction is converted to determine the total power regulation demand. According to the real-time operating status of the photovoltaic side and the energy storage side, the coordination trigger instruction is multiplied by the power proportional factor to ensure that the power regulation demand adapts to different load conditions and power generation environments. The setting of the power proportional factor needs to consider the rated power range of the photovoltaic system and the energy storage system, as well as the current operating status of the system, to avoid power fluctuations caused by over-regulation or voltage deviations caused by insufficient regulation. After this conversion process, the total power regulation demand is obtained. Based on the total power regulation demand, an optimization objective function considering the response speed objective function and the energy loss objective function is established to optimize the efficiency of power regulation while meeting the system operation constraints. The optimization objective function comprehensively considers two key factors, namely, response speed and energy loss. The response speed is a measure of the time required for the system to adjust to a new stable state when power fluctuations occur. In order to improve the dynamic response performance of the system, the optimization objective function contains a term that can minimize the adjustment time. The energy loss is a measure of the additional loss generated by the system during the power regulation process. In order to improve the overall energy efficiency, the objective function contains a term that minimizes the energy loss. Based on these two goals, the optimization objective function is constructed to allocate the power regulation tasks on the photovoltaic side and the energy storage side in the best way while ensuring the stability of the system, so as to reduce unnecessary energy loss while ensuring the regulation effect. The formula is as follows:

[0036] in, is the optimization objective function, Represents the power regulation of the system, Represents the amount of energy loss, and are weight coefficients used to balance response speed and energy loss, and is the adjustment time range. The solution of the optimization objective function needs to be carried out under certain constraints to ensure that the calculated power adjustment scheme meets the actual operating limits of the system. When solving the optimization objective function, the constraints of photovoltaic output power are considered. The output power of the photovoltaic system is affected by the light intensity, temperature and equipment characteristics. Therefore, the photovoltaic side power adjustment instruction cannot exceed the maximum allowable output power range of the photovoltaic array, otherwise it will cause MPPT (maximum power point tracking) control failure or overload operation of the photovoltaic power generation equipment. At the same time, the charging and discharging capacity of the energy storage system is limited by the battery state of charge. Therefore, the power adjustment instruction on the energy storage side needs to ensure that the battery state of charge does not exceed the safe range. If the battery state of charge is close to the upper limit, the system should reduce the charging power to prevent overcharging and damage to the battery. If the state of charge is too low, the discharge power should be reduced to ensure that the battery can maintain normal operation. Under the condition of meeting these constraints, the optimization solution algorithm is used to calculate the optimal photovoltaic side power adjustment instruction and the energy storage side power adjustment instruction, so as to ensure that the system can achieve the best energy efficiency and stability during the power adjustment process. After obtaining the photovoltaic side power regulation instruction, the voltage reference value of the photovoltaic MPPT controller is calculated according to the photovoltaic output characteristic curve. Since the maximum power point voltage of the photovoltaic system is not fixed, but is affected by the light intensity and temperature changes, a table lookup mapping is performed to obtain the maximum power point voltage value under the current environmental conditions. At the same time, considering that the photovoltaic power regulation instruction may cause the photovoltaic output power to be adjusted, a voltage offset is calculated to ensure that the photovoltaic system can maintain the operating state near the maximum power point while adjusting the power output. The maximum power point voltage value is added to the calculated offset to obtain the first reference value of the photovoltaic MPPT controller, thereby ensuring that the photovoltaic system can operate according to the new power regulation requirements. At the same time, according to the energy storage side power regulation instruction, the current reference value of the battery charge and discharge controller is calculated. Since the battery charge and discharge power is directly related to the battery voltage, the energy storage side power regulation instruction is divided by the current battery voltage value to calculate the actual charge and discharge current demand. Through this calculation method, it is ensured that the battery charge and discharge process matches the power regulation instruction, thereby optimizing the operating state of the energy storage system and improving the overall energy utilization efficiency of the system. For example, when the system detects a fluctuation in photovoltaic power, it will calculate the charging and discharging requirements of the energy storage system, and calculate the corresponding charging and discharging current through the battery voltage, thereby adjusting the working state of the energy storage system to ensure the stability of the DC bus voltage. In order to prevent excessively rapid voltage or current changes during power regulation, which may affect the stability of the system, the calculated photovoltaic MPPT controller voltage reference value and battery charge and discharge controller current reference value are slope-limited. The maximum voltage change rate is set to ensure that the voltage reference value of the photovoltaic MPPT controller does not change suddenly, thereby avoiding unstable power output of the photovoltaic system due to excessively rapid voltage changes.Similarly, the maximum current change rate is set to ensure that the current reference value of the battery charge and discharge controller is adjusted smoothly within a reasonable range, thereby preventing the energy storage system from reducing battery life or system oscillation due to excessively fast current changes. Through this slope limiting process, the stability of the power regulation process is guaranteed, and the operating stability and safety of the entire photovoltaic energy storage system are improved.

[0037] Among them, the optimization objective function is optimized under the constraints of photovoltaic output power and battery state of charge, and the photovoltaic side power regulation instructions and the energy storage side power regulation instructions are obtained, including: converting the optimization objective function into a distributed form and establishing a directed network topology, in which the photovoltaic side nodes and the energy storage side nodes are used as resource allocation units, and the state of each node is initialized to obtain an initial power allocation state vector; based on the initial power allocation state vector, a nonlinear distributed update law that converges in a predefined time is constructed, wherein the update law contains a non-homogeneous function with an exponential term to meet the user's preset convergence time requirements, and a time-varying distributed update function is obtained; according to the time-varying distributed update function, the state of each node is iteratively updated, and the candidate value of the power regulation instruction in each round of iteration is calculated, and the convergence error before and after the iteration is compared to obtain the convergence performance evaluation result; the convergence performance evaluation result is analyzed, and the convergence time parameter is dynamically adjusted based on the current system state to ensure the convergence of photovoltaic and energy storage. The convergence process is optimized while meeting the response time requirements of the energy system to obtain optimized convergence parameters; an auxiliary correction system is designed to monitor the satisfaction of the global power balance constraints in real time, and the correction amount is calculated when a constraint deviation occurs, and the correction amount is distributed to each node to obtain a correction value that satisfies the global balance constraint; the correction value is applied to the candidate value of the power adjustment instruction, and the communication topology between nodes is dynamically reconstructed according to the actual operation status of the system to adapt to the rapid changes in light intensity and load demand, so as to obtain an optimized power distribution structure; a predefined time consistency analysis is applied to the power distribution structure to ensure that each node reaches the optimal consistent state within the user-specified time, avoid the incoordination between photovoltaic output and energy storage response, and obtain a power distribution result guaranteed by time determinism; the final power adjustment ratio of the photovoltaic side and the energy storage side is determined according to the power distribution result, and the ratio is quantified and converted into a discrete control signal to obtain the final photovoltaic side power adjustment instruction and the energy storage side power adjustment instruction.

[0038] The above describes the power regulation method of the photovoltaic energy storage system in the embodiment of the present invention. The following describes the power regulation device of the photovoltaic energy storage system in the embodiment of the present invention. Figure 2 , an embodiment of a power regulating device of a photovoltaic energy storage system in an embodiment of the present invention includes: The real-time acquisition module 201 is used to perform real-time acquisition and wavelet decomposition of the voltage and current parameters of the photovoltaic array to obtain power fluctuation components of different frequencies and a power trigger signal; A power vector analysis module 202, configured to perform power vector analysis according to power fluctuation components of different frequencies and a power trigger signal, and obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy; Feedback trigger module 203, used to construct a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel based on an oscillation mode characteristic parameter set and an oscillation suppression strategy, and obtain coordinated trigger instructions for the photovoltaic side and the energy storage side; The processing module 204 is used to perform multi-objective optimization decomposition and slope limiting processing on the coordinated trigger instruction to obtain a voltage reference value of the photovoltaic MPPT controller and a current reference value of the battery charge and discharge controller.

[0039] Through the coordinated cooperation of the above-mentioned components, through multi-order polynomial fitting and wavelet decomposition technology, accurate identification and prediction of power fluctuation components of different frequencies are achieved. The quantitative analysis method based on the power vector analysis model clarifies the influence mechanism of each frequency component on the system stability. The design of the dual-channel power trigger mechanism realizes the coordinated cooperation of photovoltaic side feedforward and energy storage side feedback, and effectively suppresses power fluctuations without calculating complex power compensation values ​​on the photovoltaic side. The multi-objective optimized power allocation strategy takes into account both response speed and energy utilization efficiency, and realizes the optimal allocation of system resources. Differentiated processing strategies are adopted for power fluctuations with different frequency characteristics to avoid the risk of power oscillation caused by control mode switching. The introduction of system operation status evaluation and control strategy optimization mechanism ensures that the system maintains the best performance state during long-term operation, and significantly improves the power stability and energy utilization efficiency of photovoltaic energy storage systems under extreme conditions such as drastic changes in light or sudden changes in load.

[0040] above Figure 2 The power regulation device of the photovoltaic energy storage system in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The power regulation device of the photovoltaic energy storage system in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0041] Figure 33 is a schematic diagram of the structure of a power regulating device of a photovoltaic energy storage system provided by an embodiment of the present invention. The power regulating device 300 of the photovoltaic energy storage system may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage device terminals) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 can be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the power regulating device 300 of the photovoltaic energy storage system. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the power regulating device 300 of the photovoltaic energy storage system to implement the steps of the power regulating method of the photovoltaic energy storage system.

[0042] The power conditioning device 300 of the photovoltaic energy storage system may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It can be understood by those skilled in the art that Figure 3 The power regulating device structure of the photovoltaic energy storage system shown does not constitute a limitation on the power regulating device of the photovoltaic energy storage system provided by the present invention, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0043] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the power regulation method of the photovoltaic energy storage system.

[0044] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0045] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0046] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A power regulation method for a photovoltaic energy storage system, characterized in that: The method comprises: The voltage and current parameters of the photovoltaic array are collected in real time and decomposed by wavelet to obtain power fluctuation components of different frequencies and power trigger signals; Performing power vector analysis according to the power fluctuation components of different frequencies and the power trigger signal to obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy; Based on the oscillation mode characteristic parameter set and the oscillation suppression strategy, a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel are constructed to obtain coordinated trigger instructions for the photovoltaic side and the energy storage side; The coordinated trigger instruction is subjected to multi-objective optimization decomposition and slope limitation processing to obtain a voltage reference value of a photovoltaic MPPT controller and a current reference value of a battery charge and discharge controller.

2. The power regulation method of the photovoltaic energy storage system according to claim 1, characterized in that: The real-time acquisition and wavelet decomposition of the voltage and current parameters of the photovoltaic array to obtain power fluctuation components of different frequencies and power trigger signals include: The output voltage and current of the photovoltaic array are collected by a voltage sensor and a current sensor arranged at the output end of the photovoltaic array, and the output voltage and the output current are multiplied to obtain the photovoltaic output power; The DC bus voltage value is collected by a voltage sensor arranged on the DC bus, and a difference calculation is performed between the DC bus voltage value and a preset DC bus voltage reference value to obtain a DC bus voltage deviation; Performing adaptive bandpass filtering on the DC bus voltage deviation to obtain a power fluctuation characteristic signal; Based on the photovoltaic output power and the power fluctuation characteristic signal, multi-order polynomial fitting and wavelet decomposition are performed to obtain power fluctuation components of different frequencies and a power trigger signal.

3. The power regulation method of the photovoltaic energy storage system according to claim 2, characterized in that: The method of performing multi-order polynomial fitting and wavelet decomposition based on the photovoltaic output power and the power fluctuation characteristic signal to obtain power fluctuation components of different frequencies and a power trigger signal includes: Performing fifth-order polynomial fitting processing on the historical data sequence of the photovoltaic output power, and updating the polynomial coefficients by recursive least squares method to obtain a predicted value of the photovoltaic output power in a future sampling period; Performing weighted moving average and exponential smoothing processing on the photovoltaic output power forecast value of the future sampling period, and assigning weight coefficients for weighted calculation to obtain a medium-term power forecast value; According to the mid-term power prediction value, the power fluctuation characteristic signal is subjected to wavelet decomposition processing to extract high-frequency components, medium-frequency components and low-frequency components respectively, so as to obtain power fluctuation components of different frequencies; Based on the amplitudes of the power fluctuation components of different frequencies, an adaptive weight coefficient is allocated to the power fluctuation components of each frequency and weighted sum is performed to obtain a power compensation demand; Calculating the power stability index based on the ratio of the power compensation demand to the system rated power; The power stability index is compared with a preset threshold, and a power trigger signal is generated when the power stability index is lower than the preset threshold.

4. The power regulation method of the photovoltaic energy storage system according to claim 1, characterized in that: The performing power vector analysis according to the power fluctuation components of different frequencies and the power trigger signal to obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy includes: The system power is expressed in complex form and the photovoltaic side power vector, the battery side power vector and the load side power vector are calculated according to the phase difference information to obtain the system power vector distribution; Calculating the damping power component and the synchronous power component according to the system power angle and its derivative, and evaluating the ratio of the damping power component to the synchronous power component to obtain a system power oscillation risk determination result; A system state space model including a state variable vector including a DC bus voltage, a battery current, a photovoltaic power and a power angle and an input variable vector including a photovoltaic reference voltage and a battery reference current is established according to the system power vector distribution and the system power oscillation risk determination result to obtain a system matrix and an input matrix; Calculating eigenvalues ​​of the system matrix, and extracting real and imaginary parts from the eigenvalues ​​to obtain a system stability criterion; The power oscillation evaluation index is calculated based on the ratio of the maximum value of the real part to the minimum value of the absolute value of the imaginary part of the eigenvalue, and the contribution value of each frequency power fluctuation component to the power oscillation evaluation index is calculated respectively to obtain a set of high-frequency, medium-frequency and low-frequency oscillation contribution values; The dominant frequency component of oscillation is determined according to the high-frequency, medium-frequency and low-frequency oscillation contribution value sets, and the corresponding oscillation suppression strategy is selected according to the dominant frequency component of oscillation, and the oscillation mode characteristic parameter set is outputted at the same time.

5. The power regulation method of the photovoltaic energy storage system according to claim 4, characterized in that: The method of determining the dominant frequency component of oscillation according to the high-frequency, medium-frequency and low-frequency oscillation contribution value sets, selecting a corresponding oscillation suppression strategy according to the dominant frequency component of oscillation, and outputting an oscillation mode characteristic parameter set at the same time includes: Comparing and normalizing the high-frequency, medium-frequency and low-frequency oscillation contribution value sets, and determining the frequency component corresponding to the maximum contribution value as the oscillation dominant frequency component; Based on the dominant frequency component of the oscillation, strategy matching is performed. When the high-frequency component is dominant, a fast-response power compensation strategy is selected. When the intermediate-frequency component is dominant, a smooth-transition power allocation strategy is selected. When the low-frequency component is dominant, a long-time constant power filtering strategy is selected to obtain an oscillation suppression strategy. Performing singular value decomposition on the power fluctuation time domain signal corresponding to the dominant frequency component to obtain the oscillation modal eigenvalue; performing fast Fourier transform on the power fluctuation signal of the dominant frequency component to obtain the oscillation frequency eigenvalue; calculating the phase difference between the dominant frequency component and the adjacent frequency component to obtain the oscillation phase eigenvalue; The oscillation modal eigenvalue, the oscillation frequency eigenvalue and the oscillation phase eigenvalue are combined into an oscillation mode characteristic parameter set.

6. The power regulation method of the photovoltaic energy storage system according to claim 4, characterized in that: The step of constructing a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel based on the oscillation mode characteristic parameter set and the oscillation suppression strategy to obtain a coordinated trigger instruction for the photovoltaic side and the energy storage side includes: A feedforward compensation amount is calculated according to the difference between the predicted photovoltaic output power and the current photovoltaic output power, and a feedforward gain is set for the feedforward compensation amount, wherein the feedforward gain is dynamically adjusted according to the high-frequency oscillation contribution value in the oscillation mode characteristic parameter set, so as to construct a photovoltaic side feedforward trigger channel; A feedback compensation amount is calculated based on the DC bus voltage deviation and its change rate, and a proportional gain and a differential gain are set for the feedback compensation amount, wherein the proportional gain and the differential gain are dynamically adjusted according to the medium-frequency and low-frequency oscillation contribution values ​​in the oscillation mode characteristic parameter set to construct a storage side feedback trigger channel; Using the photovoltaic side feedforward trigger channel, generating a photovoltaic side trigger instruction according to a comparison result between the absolute value of the feedforward compensation amount and the feedforward threshold or its change rate and the change rate threshold; The energy storage side feedback trigger channel is used to generate an energy storage side trigger instruction according to a comparison result of the absolute value of the feedback compensation amount and the feedback threshold or a judgment result that the DC bus voltage deviation exceeds a safety range; The photovoltaic side trigger instruction and the energy storage side trigger instruction are combined, and the coordination coefficient is dynamically adjusted according to the oscillation suppression strategy to obtain a coordinated trigger instruction.

7. The power regulation method of the photovoltaic energy storage system according to claim 1, characterized in that: The coordinated trigger instruction is subjected to multi-objective optimization decomposition and slope limiting processing to obtain a voltage reference value of a photovoltaic MPPT controller and a current reference value of a battery charge and discharge controller, including: The coordination trigger instruction is multiplied by a power proportional factor to obtain a total power regulation demand; Establishing an optimization objective function taking into account a response speed objective function and an energy loss objective function based on the total power regulation demand; The optimization objective function is optimized under the photovoltaic output power constraint condition and the battery state of charge constraint condition to obtain the photovoltaic side power regulation instruction and the energy storage side power regulation instruction; Perform table lookup mapping according to the photovoltaic side power regulation instruction and the photovoltaic output characteristic curve, calculate the maximum power point voltage and the offset, and add the maximum power point voltage to the offset to obtain a first reference value of the photovoltaic MPPT controller; Calculate a second reference value of the battery charge and discharge controller according to the energy storage side power adjustment instruction divided by the battery voltage value; The first reference value and the second reference value are slope limited, the change rate of the first reference value is limited within the maximum voltage change rate range, and the change rate of the second reference value is limited within the maximum current change rate range, so as to obtain the voltage reference value of the photovoltaic MPPT controller and the current reference value of the battery charge and discharge controller.

8. A power regulating device for a photovoltaic energy storage system, characterized in that: Used to execute the power regulation method of the photovoltaic energy storage system according to any one of claims 1 to 7, the power regulation device of the photovoltaic energy storage system comprises: Real-time acquisition module, used for real-time acquisition and wavelet decomposition of the voltage and current parameters of the photovoltaic array to obtain power fluctuation components of different frequencies and power trigger signals; A power vector analysis module, used to perform power vector analysis according to the power fluctuation components of different frequencies and the power trigger signal to obtain an oscillation mode characteristic parameter set and an oscillation suppression strategy; A feedback trigger module, used to construct a photovoltaic side feedforward trigger channel and an energy storage side feedback trigger channel based on the oscillation mode characteristic parameter set and the oscillation suppression strategy, and obtain coordinated trigger instructions for the photovoltaic side and the energy storage side; The processing module is used to perform multi-objective optimization decomposition and slope limiting processing on the coordinated trigger instruction to obtain a voltage reference value of the photovoltaic MPPT controller and a current reference value of the battery charge and discharge controller.

9. A power regulation device for a photovoltaic energy storage system, characterized in that: The power regulation device of the photovoltaic energy storage system includes: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instruction in the memory so that the power regulation device of the photovoltaic energy storage system executes the power regulation method of the photovoltaic energy storage system according to any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the power regulation method of the photovoltaic energy storage system according to any one of claims 1 to 7 is implemented.

Images