A method for compensating load imbalance of a power system with photovoltaic power grid

Abstract

The application relates to the technical field of power systems, in particular to a power system load imbalance compensation method for photovoltaic power grid connection, which comprises the following steps: collecting photovoltaic power generation power, voltage data and three-phase current data of photovoltaic power generation at the grid connection end in each period; collecting power data of each load end at each moment in the power grid; collecting voltage and impedance of each line at each moment in the power grid; determining the waveform difference index of three-phase current in each period; obtaining the grid connection influence coefficient of photovoltaic power in each period; determining the power supply consistency coefficient and the supply-demand matching coefficient in each period; obtaining the impedance fluctuation coefficient in each period; optimizing the droop coefficient in the droop control algorithm; and compensating the load imbalance of the power system by using the optimized droop control algorithm. The application improves the compensation effect of the load imbalance of the power system.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method for compensating for load imbalance in a power system in which photovoltaic power is connected to a grid. Background Art

[0002] With the transformation of the global energy mix and the rapid development of renewable energy, photovoltaic power generation, as a clean and sustainable form of energy, is gaining widespread adoption worldwide. The large-scale integration of photovoltaic power into the grid presents new opportunities for the power system, but also presents numerous challenges. However, due to the intermittent and fluctuating nature of photovoltaic power generation, its output power is significantly affected by weather conditions, leading to increasingly prominent load imbalances in the power system.

[0003] Droop control algorithms are widely used in microgrid power regulation because they can dynamically allocate power, providing additional compensation to phases with excessive or insufficient load, thereby alleviating load imbalance. However, existing technologies for allocating power to the grid using droop control algorithms typically employ traditional fixed-coefficient droop control strategies. These strategies fail to fully consider the relationship between the randomness of wind power supply and grid load fluctuations, resulting in poor load balancing of the power system due to the set droop control parameters. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides a method for compensating for load imbalance in a power system with photovoltaic power grid access, so as to solve the existing problems.

[0005] The present invention discloses a method for compensating for unbalanced load in a photovoltaic power grid-connected power system using the following technical solutions:

[0006] An embodiment of the present application provides a method for compensating for load imbalance in a power system in which photovoltaic power is connected to a grid, the method comprising the following steps:

[0007] Collect the generated power of photovoltaic power generation at each moment in each cycle, the voltage data and three-phase current data of photovoltaic power generation at the grid access end, the power data of each load end in the grid at each moment, and the voltage and impedance of each line in the grid at each moment;

[0008] The similarity between the voltage data distribution of photovoltaic power generation at the grid access end and the sinusoidal signal in each cycle is analyzed to obtain the sinusoidal similarity coefficient of each cycle. Based on the fluctuation difference between any two phases in the three-phase current data of each cycle, the waveform difference index of the three-phase current in each cycle is determined.

[0009] a first similarity between the A-phase current and the B-phase current at different time delays, a second similarity between the B-phase current and the C-phase current at different time delays, a first difference between the first similarity and the second similarity, and a waveform difference index, to determine a grid-connection influence coefficient of the photovoltaic power of each cycle;

[0010] a load power by accumulating power at all load ends at each time, a correlation between the power generation power and the load power of each cycle to determine a power supply consistency coefficient of each cycle;

[0011] a power supply sequence and a load sequence by respectively grouping the power generation power and the load power at all times of each cycle, all peaks in the power supply sequence and the load sequence are respectively obtained, a mean value of a difference between peaks of the same sequence number in the power supply sequence and the load sequence is taken as a peak difference index of each cycle, a difference between slopes of a fitting straight line of the power supply sequence and the load sequence of each cycle is taken as a trend difference index, and the power supply consistency coefficient is combined to determine a supply-demand matching coefficient of each cycle;

[0012] a voltage mean value and an impedance mean value of each line at all times in each cycle, a second difference between the voltage mean value and the impedance mean value of each line, a difference between a dispersion degree of the voltage mean value and a dispersion degree of the impedance mean value of all lines is calculated, and the difference is taken as a third difference, and the dispersion degree of the impedance mean value is combined to determine an impedance fluctuation coefficient of each cycle;

[0013] a product of the grid-connection influence coefficient and the impedance fluctuation coefficient of each cycle, a sum value of the product and a preset value greater than 0, a ratio of the supply-demand matching coefficient of each cycle to the sum value is taken as a droop stability factor of each cycle, a droop coefficient in a droop control algorithm is adjusted based on the droop stability factor of each cycle, and the load imbalance of the power system is compensated by using the optimized droop control algorithm.

[0014] In one embodiment, the determination of the sinusoidal similarity coefficient includes:

[0015] a voltage sequence is formed by grouping voltage data of photovoltaic power generation at all times of each cycle at the grid-connection end, frequency domain information of the voltage sequence is obtained by using a frequency domain transformation algorithm, a standard sinusoidal signal of the voltage sequence is constructed, and the sinusoidal similarity coefficient is a correlation coefficient between the voltage sequence and the standard sinusoidal signal.

[0016] In one embodiment, the determination process of the waveform difference index is:

[0017] The A-phase current, the B-phase current and the C-phase current at all time points of each cycle are respectively composed of an A-phase sequence, a B-phase sequence and a C-phase sequence, a metric distance between any two sequences of the A-phase sequence, the B-phase sequence and the C-phase sequence is calculated, and the waveform difference index is a mean value of all the metric distances of each cycle.

[0018] In one embodiment, the determination of the network entry influence coefficient comprises:

[0019] The network entry influence coefficient of each cycle is positively correlated with the waveform difference index and the first difference, and is negatively correlated with the sine similarity coefficient.

[0020] In one embodiment, the power supply consistency coefficient is a correlation coefficient of the power generation and the load power at all time points of each cycle.

[0021] In one embodiment, the determination of the supply-demand matching coefficient comprises:

[0022] The supply-demand matching coefficient of each cycle is positively correlated with the power supply consistency coefficient, and is negatively correlated with the peak difference index and the trend difference index.

[0023] In one embodiment, the determination of the impedance fluctuation coefficient comprises:

[0024] The impedance fluctuation coefficient is positively correlated with the dispersion degree of the impedance mean value of all lines, and is negatively correlated with the third difference and the dispersion degree of the second difference of all lines.

[0025] In one embodiment, the second difference is a ratio of the voltage mean value to the impedance mean value of each line.

[0026] In one embodiment, the adjusting of the droop coefficient in the droop control algorithm based on the droop stability factor of each cycle comprises:

[0027] The droop coefficient in the droop control algorithm of each cycle is a product of the addition result of the normalized value of the inverse of the droop stability factor of each cycle and a preset constant greater than 0 and less than 1, and the droop coefficient of the previous cycle of each cycle.

[0028] The present application has at least the following beneficial effects:

[0029] The application collects photovoltaic power generation power, voltage data and three-phase current data of photovoltaic power generation at the grid access end in each period, collects power data of each load end at each moment in the grid, and collects voltage and impedance of each line at each moment in the grid; by collecting multiple types of grid related data, the blind area of traditional single data source is avoided, and the reliability of grid data analysis is improved; the similarity degree of voltage data distribution of photovoltaic power generation at the grid access end in each period and a sine signal is analyzed to obtain a sine similarity coefficient of each period; the determination of the sine similarity coefficient reflects the voltage harmonic content, reduces the interference to sensitive equipment, and improves the accuracy of power quality evaluation; based on the fluctuation difference between any two phases in the three-phase current data of each period, a waveform difference index of three-phase current of each period is determined; the waveform difference index quantifies the three-phase current imbalance and accurately reflects the power quality of each period, providing a basis for subsequent load imbalance compensation; the similarity of A-phase current and B-phase current, and B-phase current and C-phase current in the three-phase current data of each period under different time delays is used to determine a grid access influence coefficient of photovoltaic power of each period in combination with the sine similarity coefficient and the waveform difference index; the grid access influence coefficient reflects the influence degree of photovoltaic power access in each period, identifies the correlation between photovoltaic fluctuation and grid oscillation, and improves the accuracy of droop coefficient correction in the droop control algorithm; the power of all load ends at each moment is accumulated and recorded as load power, the correlation between the power generation power and the load power of each period is analyzed, and a power supply consistency coefficient of each period is determined; based on the data trend difference and peak value difference between the power generation power and the load power of each period, in combination with the power supply consistency coefficient, a supply-demand matching coefficient of each period is determined; the supply-demand matching coefficient reflects the matching degree of photovoltaic power generation power and total load power in each period, reflects whether the photovoltaic power can better adapt to the load demand of the grid, and is helpful to optimize the droop coefficient in the droop control algorithm; the voltage mean value and impedance mean value of all lines at all moments in each period are obtained, the difference change of the voltage mean value and the impedance mean value of each line, and the dispersion degree of the voltage mean value and the impedance mean value are analyzed, and an impedance fluctuation coefficient of each period is determined; the impedance fluctuation coefficient reflects the influence degree of grid line impedance on the voltage in the line and the difference degree between line impedances, and improves the accuracy and reliability of droop coefficient adjustment in the droop control algorithm; the droop coefficient in the droop control algorithm is optimized by comprehensively considering the grid access influence coefficient, the supply-demand matching coefficient and the impedance fluctuation coefficient, the load imbalance of the power system is compensated by using the optimized droop control algorithm, the load distribution of the power system is dynamically adjusted, the load in the power system is accurately balanced and controlled, and the stability of the grid is improved. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0031] Figure 1 A step flow chart of a power system load imbalance compensation method for photovoltaic power grid connection provided by the present application is provided.

[0032] Figure 2 A droop coefficient regulation flow chart in the droop control algorithm. DETAILED DESCRIPTION

[0033] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purposes, the specific embodiments, structures, features and effects of the power system load imbalance compensation method for photovoltaic power grid connection according to the present application are described in detail as follows in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0035] The specific scheme of the power system load imbalance compensation method for photovoltaic power grid connection provided by the present application is specifically described below in combination with the accompanying drawings.

[0036] The power system load imbalance compensation method for photovoltaic power grid connection provided by one embodiment of the present application, specifically, provides a power system load imbalance compensation method for photovoltaic power grid connection as follows, please refer to Figure 1 The method comprises the following steps:

[0037] Step S001, collecting the power generation power of photovoltaic power generation, the voltage data and three-phase current data of photovoltaic power generation at the grid access end in each period and each time, collecting the power data of each load end at each time in the power grid, and collecting the voltage and impedance of each line at each time in the power grid.

[0038] The embodiment collects the power generation of the photovoltaic power generation in real time through the power analyzer installed at the output end of the photovoltaic array; collects the voltage data and three-phase current data of the photovoltaic power generation at the grid access end through the electric energy meter, collects the power data of each load end in the grid through the electric energy meter, and takes the sum of the power collected at all load ends at the same time as the total load power of the grid, which is briefly denoted as load power. The time interval of the collected power generation, voltage, three-phase current and power data of the load end is 2 milliseconds, and the data collection period is set to 10 seconds. The time interval of the collected data and the data collection period can be set by the implementer according to the actual situation, which is not limited in the embodiment.

[0039] According to the time sequence of data collection, the power generation at all times of each cycle is combined to form a power supply sequence, the voltage data of the photovoltaic power generation at the grid access end at all times of each cycle is combined to form a voltage sequence, the A-phase current in the three-phase current at all times of each cycle is combined to form an A-phase sequence, the B-phase current is combined to form a B-phase sequence, and the C-phase current is combined to form a C-phase sequence.

[0040] Further, the voltage of each line in the grid is obtained through the electric energy meter, the impedance of each line in the grid is obtained through the impedance analyzer, the average voltage and the average impedance of each line at all times in each cycle are calculated, and are denoted as the average voltage and the average impedance of each line in each cycle.

[0041] In order to eliminate the dimensional influence between the data, all the above sequences, and the average voltage and the average impedance are normalized, and the maximum-minimum value normalization processing method is adopted in the embodiment, and the implementer can select other feasible normalization methods, which is not limited in the embodiment.

[0042] In step S002, the similarity of the voltage data distribution of the photovoltaic power generation at the grid access end to the sine signal in each cycle is analyzed to obtain the sine similarity coefficient of each cycle; and the waveform difference index of the three-phase current in each cycle is determined based on the fluctuation difference between any two phases in the three-phase current data of each cycle.

[0043] The output power of the photovoltaic power generation is greatly affected by the light intensity, and has significant volatility and uncertainty. When the photovoltaic power is connected to the grid, harmonic influence is generated, and if the harmonic content is high, it will have a great impact on the stable operation of the grid, causing the grid voltage to fluctuate or flicker. At the same time, the load at the grid end also has strong randomness. If the harmonic content generated by the photovoltaic power generation system is high and does not match the change of the load demand at the grid end, it will exacerbate the load difference of the grid, and may further cause the load imbalance problem.

[0044] Normally, the voltage in the power grid and the three-phase current waveform conform to the sine function. If the harmonic content brought by photovoltaic power is large, it will have a serious impact on the power quality of the power grid, thereby causing the voltage and current waveform to be distorted. Therefore, the embodiment evaluates the degree of adverse impact of photovoltaic power on the power grid by analyzing the waveform of the voltage and the waveform change of the three-phase current.

[0045] The voltage sequence of each cycle is subjected to fast Fourier transform to convert the time domain signal into a frequency domain signal. A standard sine signal corresponding to the voltage sequence is constructed according to the fundamental frequency, phase and amplitude of the frequency domain voltage signal. The fast Fourier transform and signal generation technology are both known technologies, and will not be described in detail herein.

[0046] The Pearson correlation coefficient between the voltage sequence of each cycle and the corresponding standard sine signal is calculated as the sine similarity coefficient of each cycle. The sine similarity coefficient can reflect the waveform similarity between the voltage sequence and the corresponding standard sine signal. The greater the sine similarity coefficient, the more the waveform of the voltage sequence approximates to a sine wave.

[0047] It should be noted that the Pearson correlation coefficient is a known technology, and the implementer can select other available correlation coefficient calculation methods, such as the Spearman similarity coefficient and the cosine similarity.

[0048] Further, in a normal alternating current system, when the three-phase load is symmetrical and balanced, the phase difference between the three-phase currents is consistent, and the waveform amplitude is consistent. Due to the instability of photovoltaic power, the power provided at different times is inconsistent, which may exacerbate the imbalance of the three-phase load in the power grid. At the same time, the photovoltaic power into the grid also brings certain harmonic pollution, which may cause fluctuations in the waveform and phase difference between the three-phase currents, and inconsistencies.

[0049] Considering the influence of the phase difference of the three-phase current, the waveform change of the three-phase current has a certain sequence in time. Therefore, the embodiment calculates the DTW distance between any two sequences in the A-phase sequence, the B-phase sequence and the C-phase sequence of each cycle, and calculates the average of all DTW distances of each cycle. The average result is recorded as the waveform difference index of each cycle. The waveform difference index can reflect the difference between the three-phase current waveforms.

[0050] It should be noted that the embodiment only provides a calculation method for measuring the distance between sequences, and the implementer can select other available distance measurement calculation methods.

[0051] Step S003, the similarity of the A-phase current and the B-phase current at different time delays in the three-phase current data of each cycle, the similarity of the B-phase current and the C-phase current at different time delays, the sine similarity coefficient and the waveform difference index are combined to determine the grid-connection influence coefficient of the photovoltaic power of each cycle.

[0052] In this embodiment, the peak detection algorithm is used to obtain all the peaks in the A-phase sequence of each cycle, and a peak position sequence is constructed in ascending order according to the position of each peak in the A-phase sequence. A first-order difference sequence of the peak position sequence is obtained, and the mean value of the first-order difference sequence is taken as the cycle interval index of the A-phase sequence of each cycle. The cycle interval index can reflect how long the A-phase current takes to reach the peak value again. In this embodiment, the peak detection algorithm is an extreme value detection algorithm, which is a known technology. The implementer can select other feasible peak detection algorithms, and this embodiment does not limit this.

[0053] For each cycle, the cross-correlation coefficient of the A-phase sequence and the B-phase sequence at a time delay k is obtained in sequence by using the cross-correlation function, where , a is the cycle interval index of the A-phase sequence of each cycle, and k is sequentially added by 1. The time delay k corresponding to the maximum value in all cross-correlation coefficients is taken as the AB phase difference of the A-phase sequence and the B-phase sequence, which is taken as the first similarity. The AB phase difference reflects the delay time of the B-phase current relative to the A-phase current in time.

[0054] For the B-phase sequence of each cycle, the same calculation method as the cycle interval index of the A-phase sequence is used to obtain the cycle interval index of the B-phase sequence. Similarly, the same calculation method as the AB phase difference is used to obtain the BC phase difference of the B-phase sequence and the C-phase sequence, which is taken as the second similarity.

[0055] The grid-connection influence coefficient of the photovoltaic power of each cycle is determined by combining the sine similarity coefficient, the waveform difference index, and the difference between the first similarity and the second similarity. The specific calculation method is as follows:

[0056] In the formula, A is the grid-connection influence coefficient of the photovoltaic power of each cycle, B is the waveform difference index of each cycle, C is the absolute value of the difference between the AB phase difference and the BC phase difference , and P is the sine similarity coefficient of each cycle. is a preset value greater than 0. In order to avoid a denominator of 0, in this embodiment, the value of The implementer can set it according to the actual situation, and the embodiment does not limit it.

[0057] It should be noted that the difference represents the difference between the two variables, which can be calculated by the absolute value of the difference, the square of the difference, the ratio, etc. The grid impact index can reflect the impact on the grid after the photovoltaic power is connected to the grid. The greater the waveform difference index, the more different the waveforms of the three-phase currents are. The greater the first difference, the greater the phase difference between the three-phase currents. The smaller the sine similarity coefficient, the greater the difference between the voltage waveform and the sine waveform. Therefore, the greater the grid impact coefficient, the greater the adverse impact of photovoltaic power on the grid after being connected to the grid. At this time, greater adjustment capability is needed to adjust the impact of photovoltaic power on the grid.

[0058] In step S004, the correlation between the power generation power and the load power of each cycle is analyzed to determine the power supply consistency coefficient of each cycle. Based on the data trend difference and the peak value difference between the power generation power and the load power of each cycle, and in combination with the power supply consistency coefficient, the supply-demand matching coefficient of each cycle is determined.

[0059] Further, since the load is mostly single-phase load, and the power consumption property and power consumption time of the load are relatively random, if the supply of photovoltaic power is consistent with the demand load of the grid, the operation of the grid is relatively stable. If the supply of photovoltaic power and the demand load fluctuate greatly and there is no corresponding change relationship, the supply of photovoltaic power at a certain moment may be significantly higher or lower than the demand load of a certain phase, thereby causing a large difference between the demand load of the phase and other phases, further exacerbating the load imbalance of the grid.

[0060] Therefore, based on the above analysis, the embodiment groups the total load power of all moments of each cycle in time sequence to form a load sequence, calculates the Pearson correlation coefficient between the power supply sequence and the load sequence of each cycle as the power supply consistency coefficient of each cycle. The power supply consistency coefficient can reflect the degree of cooperation between the photovoltaic supply power and the demand load of the grid. The smaller the power supply consistency coefficient, the lower the matching degree between the photovoltaic supply power and the demand load. When the power supply consistency coefficient is negative, it means that the connection of photovoltaic power to the grid not only cannot alleviate the load imbalance problem of the grid, but also further exacerbates the possibility of load imbalance.

[0061] All the peak values in the power supply sequence of each period are obtained by an extreme value detection algorithm, and the peak values are sorted from small to large according to the bit sequence of the peak values in the power supply sequence to form a peak value sequence, and a first difference sequence of the peak value sequence is obtained, which is recorded as a peak value difference sequence of the power supply sequence. The peak value difference sequence can reflect the change amount of the photovoltaic power supply fluctuation; the greater the element value in the peak value difference sequence is, the more the photovoltaic power supply amount is in the process of gradually increasing. For the load sequence, the same acquisition method as the peak value difference sequence of the power supply sequence is adopted to obtain the peak value difference sequence of the load sequence.

[0062] For each period, the absolute value of the difference between the elements at the same position in the peak value difference sequence of the power supply sequence and the peak value difference sequence of the load sequence is calculated, which is recorded as a first difference absolute value. The average of all the first difference absolute values of each period is taken as the peak value difference index of each period. It should be noted that if the lengths of the peak value difference sequence of the power supply sequence and the peak value difference sequence of the load sequence are inconsistent, the tail of the shorter sequence is padded with 0 to make the lengths of the peak value difference sequence of the power supply sequence and the peak value difference sequence of the load sequence equal. The peak value difference index can reflect the fluctuation difference between the power supply sequence and the load sequence, and the greater the peak value difference index is, the greater the difference between the output power fluctuation of the photovoltaic power generation and the demand load fluctuation of the power grid, and the greater the possibility of mismatch between supply and demand.

[0063] Further, the power supply sequence and the load sequence are linearly fitted by a linear fitting algorithm, and the absolute value of the difference between the slopes of the two fitted straight lines is recorded as the trend difference index of each period. The trend difference index can reflect whether the data change rate and change direction between the power supply sequence and the load sequence are consistent.

[0064] The supply-demand matching coefficient of each period is determined by combining the power supply consistency coefficient, the peak value difference index and the trend difference index of each period, and the specific calculation method is as follows:

[0065] In the formula, D is the supply-demand matching coefficient of each period; Y is the power supply consistency coefficient of each period, and 1 is added to the power supply consistency coefficient to avoid affecting the positivity of the supply-demand matching coefficient when the power supply consistency coefficient is negative, thereby affecting the correlation analysis between the data; G is the peak value difference index of each period; H is the trend difference index of each period, is a preset value greater than 0, and the purpose is to avoid a denominator of 0. In this embodiment, the implementer can set it according to the actual situation, which is not limited in this embodiment.

[0066] It should be understood that the supply-demand matching coefficient can comprehensively reflect the consistency of the change trend between the power supply sequence and the load sequence. The larger the supply-demand matching coefficient is, the better the matching degree between the photovoltaic power generation and the grid load is, the photovoltaic power can better adapt to the grid load demand, and the grid operation is more stable.

[0067] In step S005, the voltage mean value and the impedance mean value of all lines at all times in each cycle are obtained, the difference change of the voltage mean value and the impedance mean value of each line is analyzed, and the dispersion degree of the voltage mean value and the impedance mean value is analyzed, so as to determine the impedance fluctuation coefficient of each cycle.

[0068] In addition, the grid line impedance also affects the distribution of reactive power. When the impedance difference of different lines in the grid is large, the path of reactive power flow changes, resulting in excessively high voltage in some areas and excessively low voltage in other areas, causing imbalance. In order to ensure the balance of grid load, the embodiment adjusts the output of reactive power through droop control. If the line impedance difference is large, a larger droop coefficient is needed to compensate for the voltage difference, so as to ensure more balanced distribution of reactive power and thus maintain the stability of the grid. Therefore, the embodiment takes the line impedance difference and the voltage difference between each line as a control adjustment factor of the droop coefficient in the droop control algorithm.

[0069] For each cycle, the embodiment takes the ratio of the average voltage to the average impedance of each line in the grid as the impedance-voltage ratio of each line, denoted as the second difference. The dispersion degree of the impedance-voltage ratio of all lines in the grid is calculated, denoted as the impedance influence difference index of the grid. The impedance influence difference index can reflect whether the degree of influence of impedance change on voltage in the line is consistent. The smaller the impedance influence difference index is, the more consistent the influence of impedance on voltage in each line is, and the greater the influence of impedance change on voltage is, so a larger droop coefficient is needed to adjust when the line impedance difference occurs.

[0070] It should be noted that the calculation method of the dispersion degree in the embodiment is calculated by using variance. The implementer can select other feasible dispersion degree calculation methods, such as standard deviation, coefficient of variation, etc.

[0071] The variance of the average impedance of all lines in the grid in each cycle is calculated, denoted as the impedance dispersion index of the line. The variance of the average voltage of all lines in the grid in each cycle is calculated, denoted as the voltage dispersion index of the line. The absolute value of the difference between the impedance dispersion index and the voltage dispersion index is denoted as the third difference. The third difference can reflect whether the fluctuation degree of the impedance and the voltage of the line in the grid is consistent. The smaller the third difference is, the more consistent the fluctuation is, and thus the greater the influence of the impedance on the voltage is.

[0072] The impedance fluctuation coefficient of each cycle is determined according to the impedance influence difference index, the impedance dispersion index, the voltage dispersion index, and the third difference of each cycle, and the specific calculation method is as follows:

[0073] wherein, is the impedance fluctuation coefficient of each cycle in the power grid; is the impedance influence difference index of each cycle; is the impedance dispersion index of the line of each cycle; and K is the third difference of each cycle, is a preset positive number, which is used to avoid a denominator of 0, and in the embodiment The implementer can set it according to the actual situation, and the embodiment does not limit it here.

[0074] It should be understood that the impedance fluctuation coefficient can reflect the influence degree of the line impedance on the voltage in the line and the difference degree between the impedances. If the impedance fluctuation coefficient is larger, it means that the influence of the impedance on the voltage is larger, and the difference between the line impedances is larger, and the possibility of instability of the power grid is higher. At this time, a larger droop coefficient is needed to regulate the power of the power grid.

[0075] In step S006, the droop coefficient in the droop control algorithm is optimized according to the grid-connection influence coefficient, the supply-demand matching coefficient, and the impedance fluctuation coefficient, and the load imbalance of the power system is compensated by using the optimized droop control algorithm.

[0076] According to the grid-connection influence coefficient of each cycle of photovoltaic power, the supply-demand matching coefficient of each cycle, and the impedance fluctuation coefficient of each cycle, a droop stability factor of each cycle is constructed, and the specific calculation method is as follows:

[0077] wherein, L is the droop stability factor of each cycle, A is the grid-connection influence coefficient of each cycle of photovoltaic power, D is the supply-demand matching coefficient of each cycle, is the impedance fluctuation coefficient of each cycle in the power grid, is a preset positive number, which is used to avoid a denominator of 0, and in the embodiment The implementer can set it according to the actual situation, and the embodiment does not limit it here.

[0078] It should be understood that the droop stability factor can reflect whether the influence of the photovoltaic power on the power system is stable in each cycle. If the grid connection influence coefficient A and the impedance fluctuation coefficient F are larger, it indicates that the difference between the line impedances in the power grid is larger, and the adverse influence of the photovoltaic power on the power grid is larger. If the supply-demand matching coefficient D is smaller, it indicates that the matching degree between the photovoltaic power supply and the demand load of the power grid is lower. Therefore, if the droop stability factor is smaller, a larger droop coefficient is needed to adjust the load of the power grid and balance the loads between phases in time to compensate the load of the power system.

[0079] Further, in the droop control algorithm of the reactive power, the target is to adjust the output reactive power to compensate the voltage fluctuation of the power grid. The droop coefficient in the droop control algorithm is adjusted in real time in the embodiment, and specifically: , wherein R is the droop coefficient in the droop control algorithm in the current cycle, U is the droop coefficient in the droop control algorithm in the previous cycle of the current cycle, L is the droop stability factor in the current cycle, norm() is a normalization function, and r is a constant greater than 0 and less than 1, which is used to avoid the problem that the droop coefficient is 0 or too small, causing the power grid to be unable to respond to the load change in time and causing uneven load distribution. In the embodiment, r = 0.5, which can be set by the implementer according to the actual situation, and the embodiment does not limit this. The droop coefficient adjustment process in the droop control algorithm is shown in the flowchart of FIG. Figure 2

[0080] It should be noted that the initial droop coefficient is set to 0.4 in the embodiment, which can be set by the implementer; the droop coefficient in the droop control algorithm is adjusted in real time, the output of the corresponding reactive power is controlled by using the adjusted droop control algorithm, the voltage fluctuation is compensated, the load imbalance of the power system is compensated by adjusting the voltage, and the power grid is more stable. The droop control algorithm is a known technology, and the embodiment does not make a detailed description.

[0081] It should be noted that the above sequence of the embodiments is only for description, and does not represent the advantages and disadvantages of the embodiments. The specific embodiments of the present application are described above. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0082] Each embodiment in the specification is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment mainly describes the differences from other embodiments.

[0083] ​The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; the technical solutions recorded in the foregoing embodiments are modified, or some technical features are replaced equivalently, without making the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and all should be included in the protection scope of the present application.

Claims

1. A method for load imbalance compensation in a grid-connected power system with photovoltaic power, characterized by, The method comprises the following steps: Collecting photovoltaic power generation power, voltage data and three-phase current data of photovoltaic power generation at the grid access end in each period, collecting power data of each load end at each time in the grid, and collecting voltage and impedance of each line at each time in the grid; Analyzing the similarity degree of the voltage data distribution of photovoltaic power generation at the grid access end in each period and a sinusoidal signal to obtain a sinusoidal similarity coefficient of each period; the determination of the sinusoidal similarity coefficient comprises: forming a voltage sequence by using the voltage data of photovoltaic power generation at the grid access end at all times in each period, obtaining frequency domain information of the voltage sequence by using a frequency domain transformation algorithm, constructing a standard sinusoidal signal of the voltage sequence, and the sinusoidal similarity coefficient is a correlation coefficient between the voltage sequence and the standard sinusoidal signal; based on the fluctuation difference between any two phases in the three-phase current data of each period, a waveform difference index of the three-phase current of each period is determined; the determination process of the waveform difference index is: forming A-phase sequence, B-phase sequence and C-phase sequence by using A-phase current, B-phase current and C-phase current at all times in each period, respectively, calculating the metric distance between any two sequences in the A-phase sequence, the B-phase sequence and the C-phase sequence, and the waveform difference index is the average of all the metric distances in each period; The similarity of A-phase current and B-phase current under different time delays is recorded as a first similarity, the similarity of B-phase current and C-phase current under different time delays is recorded as a second similarity, the difference between the first similarity and the second similarity is analyzed and recorded as a first difference, and the grid access influence coefficient of photovoltaic power of each period is determined in combination with the sinusoidal similarity coefficient and the waveform difference index; the determination of the grid access influence coefficient comprises: the grid access influence coefficient of each period is positively correlated with the waveform difference index and the first difference, and negatively correlated with the sinusoidal similarity coefficient; The power accumulation of all load ends at each time is recorded as load power, the correlation between the power generation power and the load power of each period is analyzed to determine a power supply consistency coefficient of each period; the power supply consistency coefficient is the correlation coefficient of the power generation power and the load power at all times in each period; The power generation power and the load power at all times in each period are respectively formed into a power supply sequence and a load sequence, all peak values in the power supply sequence and the load sequence are obtained, the average of the difference between the peak values of the same number in the power supply sequence and the load sequence is taken as a peak value difference index of each period, the difference between the slopes of the fitting straight lines of the power supply sequence and the load sequence of each period is taken as a trend difference index of each period, and the supply-demand matching coefficient of each period is determined in combination with the power supply consistency coefficient; the determination of the supply-demand matching coefficient comprises: the supply-demand matching coefficient of each period is positively correlated with the power supply consistency coefficient, and negatively correlated with the peak value difference index and the trend difference index. The difference between the voltage average and the impedance average of each line in each cycle is recorded as a second difference, and the second difference is the ratio of the voltage average to the impedance average of each line; the difference between the dispersion degree of the voltage average and the dispersion degree of the impedance average of all lines is recorded as a third difference, and the impedance fluctuation coefficient of each cycle is determined in combination with the dispersion degree of the impedance average; the determination of the impedance fluctuation coefficient includes that the impedance fluctuation coefficient is positively correlated with the dispersion degree of the impedance average of all lines, and is negatively correlated with the third difference and the dispersion degree of the second difference of all lines; The product of the grid access influence coefficient and the impedance fluctuation coefficient of each cycle is calculated, the sum of the product and a preset value greater than 0 is calculated, and the ratio of the supply-demand matching coefficient of each cycle to the sum value is taken as the droop stability factor of each cycle; the droop coefficient in the droop control algorithm is adjusted based on the droop stability factor of each cycle, and the optimized droop control algorithm is used to compensate for the load imbalance of the power system; the adjustment of the droop coefficient in the droop control algorithm based on the droop stability factor of each cycle includes: calculating the addition result of the normalized value of the inverse of the droop stability factor of each cycle and a preset constant greater than 0 and less than 1, and the droop coefficient in the droop control algorithm of each cycle is the product of the addition result of each cycle and the droop coefficient of the previous cycle of each cycle.

Images