CN103955253B - Based on the photovoltaic array multimodal value maximum power point tracing method of power closed loop scanning - Google Patents

Abstract

The invention discloses a photovoltaic array multi-peak maximum power point tracking (MPPT—Maximum Power Point Tracking) method based on power closed-loop scanning. It consists of three stages. The first stage is based on the power closed-loop control to realize the global scanning of the multi-peak curve, and complete the positioning of the maximum power point; the second stage is based on the particle swarm optimization algorithm to realize the three-point cooperative variable step size disturbance observation process. , to realize the local search near the maximum power point; the third stage is the constant voltage steady-state tracking process, and at the same time, the process of the first or second stage is activated according to the judgment of the environmental change information. The present invention can realize the maximum power point tracking of the photovoltaic array in the case of single peak or multi-peak, and there is no misjudgment and oscillation problems of the disturbance observation method and the conductance increment method; at the same time, it can overcome the dynamic tracking function existing in the existing method The problem of more loss and slower dynamic process. The invention can effectively find the global maximum power point and realize fast, stable and accurate tracking of the maximum power point.

Description

Multi-peak Maximum Power Point Tracking Method for Photovoltaic Array Based on Power Closed-loop Sweep

technical field

The invention belongs to the photovoltaic power generation technology in the field of electrical engineering, and in particular relates to a method for maximum power point tracking control (MPPT—Maximum Power Point Tracking) of a photovoltaic array under uneven illumination conditions in a photovoltaic power generation system.

Background technique

As a renewable energy source, solar energy has the advantages of being widely distributed, sustainable, and non-polluting. Photovoltaic power generation technology is one of the basic ways to effectively utilize solar energy resources. At present, various photovoltaic power generation technologies, including photovoltaic grid-connected, have received strong support from governments of various countries.

Although photovoltaic power generation technology has good application prospects, as a relatively new technology, it still faces many problems to be solved, one of which is the shadow shading problem of photovoltaic arrays. The causes of shadow shading problems include inconsistencies in the output characteristics of photovoltaic cells or photovoltaic modules, different orientations and inclination angles of photovoltaic cells during engineering installation, partial shading of buildings, and dust coverage. When the photovoltaic array is shaded, the output characteristic curve of the photovoltaic array will show multi-peak characteristics. At this time, it is difficult to make the photovoltaic array work at the global maximum power point, and the operating efficiency of the photovoltaic power generation system will drop significantly; The resulting drop in efficiency of the photovoltaic array can be as high as 15%. The maximum power point tracking control of the photovoltaic array is the basic way to improve the power generation efficiency of the photovoltaic array. Under normal conditions without shadows, the output characteristic curve of the photovoltaic array is a single peak curve, and it is relatively easy to find the maximum power point of the system at this time.

Recently, the commonly used single-peak maximum power point tracking methods mainly include the perturbation and observation method, the conductance increment method, the constant voltage method, and the hysteresis comparison method. These methods are mainly suitable for the maximum power point tracking of photovoltaic systems under single-peak conditions. For the multi-peak situation caused by shadow shading, the existing single-peak maximum power point tracking methods are difficult to adapt. Therefore, it is of great engineering significance to design a maximum power point tracking method that can be applied to both single-peak and multi-peak situations.

At present, the tracking control method of the maximum power point under multi-peak conditions has become a hot research issue in photovoltaic power generation technology. There are not only academic papers that have made in-depth theoretical analysis, but also engineering methods for practical applications. For example, the invention patent application "Photovoltaic Array multi-peak maximum power point tracking method" (CN103123514A) and "PV array global maximum power point rapid optimization method under partial shadow" (CN103324239A). in,

Chinese Invention Patent Application Publication CN103123514A published on May 29, 2013 "PV Array Multi-Peak Maximum Power Point Tracking Method" is to initially select the global maximum power point (MPP) according to the open circuit voltage and short circuit current of the photovoltaic array when there is no shadow. ) where the mountain peak is located; and based on the ratio of the voltage at the maximum power point of the photovoltaic array to the number of photovoltaic modules in series when there is no shadow, select the maximum power tracking (MPPT) search step, and search for the MPPs on the left and right sides respectively until a certain MPP, whose corresponding output power value is greater than the MPP output power value on the left and right sides, that is, the MPP is considered to be the global MPP; finally, the photovoltaic array is maintained at the global MPP, and the change of the operating conditions is monitored in real time. If the operating conditions change, then Restart the multi-peak MPPT policy. However, this tracking method has the following disadvantages:

1) The implementation process needs to rely on a lot of information of the photovoltaic array: the open circuit voltage of the photovoltaic array when there is no shadow, the short-circuit current, the type of components in the photovoltaic array, the number of components connected in series, etc. For details, see step 1. This defect determines The environmental adaptability of this method is not strong;

2) The voltage closed-loop control is used to realize the scanning of the output power curve of the photovoltaic array, and the position of the maximum power point must be determined by point-by-point scanning, which leads to the slow dynamic scanning process;

3) The local search in this method adopts the "disturbance observation method" and "conductance increment method". These two methods have the problems of misjudgment of dynamic process and small-scale oscillation in steady state tracking process, which lead to the dynamic stability and steady-state accuracy is not high;

4) The basic idea of this method is as in the 2003 IEEE document "A Study on a Two Stage Maximum Power Point Tracking Control of a Photovoltaic System under Partially Shaded Insolation Conditions" ("Research on a Two-Stage Maximum Power Point Tracking Method for Photovoltaic Systems under Partially Shaded Conditions") "——Proceedings of the 2003 IEEE Energy Society Plenary Meeting), it will make a misjudgment in some cases, that is, the basis for "selecting the scope of the existence of the global MPP" is not universally applicable, and the result of the judgment is sometimes wrong of;

5) The detection of dynamic changes in the environment and the judgment conditions for the restart of the MPPT method are not given, that is, the algorithm is incomplete.

Chinese Invention Patent Application Publication CN103324239A published on September 25, 2013 "Fast Optimal Method for Photovoltaic Array Global Maximum Power Point under Partial Shading" adopts "modified fruit fly algorithm" to realize global search and "improved golden separation method" Complete a local search. Its main shortcomings include:

1) The global search strategy adopted - "modified fruit fly algorithm" is similar to "PSO algorithm". They are both evolutionary algorithms. For the extreme value search problem of the array P-V curve, its rapidity is not obvious. At the same time, the search performance of the "modified fruit fly algorithm" depends on the selection of the initial operating point, and there is a situation where it cannot converge to the global maximum power point;

2) The voltage closed-loop search is used, and the realization process must also adopt a point-by-point scanning process. During the search process, the power values of many operating points are much smaller than the power of the maximum power operating point, so the dynamic power loss is relatively large, and there are also operating points Multiple large fluctuations in voltage;

3) The detection of dynamic changes in the environment and the judgment conditions for the restart of the MPPT method are not given, and the algorithm is not complete.

Contents of the invention

The technical problem to be solved by the present invention is to provide a fast and stable track-to The global maximum power point under multi-peak conditions, in order to improve the power generation efficiency of photovoltaic arrays, a multi-peak maximum power point tracking method for photovoltaic arrays based on power closed-loop scanning.

In order to solve the technical problem of the present invention, the adopted technical solution is: the photovoltaic array multi-peak maximum power point tracking method based on power closed-loop scanning includes on-line detection of the output voltage and output current of the photovoltaic array, especially,

Step 1, power closed-loop control scanning: First, through online detection of the output voltage and output current of the photovoltaic array to obtain the real-time output power of the photovoltaic array output, and then use the power closed-loop control to make the real-time output power track the reference power, so as to obtain the maximum power point tracking circuit control signal;

Step 2. Determine whether the output voltage of the photovoltaic array is lower than the minimum operating voltage of the photovoltaic inverter? If less than, go to step 3, otherwise repeat step 1 and step 2;

Step 3, three-point cooperative variable step size local search: First, based on the maximum output power of the photovoltaic array and its corresponding output voltage value obtained in steps 1 and 2, set three initial operating points, and then use the voltage closed-loop control to make The photovoltaic array works to three operating points in turn, and at the same time, the output power of each operating point is obtained by using the output voltage and output current of the photovoltaic array detected online. At the end of the maximum power point tracking period, the global maximum power point information and each The maximum power point information experienced by each working point, when the three working points work in turn, use the obtained information of the three working points to jointly determine the working voltage of the next round of three working points;

Step 4, judge whether the three operating points are close enough according to the voltage values of the three operating points? If the following conditions: P _m ≠ 0, U _m - U _m1 < U _step /10, U _m - U _m2 < U _step /10, U _m - U _m3 < U _step /10 are satisfied at the same time, among them P _m is the power at the known maximum power point, U _m is the voltage at the known maximum power point, U _m1 , U _m2 , and U _m3 are the voltage values at the maximum power points experienced by the three operating points respectively, and U _step is the initial disturbance Step size, then turn to step 5 if it is close enough, otherwise repeat the above steps 3 and 4;

Step 5, constant voltage steady-state tracking: first use the voltage closed-loop control to stabilize the output voltage of the photovoltaic array at the maximum power point determined in steps 3 and 4, and then use the real-time detection of the output voltage and output current of the photovoltaic array to calculate the real-time Output power, after that, by comparing the real-time output power with the maximum output power obtained in steps 3 and 4, calculate the relative power variation and the cumulative relative power variation;

Step 6, based on the relative power variation, determine whether the environment has changed drastically? If yes, restart step 1;

Step 7, based on the cumulative relative power change, determine whether the environment has changed slowly? If so, restart step 3, otherwise go to step 5.

As a further improvement of the multi-peak maximum power point tracking method for photovoltaic arrays based on power closed-loop scanning:

The process of the power closed-loop control in step 1 is to update the global maximum power point information by comparing the real-time output power with the known global maximum power after each time the real-time output power is obtained, that is, if P=U _pv *I _pv ＞P _m , then U _m ＝U _pv , P _m ＝P,, where P is the real-time output power, U _pv is the real-time detected photovoltaic array output voltage, and I _pv is the real-time detected photovoltaic array output current , U _m is the voltage at the known maximum power point, and P _m is the power at the known maximum power point.

The reference power in the step 1 is determined by the formula P _r (k)=P _r (k-1)+P _step , where P _r (k) is the current reference power to be tracked, P _r (k- 1) is the reference power of the last tracking, and P _step is the step size of the reference power change.

The three initial working points in step 3 are determined according to U ₁ (k)=U _m , U ₂ (k)=U _m -U _step , U ₃ (k)=U _m +U _step , where U ₁ (k) is the voltage of the first working point, U ₂ (k) is the voltage of the second working point, U ₃ (k) is the voltage of the third working point, U _m is the voltage of the known maximum power point Voltage and U _step are the initial disturbance step size.

The working voltage of the next round of three working points in the step 3 is determined according to the following formula:

In the formula, ΔU ₁ (k), ΔU ₂ (k), ΔU ₃ (k) are the voltage disturbances in the current maximum power point tracking period of the three operating points, ΔU ₁ (k-1), ΔU ₂ (k-1 ), ΔU ₃ (k-1) is the voltage disturbance in the maximum power point tracking cycle before the three operating points, U _m1 , U _m2 , U _m3 are the maximum power point voltage values experienced by the three operating points respectively, U ₁ (k-1), U ₂ (k-1), U ₃ (k-1) are the voltages of the maximum power point tracking cycle before the three operating points, ω∈(0,1), c ₁ ∈(0, 1), c ₂ ∈ (0,1) is the adjustment parameter of the working point voltage step.

The relative power change and cumulative relative power change in step 5 are calculated according to the following formula: E _p =1-U _pv *I _pv /P _m , E=ΣE _p , where E _p is the current maximum power The relative power variation of the point tracking period, E is the cumulative relative power variation of multiple tracking periods.

When step 1 is restarted in step 6, the power reference value is set according to P _r (k)=(U _pv *I _pv )/2, and the maximum power point power P _m =0 is reset at the same time, where U _pv is the output voltage of the photovoltaic array detected in real time, and I _pv is the output current of the photovoltaic array detected in real time.

The step of judging that the environment in step 7 changes slowly is as follows:

(1) Judging whether the steady-state tracking cycle time recorded by the timer exceeds the specified time, if overtime, reset the timer and the accumulated relative power variation, otherwise continue timing;

(2) If within the specified time, the relative change in cumulative power E>0.05, it means that the environment has changed slightly, that is, there is a slight offset between the actual maximum power point and the stored maximum power point, then restart step 3 Do a local search.

The multi-peak maximum power point tracking method of a photovoltaic array based on power closed-loop scanning disclosed by the present invention quickly realizes the positioning of the global maximum power point under the multi-peak situation, and realizes the accurate tracking of the global maximum power point, and its beneficial effects are concrete Reflected in:

1. It does not rely on any known photovoltaic array information, and operates entirely based on online detection information.

2. Use the power closed-loop control to control the instability of the "valley point" area on the output power curve of the photovoltaic array, automatically skip the area where there is no maximum power point, and determine the position of the maximum power point without point-by-point scanning, and the dynamic process is fast , the power loss is small.

3. The scope of the global MPP search is not pre-selected, and the search process does not depend on the starting point, so there is no phenomenon that it cannot converge to the global maximum power point.

4. The PSO algorithm-based "three-point collaborative variable step-size search method" is adopted near the global maximum power point. By using the convergence characteristics of this method, the misjudgment problem existing in the dynamic process of the existing method is overcome, and the steady-state power is avoided. Oscillation, improved steady-state tracking accuracy. The actual test results show that the tracking accuracy based on the PSO algorithm is as high as 99.9%, while the steady-state accuracy of the traditional "perturbation and observation method" and "conductance incremental method" are both less than 98%.

5. In a dynamic situation, the search process can be restarted according to the degree of environmental changes, realizing a rapid response to environmental changes.

6. It must be pointed out that the power closed-loop control in step 1 of this method is locally stable on the output curve of the photovoltaic array. When the reference power is greater than the maximum power point of the output curve of the photovoltaic array, the output voltage of the photovoltaic array will slide. The lowest operating voltage to the photovoltaic inverter, thus realizing the scanning of the output curve of the photovoltaic array. Different from the prior art, other existing methods only update the maximum power point information once every maximum power point tracking period, while the power closed-loop control of this method updates once every sampling period. Even if other methods update the maximum power point information every sampling period, it has no practical significance. The fundamental reason is that the above-mentioned "closed-loop power control is locally stable on the output curve of the photovoltaic array".

Description of drawings

Fig. 1 is the general flowchart of the present invention.

Fig. 2 is a schematic circuit diagram for realizing the present invention.

Fig. 3 is a schematic diagram of the static multi-peak tracking process.

Fig. 4 is a schematic diagram of the restart process when the environment changes suddenly.

Fig. 5 is a specific implementation flow chart of the present invention.

Figure 6 is a flow chart of the sampling control program.

FIG. 7 is a flow chart of restarting the three-point cooperative search process-initialization program.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Apparently, the described embodiments are only part of the embodiments of the present invention. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative work all belong to the protection scope of this patent.

Embodiments of the present invention provide a photovoltaic array multi-peak maximum power point tracking method based on power closed-loop scanning to solve the problems of slow tracking process, high energy loss and failure to track the maximum power point existing in the prior art.

The hardware circuit of the present invention should include detection circuits for the output voltage, output current and DC bus voltage of the photovoltaic array. The photovoltaic grid-connected inverter adopts a two-stage structure, the front stage is a DC-DC conversion circuit, which is used to complete the maximum power point tracking; the latter stage is a DC-AC inverter circuit, which realizes the photovoltaic array injection through the stable control of the bus voltage Energy and inverter output energy balance. When the system is powered on, the initialization of the maximum power point tracking circuit and program is completed, and the initialization of the variables of the maximum power point tracking program is completed.

Figure 2 is a circuit scheme for implementing the present invention. The circuit scheme includes a photovoltaic array, a detection circuit for the output voltage U _pv of the photovoltaic array, a detection circuit for the output current I _pv of the photovoltaic array, and a BOOST circuit composed of an inductor L, a switch tube T, a diode D and a DC bus capacitor C _DC , and Grid inverter circuit, maximum power point tracking control circuit. The maximum power point tracking method disclosed in the present invention obtains the duty ratio signal for controlling the BOOST circuit through the maximum power point tracking operation according to the U _pv and I _pv information detected online. If the duty ratio signal for controlling the BOOST circuit increases, the I _pv Increase, U _pv decreases; On the contrary, if the duty cycle decreases, I _pv decreases, U _pv increases.

For the static multi-peak situation shown in Figure 3, when the hardware system on which the present invention relies is powered on, the BOOST circuit and the grid-connected inverter circuit have not yet worked, and the operating point of the photovoltaic array is located at point a in Figure 3, that is, at the open circuit voltage . At this time, it is set that the sampling control period of the system is T _S =50 uS, and the maximum power point tracking period T _mppt =1 second. Each sampling control cycle executes a sampling control program to collect the output voltage and output current of the photovoltaic array; each maximum power point tracking cycle executes a maximum power point tracking program to change the reference power or reference voltage once. When programming, define S _mppt as the identification of the stage where the maximum power point is tracked.

For the multi-peak P-V curve as shown in Figure 3, the basic steps of the multi-peak maximum power point tracking method disclosed by the present invention are as follows:

See Figure 1, Figure 2, Figure 3, Figure 4, Figure 5, Figure 6 and Figure 7,

Step 1, power closed-loop control sweep:

In each maximum power point tracking cycle, the maximum power point tracking method gives a reference power value in the manner of "current reference power value = last reference power value + power step size". Before the next maximum power point tracking cycle arrives, the sampling control program will be called once in each control cycle.

In the sampling process, the output voltage and output current of the photovoltaic array are detected online to obtain the real-time output power of the photovoltaic array. Then, the power closed-loop control is used to make the real-time output power track the reference power, so as to obtain the control signal of the maximum power point tracking circuit.

The process of power closed-loop control is to update the global maximum power point information by comparing the real-time output power with the known global maximum power after each real-time output power is obtained, that is, if P=U _pv *I _pv >P _m , Then U _m = U _pv , P _m = P, where P is the real-time output power, U _pv is the real-time detected photovoltaic array output voltage, I _pv is the real-time detected photovoltaic array output current, and U _m is the known maximum The voltage at the power point, P _m is the power at the known maximum power point. The reference power of power closed-loop control is determined by the formula P _r (k)=P _r (k-1)+P _step , where P _r (k) is the current reference power to be tracked, and P _r (k-1) is The last tracked reference power, P _step is the step size of the reference power change. Since the power closed-loop control is locally stable on the output curve of the photovoltaic array, when the reference power is greater than the maximum power point of the output curve of the photovoltaic array, the output voltage of the photovoltaic array will slide to the minimum operating voltage of the photovoltaic inverter, realizing the photovoltaic Sweep of the array output curve.

Step 2. Determine whether the output voltage of the photovoltaic array is lower than the minimum operating voltage of the photovoltaic inverter? If it is less than, go to step 3, otherwise repeat step 1 and step 2.

By repeating step 1 and step 2, the working point of the system passes through a→b→c→d→e→f→g→h (U _min ) sequentially in Figure 3, realizing the scanning of the output voltage range of the photovoltaic array, and realizing The location of the global maximum power point M. For the area between e→f point and g→h(U _min ) point in Figure 3, the power closed-loop control completes the scanning of e→f point within one maximum power point tracking cycle. After the scanning process ends, the maximum power point information (U _m , P _m ) is the information of point "1" near the global maximum power point M.

Step 3, three-point cooperative variable step size local search:

First, based on the maximum output power point information (U _m , P _m ) of the photovoltaic array obtained in step 1 and step 2, set three initial operating points; the three initial operating points are according to U ₁ (k)=U _m , U ₂ (k)=U _m -U _step , U ₃ (k)=U _m +U _step is determined, where U ₁ (k) is the voltage of the first working point, U ₂ (k) is the voltage of the second working point , U ₃ (k) is the voltage at the third operating point, U _m is the voltage at the known maximum power point, and U _step is the initial disturbance step size. In the next three periods of the maximum power point, through the voltage closed-loop control, the photovoltaic array will work to three operating points in turn (the variable C _W is the current operating point mark), and at the same time, the output voltage and output voltage of the photovoltaic array detected online will be used. current to obtain the output power at each operating point. At the end of the maximum power point tracking period, the global maximum power point information and the maximum power point information experienced by each operating point are updated. After the three working points work in turn, use the obtained information of the three working points to jointly determine the working voltage of the next round of three working points; where the working voltage of the next round of three working points is determined according to the following formula:

In the formula, ΔU ₁ (k), ΔU ₂ (k), ΔU ₃ (k) are the voltage disturbances in the current maximum power point tracking period of the three operating points, ΔU ₁ (k-1), ΔU ₂ (k-1 ), ΔU ₃ (k-1) is the voltage disturbance in the maximum power point tracking cycle before the three operating points, U _m1 , U _m2 , U _m3 are the maximum power point voltage values experienced by the three operating points respectively, U ₁ (k-1), U ₂ (k-1), U ₃ (k-1) are the voltages of the maximum power point tracking cycle before the three operating points, ω∈(0,1), c ₁ ∈(0, 1), c ₂ ∈ (0,1) is the adjustment parameter of the working point voltage step.

Step 4, judge whether the three operating points are close enough according to the voltage values of the three operating points? If the following conditions: P _m ≠ 0, U _m - U _m1 < U _step /10, U _m - U _m2 < U _step /10, U _m - U _m3 < U _step /10 are satisfied at the same time, among them P _m is the power at the known maximum power point, U _m is the voltage at the known maximum power point, U _m1 , U _m2 , and U _m3 are the voltage values at the maximum power points experienced by the three operating points respectively, and U _step is the initial disturbance Step size, then turn to step 5 if it is close enough, otherwise repeat the above steps 3 and 4. By repeating steps 3 and 4, the operating point of the system will converge to the global maximum power point M, that is, U _m = U _M , P _m = P _M , where U _M and P _M are the voltage and power of point M respectively.

Step 5, constant voltage steady-state tracking:

First use voltage closed-loop control to stabilize the output voltage of the photovoltaic array at the maximum power point determined in steps 3 and 4, that is, U _pv =U _m =U _M . Then use the real-time detection of the output voltage and output current of the photovoltaic array to calculate the real-time output power. Afterwards, by comparing the real-time output power with the maximum output power obtained in steps 3 and 4, the relative power change and the cumulative relative power change are calculated according to the following formula: E _p =1-U _pv *I _pv /P _m , E= ΣE _p , where E _p is the relative power variation of the current maximum power point tracking period, and E is the cumulative relative power variation of multiple tracking periods. When constant voltage steady-state tracking is performed, the maximum power point information is no longer updated.

Step 6, based on the relative power variation, determine whether the environment has changed drastically? If so, first set the power reference value P _r (k)=(U _pv *I _pv )/2, and at the same time reset the maximum power point power P _m =0, where U _pv is the real-time detected photovoltaic array output voltage, I _pv is the output current of the photovoltaic array detected in real time. Then restart step 1 to re-scan. In the two cases shown in Figure 4, step 1 will be restarted.

Step 7, based on the cumulative relative power change, determine whether the environment has changed slowly? If so, restart step 3, otherwise go to step 5. When judging whether the environment changes slowly, first judge whether the steady-state tracking cycle time recorded by the timer exceeds the specified time, if it is overtime, reset the timer and the accumulated relative power change, otherwise continue timing; if within the specified time, the accumulated power If the relative change E>0.05, it means that the environment has changed slightly, that is, there is a slight offset between the actual maximum power point and the stored maximum power point, then restart step 3 to perform a three-point cooperative local search with variable step size.

The above search process is a specific implementation process of the present invention, and the relevant flow charts of the maximum power point tracking process are shown in FIG. 5 , FIG. 6 and FIG. 7 .

Claims (8)

1. A photovoltaic array multi-peak maximum power point tracking method based on power closed-loop scanning comprises the steps of detecting the output voltage and the output current of a photovoltaic array on line, and is characterized by comprising the following steps:

step 1, firstly, detecting output voltage and output current of a photovoltaic array on line to obtain real-time output power output by the photovoltaic array, and then utilizing power closed-loop control to enable the real-time output power to track reference power so as to obtain a control signal of a maximum power point tracking circuit;

step 2, judging whether the output voltage of the photovoltaic array is smaller than the lowest working voltage of the photovoltaic inverter, if so, turning to step 3, otherwise, repeating the step 1 and the step 2;

step 3, setting three initial working points based on the maximum output power of the photovoltaic array and the corresponding output voltage value thereof obtained in the step 1 and the step 2, sequentially working the photovoltaic array to the three working points by using voltage closed-loop control, simultaneously obtaining the output power of each working point by using the output voltage and the output current of the online detected photovoltaic array, updating the global maximum power point information and the maximum power point information experienced by each working point when the maximum power point tracking period is finished, and jointly determining the working voltage of the next round of three working points by using the obtained information of the three working points after the three working points work in turn;

step 4, judging whether the three working points simultaneously meet the following conditions according to the voltage values of the three working points: p_m≠0、U_m-U_m1＜U_step/10、U_m-U_m2＜U_step/10、U_m-U_m3＜U_step/10, wherein P_mPower, U, of known maximum power point_mVoltage, U, of known maximum power point_m1、U_m2、U_m3Respectively, the voltage value, U, at the maximum power point experienced by each of the three operating points_stepIf the disturbance step length is the initial disturbance step length, the step 5 is carried out, and if the disturbance step length is not the initial disturbance step length, the step 3 and the step 4 are repeated;

step 5, firstly, stabilizing the output voltage of the photovoltaic array on the maximum power point determined in the step 3 and the step 4 by using voltage closed-loop control, then calculating real-time output power by using the output voltage and the output current of the photovoltaic array detected in real time, and then calculating the relative power variation and accumulating the relative power variation by comparing the real-time output power with the maximum output power obtained in the step 3 and the step 4;

step 6, judging whether the environment is changed violently or not based on the relative power variation, and if so, restarting the step 1;

and 7, judging whether the environment slowly changes or not based on the accumulated relative power variation, if so, restarting the step 3, and otherwise, turning to the step 5.

2. The method as claimed in claim 1, wherein the power closed-loop control in step 1 is performed by comparing the real-time output power with the known global maximum power to update the global maximum power point information after each time the real-time output power is obtained, i.e. if P ═ U-_pv*I_pv＞P_mThen U is_m＝U_pv,P_mP in the formula is real-time output power, U_pvPhotovoltaic array output voltage, I, for real-time detection_pvPhotovoltaic array output current, U, for real-time detection_mVoltage, P, of known maximum power point_mIs the power at which the maximum power point is known.

3. The method of claim 1, wherein the reference power in step 1 is represented by formula P_r(k)＝P_r(k-1)+P_stepDetermining P in the formula_r(k) For the current reference power, P, to be tracked_r(k-1) reference power, P, for last tracking_stepIs a step size of the reference power change.

4. The method as claimed in claim 1, wherein the three initial operating points in step 3 are according to U₁(k)＝U_m，U₂(k)＝U_m-U_step，U₃(k)＝U_m+U_stepDetermine U therein₁(k) Voltage, U, at a first operating point₂(k) Voltage, U, of the second operating point₃(k) Voltage, U, at a third operating point_mVoltage, U, of known maximum power point_stepIs the initial perturbation step size.

5. The method for tracking the maximum power point of the photovoltaic array with multiple peaks based on the power closed-loop scanning as claimed in claim 1, wherein the operating voltages of the next three operating points in the step 3 are determined according to the following formula:

in the formula₁(k)、ΔU₂(k)、ΔU₃(k) Voltage disturbance quantity, delta U, of tracking period of current maximum power point of three working points₁(k-1)、ΔU₂(k-1)、ΔU₃(k-1) is the voltage disturbance quantity of the maximum power point tracking period before the three working points, U_m1、U_m2、U_m3For maximum power point voltage values, U, experienced by each of the three operating points₁(k-1)、U₂(k-1)、U₃(k-1) is the voltage of the maximum power point tracking period before three working points, and omega belongs to (0,1) and c₁∈(0,1)、c₂And epsilon (0,1) is a working point voltage step length adjusting parameter.

6. The method of claim 2, wherein the relative power variation and the accumulated relative power variation in step 5 are calculated according to the following equations: e_p＝1-U_pv*I_pv/P_m，E＝ΣE_pIn the formula E_pThe relative power variation of the current maximum power point tracking period is E, and the accumulated relative power variation of a plurality of tracking periods is E.

7. The method of claim 1, wherein the power reference value is P when restarting step 1 in step 6_r(k)＝(U_pv*I_pv) Setting/2 while resetting maximum power point power P_m0, U in the formula_pvPhotovoltaic array output voltage, I, for real-time detection_pvAnd outputting current for the photovoltaic array detected in real time.

8. The method for tracking the maximum power point of the photovoltaic array with multiple peaks based on the power closed-loop scanning as claimed in claim 1, wherein the step of determining that the environment in step 7 has changed slowly comprises,

(1) judging whether the steady-state tracking period time recorded by the timer exceeds a specified time, if so, resetting the timer and accumulating the relative power variation, otherwise, continuing to time;

(2) if the accumulated power relative change E is larger than 0.05 in a specified time, which indicates that the environment slightly changes, namely a slight deviation occurs between the actual maximum power point and the memorized maximum power point, the step 3 is restarted to perform local search.