Photovoltaic global maximum power tracking method based on hyperbolic model

Abstract

The invention discloses a photovoltaic global maximum power tracking method based on a hyperbolic model. The method comprises the following steps: establishing a fitting model of a photovoltaic output curve through real-time sampling data; and dynamically combining an online fitting model with a perturbation and observation method by adopting a self-adaptive weight coefficient. Compared with an existing photovoltaic maximum power tracking method, the method has the advantages that the tracking speed is higher, the search range and the voltage track in the search process can be effectively reduced, the method is suitable for the local shading working condition with the environment changing frequently, and the overall performance of a photovoltaic system is improved.

Description

Technical Field

[0001] The present invention belongs to the field of control technology in power electronics technology and specifically relates to a photovoltaic global maximum power tracking method based on a hyperbolic model. Background Art

[0002] In photovoltaic power generation systems, photovoltaic cells convert solar energy into clean, pollution-free electricity. As light and temperature conditions change, the output of photovoltaic cells exhibits nonlinear and fluctuating characteristics. In practical applications, multiple photovoltaic modules are usually connected in series to form a photovoltaic array with a higher voltage level. Under partial shading conditions, the output of the photovoltaic array has multiple local power peaks. Therefore, studying the photovoltaic maximum power point tracking (MPPT) method under multiple peaks is a key issue in achieving efficient conversion of photovoltaic power.

[0003] Classic MPPT algorithms include the perturbation observation method and the incremental conductance method, which can only converge to a local peak point and are therefore not suitable for partial shading conditions. Some global MPPT algorithms based on meta-inspired optimization algorithms search for the optimal power value by distributing search particles throughout the search space [1]. Such algorithms have the advantage of not requiring advance knowledge of photovoltaic parameter information, but their response speed and convergence performance depend on parameter adjustment and require trial and error. In addition, each particle is evaluated during each iteration, which results in a long tracking time for the photovoltaic maximum power point. Scanning maximum power tracking algorithms can quickly converge to the power peak point, but rapid charging and discharging can lead to component damage and aging problems [2]. On the other hand, the output power needs to be continuously sampled at high frequencies during the scanning process. Some algorithms improve the classic MPPT algorithm, narrow the operating point search range, and thus improve the convergence speed of power tracking [3], [4]. In addition, the composite MPPT method that combines artificial intelligence and mathematical models can present better comprehensive performance [5], [6]. However, the existing MPPT algorithms still have shortcomings in terms of search speed, convergence performance, sensor requirements, and controller computation.

[0004] Reference Patents:

[0005] [1] Xia Xiangzi, Zhang Chun, Zhou Yuan. Maximum power point tracking method for photovoltaic array output power under partial shading conditions [P]. Anhui Province: CN202410912229.6, 2024-10-11.

[0006] [2] Ye Jianying, Wan Wei, Shu Yizhan, et al. A maximum power tracking device and method for photovoltaic power generation system [P]. Fujian Province: CN202311808939.6, 2024-03-15.

[0007] [3]Xiong Yuansheng, Liu Chunyuan, Zhang Huanle, et al. A method for tracking the maximum power point of multi-peak photovoltaic module strings[P]. Zhejiang Province: CN202311648410.2, 2024-02-06.

[0008] [4] Yun Ping. A global maximum power point tracking method, power optimizer and photovoltaic inverter [P]. Anhui Province: CN202111386502.9, 2024-04-12.

[0009] [5] Yin Haoran, Song Bo, Huang Bo. A photovoltaic power generation system and method for improving maximum power point tracking efficiency [P]. Shaanxi Province: CN202410884672.7, 2024-09-20.

[0010] [6] Xiao Yiping, Shen Zongtao, Shu Jun, et al. A photovoltaic array multi-peak maximum power point tracking control method and system [P]. Hubei Province: CN202311450529.9, 2024-05-10. Summary of the invention

[0011] In view of the above-mentioned deficiencies in the prior art, the present invention provides a photovoltaic global maximum power tracking method based on a hyperbola model, which can achieve rapid power tracking, narrow the voltage trajectory and range during the search process, is suitable for complex working conditions with frequent environmental changes, and improves the overall performance of the photovoltaic system.

[0012] In order to achieve the above object, the technical solution adopted by the present invention is:

[0013] A photovoltaic global maximum power tracking method based on a hyperbolic model comprises the following steps:

[0014] S1, determine whether the change of photovoltaic output power exceeds the power threshold. If not, continue to apply the perturbation observation method to maintain steady-state operation. If yes, start global maximum power point tracking and go to S2;

[0015] S2. Determine the starting search direction according to the starting photovoltaic voltage and the voltage threshold. If the starting photovoltaic voltage is less than the threshold, first search in the direction of decreasing voltage, then return to the starting point for an ascending search; if the starting photovoltaic voltage is greater than the threshold, first search in the direction of increasing voltage, then return to the starting point for a descending search;

[0016] S3, after both the ascending and descending searches are completed, the steady-state operation is maintained at the recorded global maximum power point;

[0017] In S2, during the upward or downward search process, if it is detected that the photovoltaic voltage is less than the voltage boundary threshold, or the photovoltaic current is less than the current boundary threshold, the search in the current direction is completed. Voltage boundary threshold V minand the current boundary threshold I min Determined by the following formula:

[0018]

[0019] Among them, P GMPP is the maximum power recorded during the search process, I GMPP,STC,array is the maximum power point current of the photovoltaic array under standard lighting conditions, V oc,STC,array It is the open circuit voltage of the photovoltaic array under standard lighting conditions.

[0020] In S2, the descending search algorithm cyclically executes the local maximum power point tracking and short-circuit current estimation algorithms, and the ascending search algorithm cyclically executes the region segmentation point voltage estimation and the local maximum power point current estimation algorithms.

[0021] The algorithm flow of the local maximum power point tracking includes the following steps:

[0022] The photovoltaic output hyperbolic model parameters are calculated based on the real-time sampling data. When the voltage and current data in the sampling window are (V1, I1), (V2, I2) and (V3, I3), the photovoltaic output resistance R1, R2 and R3 corresponding to the sampling data are calculated and determined by the following formula:

[0023]

[0024] The model parameters k1, k2, k3 are calculated based on the photovoltaic sampling data and are determined by the following formula:

[0025]

[0026] Thus, a photovoltaic output hyperbolic model is established, and the relationship between the photovoltaic output equivalent resistance and voltage is described as follows:

[0027]

[0028] Next, estimate the local maximum power point voltage V LMPP,est , determined by the following formula:

[0029]

[0030] Then the reference voltage V of the local maximum power tracking algorithm is ref_LMPPT Determined by the following formula:

[0031] V ref_LMPPT =(1-α)V po +αV LMPP,est

[0032] Where V poThe classical perturbation observation method is used to obtain the reference value of the PV voltage after disturbance, and α is the weight coefficient of the dynamic combination of disturbance observation and model estimation.

[0033] The weight coefficient α is adjusted according to the change of photovoltaic output power and is determined by the following formula:

[0034]

[0035] Where T po is the perturbation period in the perturbation-observation algorithm, T α is the adjustment period of the weight coefficient, α prev is the weight coefficient of the previous regulation cycle, P is the photovoltaic output power, P prev It is the output power history value of the previous regulation cycle.

[0036] In the short-circuit current estimation algorithm, the reference current I ref_sc Determined by the following formula:

[0037]

[0038] Among them, I ref is the current reference value, I mpp is the local maximum power point current in the current area, I sc,STC is the short-circuit current of the photovoltaic module under standard illumination conditions, I MPP,STC is the photovoltaic maximum power point current under standard lighting conditions.

[0039] In the algorithm for estimating the voltage at the region segmentation point, the reference voltage V ref_SDP Determined by the following formula:

[0040]

[0041] In the algorithm for estimating the local maximum power point current, the reference current I ref_MPP Determined by the following formula:

[0042]

[0043] Among them, I sc It is the short-circuit current of the photovoltaic module corresponding to the current search area.

[0044] Compared with the prior art, the photovoltaic global maximum power tracking method based on the hyperbolic model disclosed in this application has the following beneficial effects:

[0045] 1. The present invention can achieve rapid tracking of the global maximum power point under partial shading conditions and reduce the energy loss of photovoltaic power generation by reducing the search space and voltage trajectory in the photovoltaic maximum power tracking process;

[0046] 2. The present invention can realize stable operation of the photovoltaic system at the global maximum power point, reduce steady-state power oscillation, realize efficient conversion of photovoltaic power generation, and improve the stability of energy output of the photovoltaic system;

[0047] 3. The present invention can reduce the energy disturbance of the photovoltaic system to the power grid by improving the photovoltaic maximum power tracking speed and suppressing the steady-state waveform of photovoltaic power, so that the power system can resume normal operation more quickly when the operating conditions change, realize the reliable support of the photovoltaic unit for the system load, and improve the stability of the power grid and the power quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments and technical solutions of the present invention, the drawings required for use are briefly introduced below.

[0049] Figure 1 The present invention is a flowchart of the steps of the photovoltaic global maximum power tracking method based on the hyperbolic model.

[0050] Figure 2 The present invention is an algorithm flow chart of the photovoltaic global maximum power tracking method based on the hyperbolic model.

[0051] Figure 3 Schematic diagram of the photovoltaic power generation system.

[0052] Figure 4 Schematic diagram of the global search process under partial shading conditions.

[0053] Figure 5 PV characteristic curve of photovoltaic output under different local shading conditions.

[0054] Figure 6 This is the experimental waveform when the photovoltaic operating point starts from the open circuit voltage under PSC1.

[0055] Figure 7 It is the experimental waveform when switching from PSC1 to PSC2 under partial shading conditions.

[0056] Figure 8 It is the experimental waveform when switching from PSC2 to PSC3 under partial shading conditions.

[0057] Fig. 9 It is the experimental waveform when switching from PSC3 to PSC4 under partial shading conditions. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention are further described below in conjunction with the accompanying drawings and tables. It should be understood that the description of the embodiments of the present invention herein is not intended to limit the protection scope of the present invention.

[0059] The present invention provides a photovoltaic global maximum power tracking method based on a hyperbolic model. Figure 1 The present invention is a flowchart of a photovoltaic global maximum power tracking method based on a hyperbolic model, comprising the following steps:

[0060] S1, determine whether the change of photovoltaic output power exceeds the power threshold. If not, continue to apply the perturbation observation method to maintain steady-state operation. If yes, start global maximum power point tracking and go to S2;

[0061] S2. Determine the starting search direction according to the starting photovoltaic voltage and the voltage threshold. If the starting photovoltaic voltage is less than the threshold, first search in the direction of decreasing voltage, then return to the starting point for an ascending search; if the starting photovoltaic voltage is greater than the threshold, first search in the direction of increasing voltage, then return to the starting point for a descending search;

[0062] S3. After both the ascending and descending searches are completed, the steady-state operation is maintained at the recorded global maximum power point.

[0063] Figure 2 The present invention is an algorithm flow chart of the photovoltaic global maximum power tracking method based on the hyperbolic model. Figure 3 The schematic diagram of the photovoltaic power generation system includes a photovoltaic array, a DC conversion circuit, a DC bus, a load, and a controller. The photovoltaic array converts solar energy into electrical energy, and the DC conversion circuit, under the control of the drive signal generated by the controller, converts the photovoltaic power to the output side to supply energy to the load on the DC bus.

[0064] Figure 4 The global search process diagram under partial shading conditions is shown in Figure 1. Point A in the figure is the starting photovoltaic working point. In this embodiment, the starting voltage threshold is designed to be 0.5V. oc,array , where V oc,array is the open circuit voltage of the photovoltaic array under standard illumination. In addition, at point A, it is detected that the difference between the current photovoltaic output power and the photovoltaic power of the previous sampling period is greater than the power threshold.

[0065] The global maximum power tracking is started from point A by S1.

[0066] According to S2, the starting photovoltaic voltage is less than the threshold, so a downward search is first performed in the direction of voltage reduction, and then an upward search is performed back to the starting point.

[0067] During the search process, the photovoltaic voltage and current are sampled in real time to calculate the photovoltaic output power P GMPP , and thus calculate the voltage boundary threshold V min and the current boundary threshold I min , determined by the following formula:

[0068]

[0069] The short-circuit current is estimated from point A. Substitute the sampled current at point A as the local maximum power point current into equation (2) to calculate the short-circuit current estimate, which is determined by the following equation:

[0070]

[0071] This estimated value is used as the reference current and tracked to point B through current closed-loop control.

[0072] Next, start local maximum power tracking from point B. Based on the disturbance observation algorithm, the photovoltaic output voltage and current are continuously sampled, and the sampling data (V1, I1), (V2, I2), (V3, I3) are continuously updated by moving the sampling window. Substitute the sampling data into formula (4) and calculate the photovoltaic output resistance R1, R2, R3 corresponding to the sampling data, which is determined by the following formula:

[0073]

[0074] The model parameters k1, k2, k3 are calculated based on the photovoltaic sampling data and are determined by the following formula:

[0075]

[0076] Thus, a photovoltaic output hyperbolic model is established, and the relationship between the photovoltaic output equivalent resistance and voltage is described as follows:

[0077]

[0078] Next, estimate the local maximum power point voltage V LMPP,est , determined by the following formula:

[0079]

[0080] The estimated local maximum power point voltage V LMPP,est The voltage reference value V obtained by the perturbation observation method po Substitute into equation (7) as the reference value for voltage closed-loop control, the reference voltage V ref_LMPPT Determined by the following formula:

[0081] V ref_LMPPT =(1-α)V po +αV LMPP,est (7)

[0082] In the above formula, α is the weight coefficient of online regulation. In the local maximum power tracking algorithm, in each regulation cycle, according to the change of photovoltaic output power, α is determined by the following formula.

[0083]

[0084] Where T po is the perturbation period in the perturbation-observation algorithm, T α is the adjustment period of the weight coefficient, α prev is the weight coefficient of the previous regulation cycle, P is the photovoltaic output power, P prev It is the output power history value of the previous regulation cycle.

[0085] In the local maximum power tracking algorithm, the voltage reference value is updated in each voltage disturbance cycle, so that the operating point gradually approaches the local maximum power point in the current search area. In the process of tracking from point B to the local maximum power point, the recorded power peak becomes larger, so the voltage boundary threshold V calculated by formula (1) min When the photovoltaic working point is searched to point B', it is detected that the photovoltaic voltage is less than the voltage boundary threshold V min , then the descending search process ends. The photovoltaic working point returns to the starting working point of point A, also recorded as point C.

[0086] According to S2, the photovoltaic working point starts from point C and searches upward in the direction of voltage increase. min Substituting into the photovoltaic output hyperbolic model, the estimated value of the voltage at the regional segmentation point is solved by equation (9) as the reference voltage, V ref_SDP Determined by the following formula.

[0087]

[0088] Under voltage closed-loop control, the photovoltaic operating point reaches point D near the regional division point. Next, the output current at point D is sampled as the photovoltaic module short-circuit current corresponding to the search area, and is substituted into equation (10) to solve the local maximum power point current as the current reference value, I ref_MPP Determined by the following formula:

[0089]

[0090] The photovoltaic operating point is tracked to the local maximum power point at point E through current closed-loop control.

[0091] Since the current at point E is still greater than the current boundary threshold I min Therefore, the regional segmentation point estimation algorithm process is repeated. The photovoltaic output hyperbolic model and current boundary threshold I established in this area are calculated by formula (9). min The intersection voltage is , then the photovoltaic operating point tracks to point F.

[0092] The photovoltaic operating point samples the current at point F, and then substitutes it into equation (10) to estimate the current at the local maximum power point F' on the right as the current reference value. However, since it is detected that the calculated current reference value is less than the current boundary threshold, the upward search process ends. The photovoltaic operating point stops moving in the direction of increasing voltage at point F, but returns to the power peak point recorded during the global search process, that is, point E.

[0093] After the photovoltaic working point returns to point E, the local maximum power point tracking algorithm is performed to make the photovoltaic system operate steadily at the global maximum power point. According to S1, the fluctuation of the current photovoltaic output power relative to the power value recorded at the start of steady-state operation is detected in each voltage disturbance cycle. If the power fluctuation is greater than the preset threshold, the algorithm process of the photovoltaic global maximum power tracking method based on the hyperbolic model is restarted; if the power fluctuation is less than the preset threshold, the steady-state operation is continued.

[0094] In order to more clearly present the implementation effect of the photovoltaic global maximum power tracking method based on the hyperbolic model provided by the present invention, an experimental study was conducted on the photovoltaic maximum power tracking effect under different shading conditions. In the experiment, four shading conditions, PSC1-PSC4, were used. Figure 5 PV output characteristic curve under different local shading conditions.

[0095] Figure 6 This is the experimental waveform when the photovoltaic operating point starts from the open circuit voltage under PSC1. Since the starting operating point is at the open circuit voltage, only the descending search algorithm is executed. The photovoltaic operating point searches from the open circuit voltage of 120V in the direction of voltage reduction, and the global search is completed when the voltage drops to 50V. The photovoltaic operating point tracking converges to the global maximum power point at 53V, the output power is 195W, and the search time is about 0.65s. During the maximum power tracking process, the weight coefficient α is automatically adjusted in the range of 0 to 1, realizing the dynamic combination of the photovoltaic output hyperbolic model and the perturbation observation method, thereby greatly improving the tracking speed. After the global search is completed, the photovoltaic operating point stabilizes at the global maximum power point. It can be seen from the figure that after entering the steady state, the value of the weight coefficient α finally stabilizes to 1, which effectively suppresses the power oscillation caused by the traditional perturbation observation method and achieves stable and efficient photovoltaic power output.

[0096] Figure 7This is the experimental waveform when switching from PSC1 to PSC2 under partial shading conditions. Under this condition, the global maximum power point switches from the second local power peak point to the third local power peak point. Before the environmental condition is switched, the photovoltaic operating point is stable at the global maximum power operating point of PSC1. After the environmental condition is switched, the controller detects that the photovoltaic output power fluctuation exceeds the threshold, and then starts the global maximum power tracking algorithm. As can be seen from the figure, the photovoltaic operating point first searches downward, then searches upward, and finally converges to the global maximum power point. The search time is 0.5s, the steady-state tracking efficiency is 99.62%, the voltage trajectory during the search process is 42V, the voltage fluctuation range is 26.9V, and the energy loss is 21.28J.

[0097] Figure 8 This is the experimental waveform when switching from PSC2 to PSC3 under partial shading conditions. Under this condition, the global maximum power point switches from the third local power peak point to the fourth local power peak point. The controller first performs an upward search algorithm, then a downward search algorithm, and converges to the global maximum power point after a response time of 0.63s. During steady-state operation, the power tracking accuracy is as high as 99.67%. The voltage trajectory during the search process is 136.1V, the voltage fluctuation amplitude is 51.46V, and the energy loss is 23.63J.

[0098] Fig. 9 This is the experimental waveform when switching from PSC3 to PSC4 under partial shading conditions. Under this condition, the global maximum power point switches from the fourth local power peak point to the first local power peak point. As can be seen from the figure, after the downward search is completed, the photovoltaic working point quickly reaches the search boundary threshold, thereby converging to the global maximum power point. The search process takes 0.75s, and the tracking efficiency in the steady state is 99.72%. The voltage trajectory and voltage fluctuation amplitude during the search process are 51.9V and 50.6V, respectively, and the energy loss is 23.5J.

[0099] Through the above embodiments, it can be verified that the photovoltaic global maximum power tracking method based on the hyperbolic model provided by the present invention is feasible and effective, can realize fast global maximum power tracking under partial shading conditions, and effectively shorten the photovoltaic voltage trajectory and fluctuation amplitude during the search process. Therefore, the photovoltaic global maximum power tracking method based on the hyperbolic model can alleviate the photovoltaic energy waste caused by frequent changes in the lighting environment in actual application scenarios, and the problem that the photovoltaic output power disturbance affects the system stability and power quality, thereby further improving the comprehensive performance of the photovoltaic power generation system.

[0100] Obviously, the above embodiments are only examples for clear explanation, and are not intended to limit the implementation methods. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A photovoltaic global maximum power tracking method based on a hyperbolic model, characterized in that: The following steps are involved: S1, determine whether the change of photovoltaic output power exceeds the power threshold. If not, continue to apply the perturbation observation method to maintain steady-state operation. If yes, start global maximum power point tracking and go to S2; S2. Determine the starting search direction according to the starting photovoltaic voltage and the voltage threshold. If the starting photovoltaic voltage is less than the threshold, first search in the direction of decreasing voltage, then return to the starting point for an ascending search; if the starting photovoltaic voltage is greater than the threshold, first search in the direction of increasing voltage, then return to the starting point for a descending search; S3. After both the ascending and descending searches are completed, the steady-state operation is maintained at the recorded global maximum power point.

2. The photovoltaic global maximum power tracking method based on a hyperbolic model according to claim 1, characterized in that: In S2, during the upward or downward search process, if it is detected that the photovoltaic voltage is less than the voltage boundary threshold, or the photovoltaic current is less than the current boundary threshold, the search in the current direction is completed. Voltage boundary threshold V min and the current boundary threshold I min Determined by the following formula: Among them, P GMPP is the maximum power recorded during the search process, I GMPP,STC,array is the maximum power point current of the photovoltaic array under standard lighting conditions, V oc,STC,array It is the open circuit voltage of the photovoltaic array under standard lighting conditions.

3. The photovoltaic global maximum power tracking method based on a hyperbolic model according to claim 1, characterized in that: In S2, the descending search algorithm cyclically executes the local maximum power point tracking and short-circuit current estimation algorithms, and the ascending search algorithm cyclically executes the region segmentation point voltage estimation and the local maximum power point current estimation algorithms.

4. The photovoltaic global maximum power tracking method based on a hyperbolic model according to claim 3, characterized in that: The algorithm flow of the local maximum power point tracking includes the following steps: The photovoltaic output hyperbolic model parameters are calculated based on the real-time sampling data. When the voltage and current data in the sampling window are (V1, I1), (V2, I2) and (V3, I3), the photovoltaic output resistance R1, R2 and R3 corresponding to the sampling data are calculated and determined by the following formula: The model parameters k1, k2, k3 are calculated based on the photovoltaic sampling data and are determined by the following formula: Thus, a photovoltaic output hyperbolic model is established, and the relationship between the photovoltaic output equivalent resistance and voltage is described as follows: Next, estimate the local maximum power point voltage V LMPP,est , calculated by the following formula; Then the reference voltage V of the local maximum power tracking algorithm is ref_LMPPT Determined by the following formula: V ref_LMPPT =(1-α)V po +αV LMPP,est Where V po The classical perturbation observation method is used to obtain the reference value of the PV voltage after disturbance, and α is the weight coefficient of the dynamic combination of disturbance observation and model estimation.

5. The photovoltaic global maximum power tracking method based on a hyperbolic model according to claim 3, characterized in that: In the short-circuit current estimation algorithm, the reference current I ref_sc Determined by the following formula: Among them, I mpp is the local maximum power point current in the current area, I sc,STC is the short-circuit current of the photovoltaic module under standard illumination conditions, I MPP,STC is the photovoltaic maximum power point current under standard lighting conditions.

6. The photovoltaic global maximum power tracking method based on a hyperbolic model according to claim 3, characterized in that: In the algorithm for estimating the voltage at the region segmentation point, the reference voltage V ref_SDP Determined by the following formula: Among them, I min is the current boundary threshold calculated from the sampled data.

7. The photovoltaic global maximum power tracking method based on a hyperbolic model according to claim 3, characterized in that: In the algorithm for estimating the local maximum power point current, the reference current I ref_MPP Determined by the following formula: Among them, I sc It is the short-circuit current of the photovoltaic module corresponding to the current search area.

8. The photovoltaic global maximum power tracking method based on a hyperbolic model according to claim 4, characterized in that: The weight coefficient α is adjusted according to the change of photovoltaic output power and is determined by the following formula: Where T po is the perturbation period in the perturbation-observation algorithm, T α is the adjustment period of the weight coefficient, α prev is the weight coefficient of the previous regulation cycle, P is the photovoltaic output power, P prev It is the output power history value of the previous regulation cycle.

Images