A Multi-Peak MPPT Method Based on Improved PSFOA-GBS Algorithm and Improved Incremental Conductivity Method

Abstract

This invention discloses a multi-peak MPPT method based on an improved PSFOA-GBS algorithm and an improved incremental conductance method. The improved PSFOA-GBS algorithm uses optimal point set theory to generate a uniformly distributed initial population of individuals, then integrates mixed behaviors to achieve global optimization, quickly locating the region near the global maximum power point. It outputs the duty cycle signal corresponding to the globally optimal individual and checks if its threshold is less than a fluctuation threshold. If less, it switches to the improved incremental conductance method for local search of the photovoltaic maximum power point. It collects the output power of the photovoltaic array, calculates the power change percentage, and checks if it is greater than or equal to a set power threshold. If greater than or equal to, it restarts the PSFOA-GBS algorithm to track and search for the photovoltaic maximum power point. This invention uses an improved high-gain secondary boost DC-DC converter to replace the traditional Boost converter, further improving the power generation efficiency and tracking accuracy of the photovoltaic system under complex lighting conditions, effectively adapting to the voltage boost requirements of the photovoltaic array under local shading.

Description

Technical Field

[0001] This invention relates to the field of power point optimization in photovoltaic systems, specifically to a multi-peak MPPT method based on an improved PSFOA-GBS algorithm and an improved incremental conductance method. Background Technology

[0002] In actual operation, local shading (such as building or tree obstruction) in photovoltaic power generation systems can cause the power efficiency (PU) curve of the photovoltaic array to exhibit a multi-peak pattern. Traditional MPPT algorithms (such as perturbation and observation P&O and traditional incremental conductance method INC) can only track the local maximum power point (LMPP) and cannot locate the global maximum power point (GMPP), resulting in a decrease in system power generation efficiency. Traditional boost converters also have the problem of limited gain.

[0003] To address the multi-peak problem, existing technologies employ swarm optimization algorithms (such as Particle Swarm Optimization (PSO) and Genetic Algorithm (GA)) for global optimization. However, these algorithms have limitations: PSO is prone to getting trapped in local optima, and GA has slow convergence speed; the standalone Parrot Optimization Algorithm (POA) uses random initialization, resulting in uneven population distribution and insufficient exploration capabilities in high-dimensional scenarios; while the standalone Heron Optimization Algorithm (SBOA) exhibits strong adaptability to high dimensions, its initial population randomness leads to a high risk of local optima. Furthermore, even if a global optimization algorithm finds the vicinity of the GMPP region, the lack of a high-precision local tracking mechanism can still result in steady-state power loss due to oscillations. Therefore, there is an urgent need for a hybrid MPPT algorithm that combines "global optimization breadth" and "local tracking accuracy," adaptable to high-gain converters, to improve voltage gain and system reliability under low-voltage input conditions, and to address the multi-peak challenge under complex lighting conditions. Summary of the Invention

[0004] Purpose of the invention: To address the problems mentioned in the background art, this invention discloses a multi-peak MPPT method based on an improved PSFOA-GBS algorithm and an improved incremental conductance method. By combining the improved Parrot-Eagle Fusion Optimization Algorithm (PSFOA-GBS) with the improved incremental conductance method, the multi-peak interference caused by local shading is effectively offset. At the same time, by utilizing the characteristics of a high-gain converter, the voltage boost requirement of the photovoltaic array under low light conditions is met, enabling the photovoltaic power generation system under local shading to maintain a high power generation efficiency.

[0005] Technical solution:

[0006] This invention discloses a multi-peak MPPT method based on an improved PSFOA-GBS algorithm and an improved incremental conductance method. The method includes the following steps:

[0007] S1: Based on the circuit structure of the photovoltaic array and the characteristics of the improved high-gain double boost DC-DC converter, the parameters of the improved PSFOA-GBS algorithm are initialized.

[0008] S1.1: Improvements to the PSFOA-GBS algorithm: Introduce a best-point set initialization strategy; integrate mixed behaviors to update individual positions; perform adaptive weight adjustment using the globally optimal individual;

[0009] S2: Execute PSFOA-GBS global optimization, output the duty cycle signal corresponding to the global optimal individual, control the output power of the photovoltaic array based on the duty cycle signal and calculate the threshold. If the threshold is less than the preset fluctuation threshold: switch to the improved conductivity incremental method; otherwise, return to S1.1.

[0010] S3: Based on the improved incremental conductivity method, perform local precise tracking of the power point, use an adaptive step size formula to adjust the duty cycle, collect the output power of the photovoltaic array, and if the percentage change in power is greater than or equal to the set threshold, return to re-track and search for the maximum power point.

[0011] Furthermore, the improved high-gain double-boost DC-DC converter described in S1 possesses characteristics such as high voltage gain, low switching voltage stress, and common ground. Its ideal voltage gain formula is:

[0012]

[0013] Among them, V in D represents the input voltage of the improved high-gain double-boost DC-DC converter, and V represents the duty cycle of the IGBT switch in the improved high-gain double-boost DC-DC converter. o This refers to the output voltage of the improved high-gain secondary boost DC-DC converter.

[0014] Furthermore, the optimal point set initialization strategy introduced in S1.1 includes the following specific steps:

[0015] The search space of PSFOA-GBS is defined as the duty cycle range of the improved high-gain double-boost DC-DC converter, from 0 to 1, i.e., the individual position. Corresponding duty cycle;

[0016] The initial individual is generated using the optimal point set theory, and the formula is as follows:

[0017]

[0018] in, For the first The initial duty cycle of each individual; =1 is the upper limit of the duty cycle; =0 sets the lower limit of the duty cycle; mod() is the function to extract the decimal part; is the good point set generation factor, N is the population size; k is the good point set adjustment parameter;

[0019] Initialize other parameters of PSFOA-GBS: Levy distribution parameter in the hybrid behavior = 1.5; fitness function , is the individual corresponding photovoltaic output power.

[0020] Furthermore, the specific process of updating the individual position by fusing the hybrid behavior in S1.1 is as follows:

[0021] Initialize individuals according to the good point set initialization strategy to form the initial population individual positions with uniform distribution; update the individual positions by fusing the hybrid behavior, that is, each individual randomly executes the hybrid behavior of "foraging - hunting", "staying - escaping", "communicating - running away", "fear - avoiding" to update the individual positions; perform individual update and global optimal individual update: for each updated individual , collect the output power of the photovoltaic array as the fitness; if > , then update the individual , similarly if > , then update , otherwise .

[0022] Furthermore, the calculation of the fluctuation threshold and the threshold preset in S2 are specifically as follows:

[0023] The specific formula for calculating the fluctuation threshold is:

[0024]

[0025] where Th is the fluctuation threshold, P t = U t · I t , P t-1 = U t-1 · I t-1 are the photovoltaic output powers at the current moment and the previous moment respectively; the preset fluctuation threshold Tho = 0.05, if Th < Tho, then switch to the improved conductance increment method.

[0026] Furthermore, the specific steps of the improved conductance increment method are as follows: by setting the switching condition, after finding near the initial maximum power point in the later stage of iterative convergence, perform local precise search for the maximum power point and introduce and calculate the adaptive step size:

[0027] The output voltage and output current of the photovoltaic array are measured and collected, and the voltage change dU and current change dI of adjacent operating points of the photovoltaic array are determined based on the output voltage and output current.

[0028] When the voltage change dU=0, if the current change dI=0 at the same time, the local search will track the maximum power point and terminate, indicating that the photovoltaic power generation system is already working at the photovoltaic maximum power point.

[0029] If dU≠0, then calculate and compare the conductance increment dI / dU and the instantaneous conductance I / U to perform a local search;

[0030] When the voltage change dU≠0, if the conductance increment is equal to the instantaneous conductance, i.e. dI / dU=-I / U, then the local search will reach the maximum power point and terminate, indicating that the photovoltaic power generation system is already operating at the photovoltaic maximum power point.

[0031] If the conductance increment is greater than the instantaneous conductance (i.e., dI / dU > -I / U), then the local search has not tracked the maximum power point, and the step size is reduced to perform the local search again.

[0032] If the conductance increment is less than the instantaneous conductance (i.e., dI / dU < -I / U), then the local search has not tracked the maximum power point, and the step size is increased to perform the local search again.

[0033] Furthermore, the decrease and increase of the step size are achieved through adaptive compensation calculation, and the adaptive step size calculation formula is as follows:

[0034]

[0035] in, The duty cycle at the current moment is the conductivity increment method. This represents the change in duty cycle. The duty cycle at the next moment is the conductivity increment method;

[0036] The formula for calculating the change in duty cycle is:

[0037]

[0038] in, dP / dU is the current across the photovoltaic panel, dP / dU is the rate of change of the PU characteristic of the photovoltaic panel, and C is a constant.

[0039] Furthermore, the restart condition for determining whether to return to re-track and search for the maximum power point if the power change percentage mentioned in S3 is greater than or equal to a set threshold is as follows:

[0040]

[0041] in, The percentage change in power. The current sampling power, For maximum power, To set a threshold.

[0042] Beneficial effects:

[0043] 1. This invention uses an improved high-gain double-boost DC-DC converter to replace the traditional Boost converter. The high-gain characteristics of the improved high-gain double-boost DC-DC converter solve the problem of insufficient gain of the traditional converter when the photovoltaic input voltage is low. The low switching stress characteristics reduce system losses. The continuous input current and common ground characteristics are adapted to the output characteristics of the photovoltaic array. In synergy with the hybrid MPPT algorithm, it realizes the dual advantages of "high-gain voltage boost + high-precision MPP tracking", which further improves the overall performance of the photovoltaic power generation system under complex lighting conditions.

[0044] 2. This invention improves the PSFOA-GBS algorithm by introducing a best-point set initialization strategy and fusion hybrid behavior, and by adjusting the adaptive weights of the globally optimal individuals. This further enhances the algorithm's global search capability, avoids getting trapped in local optima, improves the performance of the Parrot algorithm in photovoltaic power generation systems, and makes the algorithm more effective in optimizing the system's output power.

[0045] 3. This invention addresses the problem of slow convergence and large oscillation when approaching the MPP caused by the fixed step size of the traditional incremental conductance method. In the later stage of the iteration, an improved incremental conductance method is incorporated. By setting switching conditions, after finding the vicinity of the initial maximum power point in the later stage of the iteration convergence, a local precise search for the maximum power point is performed. At the same time, an adaptive step size is introduced, which improves the global search accuracy in the early stage of the iteration process and increases the local search speed in the later stage of the iteration. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the MPPT simulation circuit for a photovoltaic array according to an embodiment of the present invention;

[0047] Figure 2 This is a flowchart illustrating the overall process of the method of the present invention.

[0048] Figure 3 This is a graph showing the output characteristics of a photovoltaic array under uniform illumination and illumination shading, based on the improved PSFOA-GBS algorithm and the improved incremental conductivity method of the multi-peak MPPT method in this invention.

[0049] Figure 4 This is a comparison curve of the output power of the multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved conductance increment method of the present invention with that of the prior art under partial shading. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] This invention discloses a multi-peak MPPT method based on an improved PSFOA-GBS algorithm and an improved incremental conductance method. Firstly, as follows... Figure 1 As shown, the photovoltaic power generation system includes a photovoltaic array, an MPPT controller, and an improved high-gain double-boost DC-DC converter. The MPPT controller calculates the real-time output power by sampling the real-time output voltage and current of the photovoltaic array, thereby outputting a PWM control signal with a duty cycle. The PWM control signal is amplified by a drive circuit to effectively drive the IGBT switches in the improved high-gain double-boost DC-DC converter, thus controlling the duty cycle of the IGBT switches. By controlling the output of the photovoltaic array after the converter through the duty cycle signal output by the MPPT controller, the photovoltaic power generation system can operate stably at its maximum power point. The improved high-gain double-boost DC-DC converter features high voltage gain, low switching voltage stress, and common ground characteristics. Its ideal voltage gain formula is:

[0052]

[0053] Among them, V in D represents the input voltage of the improved high-gain double-boost DC-DC converter, and V represents the duty cycle of the IGBT switch in the improved high-gain double-boost DC-DC converter. o This refers to the output voltage of the improved high-gain secondary boost DC-DC converter.

[0054] like Figure 2 As shown, the steps of the multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method are as follows:

[0055] Step 1: Based on the circuit structure of the photovoltaic array and the characteristics of the improved high-gain secondary boost DC-DC converter, the parameters of the improved Parrot-Eagle Fusion Optimization Algorithm (PSFOA-GBS) are initialized, including initializing the duty cycle search range, the population size of PSFOA-GBS, the optimal point set adjustment parameters, the hybrid behavior update coefficient, and defining the fitness function of PSFOA-GBS.

[0056] The duty cycle range of 0 to 1 is the conduction time of the IGBT switching transistor controlled by the MPPT controller, which is equivalent to the working voltage of the photovoltaic array after passing through the converter. That is, the duty cycle corresponding to the position parameters of the individual in the parrot-eagle fusion optimization algorithm is used to replace the working voltage of the photovoltaic array after passing through the converter.

[0057] The initial population is generated using the optimal point set theory, with the following formula:

[0058]

[0059] in, For the first The initial duty cycle of each individual; =1 is the upper limit of the duty cycle; =0 sets the lower limit of the duty cycle; mod() is the function to extract the decimal part; is the optimal point set generation factor (N is the population size); k is the optimal point set adjustment parameter;

[0060] Initialize other PSFOA-GBS parameters: Set the Levy distribution parameters in the mixed behavior. =1.5; fitness function , ( For individuals The corresponding photovoltaic output power.

[0061] Step 2: Perform PSFOA-GBS global optimization based on the initialized parameters: Generate uniformly distributed initial population individual positions through the optimal point set theory, calculate individual fitness, and each individual randomly performs mixed behaviors such as "foraging-hunting", "staying-escape", "communication-running", and "fear-avoidance" to update the individual position and update the global optimal individual;

[0062] The initial individuals are generated using the optimal point set theory to form the initial population, as shown in the formula:

[0063]

[0064] in, For the first The initial duty cycle of each individual; =1 is the upper limit of the duty cycle; =0 sets the lower limit of the duty cycle; mod() is the function to extract the decimal part; is the optimal point set generation factor, N is the population size, and k is the optimal point set adjustment parameter;

[0065] Merging and mixing behaviors to update individual location:

[0066] The formula for updating foraging-hunting behavior (global exploration) is:

[0067]

[0068] in, For the t-th iteration, the... The duty cycle of each individual; Let be the duty cycle of the globally optimal individual in the t-th iteration; ( , ); The average duty cycle of individuals in the t-th iteration; t is the current iteration number;

[0069] The update formula for the stay-escape behavior (partial transition) is:

[0070]

[0071] in, These are random numbers distributed according to a standard normal distribution. It is a vector of all 1s with dimension dim = 1.

[0072] The update formula for communication-running behavior (group information transmission) is:

[0073]

[0074] Where P~U(0,1) is the uniform random probability; 0.2 is the step size scaling factor;

[0075] The update formula for fear-avoidance behavior (local optimal avoidance) is:

[0076]

[0077] For each updated individual Collect the output power of the photovoltaic array As fitness; if > Then update the individual Similarly, if > Then update ,otherwise .

[0078] Step 3: Output the duty cycle signal corresponding to the globally optimal individual, which is the voltage (U) output by the photovoltaic array at the previous time (t-1) and the current time (t). t-1 U t ) and current (I) t-1 I t ) and power (P) t-1 P t Calculate the threshold and determine whether it is less than the preset fluctuation threshold; if it is less, switch to the improved incremental conductance method; otherwise, return to step two to continue iterating.

[0079] The specific formula for calculating the fluctuation threshold is as follows:

[0080]

[0081] Where Th is the fluctuation threshold, P t =U t ·I t P t-1 =U t-1 ·I t-1 These are the photovoltaic output power at the current moment and the previous moment, respectively; the preset fluctuation threshold Tho = 0.05 (i.e., 5%).

[0082] Step 4: Perform local precise tracking based on the improved incremental conductance method: Determine the duty cycle adjustment direction, and adjust the duty cycle using an adaptive step size formula to make the photovoltaic array operate at the duty cycle corresponding to the global maximum power point, thereby achieving stable tracking of GMPP;

[0083] The specific steps of the improved incremental conductance method are as follows:

[0084] The output voltage and output current of the photovoltaic array are measured and collected, and the voltage change dU and current change dI of adjacent operating points of the photovoltaic array are determined based on the output voltage and output current.

[0085] If dU=0, then a local search is performed by determining the current change dI between adjacent operating points of the photovoltaic array;

[0086] When the voltage change dU=0, if the current change dI=0 at the same time, the local search will track the maximum power point and terminate, indicating that the photovoltaic power generation system is already working at the photovoltaic maximum power point.

[0087] If dI > 0 at this time, the local search has not tracked the maximum power point, and the step size is reduced to start the local search again;

[0088] If dI < 0 at this time, the local search has not tracked the maximum power point, and the step size is increased to perform the local search again;

[0089] If dU≠0, then calculate and compare the conductance increment dI / dU and the instantaneous conductance I / U to perform a local search;

[0090] When the voltage change dU≠0, if the conductance increment is equal to the instantaneous conductance, i.e. dI / dU=-I / U, then the local tracking reaches the maximum power point and terminates, indicating that the photovoltaic power generation system is working at the photovoltaic maximum power point.

[0091] If the conductance increment is greater than the instantaneous conductance (i.e., dI / dU > -I / U), then the local search has not tracked the maximum power point, and the step size is reduced to perform the local search again.

[0092] If the conductance increment is less than the instantaneous conductance (dI / dU < -I / U), then the local search has not found the maximum power point, and the step size is increased to perform the local search again.

[0093] The step size for the incremental conductance method is:

[0094]

[0095] The formula for calculating the change in duty cycle is:

[0096]

[0097] in, This represents the change in duty cycle using the conductivity increment method. dP / dU is the current across the photovoltaic panel, dP / dU is the rate of change of the PU characteristic of the photovoltaic panel, and C is a constant, which is selected as 0.001 in this embodiment.

[0098] Step 5: Collect the output power of the photovoltaic array, calculate the power change percentage with the maximum output power, and determine whether the power change percentage is greater than or equal to the set power threshold. If it is greater than or equal to the threshold, restart the improved parrot-egret fusion optimization algorithm to track and search for the maximum power point of the photovoltaic array.

[0099] The restart determination condition is:

[0100]

[0101] in, The percentage change in power. The current sampling power, For maximum power, To set a power threshold.

[0102] Figure 3 The PU characteristic curve of the photovoltaic array in the example is shown. During the simulation, the temperature was set to a constant 25°C, and the irradiance received by the three photovoltaic modules were 1000, 800, and 600 W / m², respectively. 2 The maximum power point is P MAX =4591W. Comparisons were made using the improved Parrot-Eagle fusion optimization algorithm and the improved incremental conductance method (this invention), as well as the Particle Swarm Optimization and Gravity Search Hybrid Algorithm (PSOGSA) and the Grey Wolf Search Algorithm (GWO). The simulation time was set to 1 second. The output power curves of the three methods under partial occlusion are shown below. Figure 4 As shown.

[0103] The MPPT system based on the improved parrot-egret fusion optimization algorithm and the improved incremental conductance method tracked the global GMPP after 0.29s with a power of 1116.2W and a tracking accuracy of 99%. The MPPT system based on the particle swarm optimization and gravity search hybrid algorithm (PSOGSA) tracked the global GMPP after 0.6s with a power of 1114W and a tracking accuracy of 98.3%. The MPPT system based on the gray wolf search algorithm (GWO) converged to the vicinity of the global maximum power point in 0.71s, but the power fluctuation was too severe, with the final power ranging from 1065.6 to 1116.2W and a tracking accuracy of 94.51% to 99%. In comparison, it can be seen that the method of the present invention has a faster convergence speed, higher accuracy, and better stability.

[0104] The foregoing description of the embodiments enables those skilled in the art to make or use the present invention. Various modifications to the embodiments will be readily apparent to those skilled in the art. The general principles of the invention may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention should not be limited to the embodiments shown herein, but should cover the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-peak MPPT method based on an improved PSFOA-GBS algorithm and an improved incremental conductance method, characterized in that, The method includes the following steps: S1: Based on the circuit structure of the photovoltaic array and the characteristics of the improved high-gain double boost DC-DC converter, the parameters of the improved PSFOA-GBS algorithm are initialized. S1.1: Improvements to the PSFOA-GBS algorithm: Introduce a best-point set initialization strategy; integrate mixed behaviors to update individual positions; perform adaptive weight adjustment using the globally optimal individual; S2: Perform PSFOA-GBS global optimization, output the duty cycle signal corresponding to the global optimal individual, control the output power of the photovoltaic array based on the duty cycle signal and calculate the threshold. If the threshold is less than the preset fluctuation threshold: switch to the improved conductivity incremental method. Otherwise, return to S1.1; S3: Based on the improved incremental conductivity method, perform local precise tracking of the power point, use an adaptive step size formula to adjust the duty cycle, collect the output power of the photovoltaic array, and if the percentage change in power is greater than or equal to the set threshold, return to re-track and search for the maximum power point.

2. The multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method according to claim 1, characterized in that, The improved high-gain double-boost DC-DC converter described in S1 features high voltage gain, low switching voltage stress, and common ground. Its ideal voltage gain formula is: ； Among them, V in D represents the input voltage of the improved high-gain double-boost DC-DC converter, and V represents the duty cycle of the IGBT switch in the improved high-gain double-boost DC-DC converter. o This refers to the output voltage of the improved high-gain secondary boost DC-DC converter.

3. The multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method according to claim 2, characterized in that, The optimal point set initialization strategy introduced in S1.1 includes the following steps: The search space of PSFOA-GBS is defined as the duty cycle range of the improved high-gain double-boost DC-DC converter, from 0 to 1, i.e., the individual position. Corresponding duty cycle; The initial individual is generated using the optimal point set theory, and the formula is as follows: ； in, For the first The initial duty cycle of each individual; =1 is the upper limit of the duty cycle; =0 sets the lower limit of the duty cycle; mod() is the function to extract the decimal part; is the optimal point set generation factor, N is the population size, and k is the optimal point set adjustment parameter; Initialize other PSFOA-GBS parameters: Levy distribution parameters in mixed behavior =1.5; fitness function , For individuals The corresponding photovoltaic output power.

4. The multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method according to claim 3, characterized in that, The fusion and hybrid behavior update of individual position described in S1.1 specifically includes: Individuals are initialized according to the optimal point set initialization strategy to form a uniformly distributed initial population of individual positions. Individual positions are updated by fusing mixed behaviors, i.e., each individual randomly performs a mixed behavior of "foraging-hunting," "staying-escape," "communicating-running," and "fear-avoidance" to update its position. Individual updates and global optimal individual updates are then performed: for each updated individual... Collect the output power of the photovoltaic array As fitness; if > Then update the individual Similarly, if > Then update ,otherwise .

5. The multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method according to claim 4, characterized in that, The fluctuation threshold calculation and threshold preset described in S2 are as follows: The specific formula for calculating the fluctuation threshold is as follows: ； Among them, Th is the fluctuation threshold, P t =U t ·I t 、P t-1 =U t-1 ·I t-1 are the photovoltaic output powers at the current moment and the previous moment respectively; the preset fluctuation threshold Tho = 0.

05. If Th < Tho, switch to the improved conductance increment method.

6. The multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method according to claim 5, characterized in that, The specific steps of the improved incremental conductance method are as follows: by setting switching conditions, after finding the vicinity of the initial maximum power point in the later stage of iterative convergence, a local precise search for the maximum power point is performed, and an adaptive step size is introduced and calculated. The output voltage and output current of the photovoltaic array are measured and collected, and the voltage change dU and current change dI of adjacent operating points of the photovoltaic array are determined based on the output voltage and output current. When the voltage change dU=0, if the current change dI=0 at the same time, the local search will track the maximum power point and terminate, indicating that the photovoltaic power generation system is already working at the photovoltaic maximum power point. If dU≠0, then calculate and compare the conductance increment dI / dU and the instantaneous conductance I / U to perform a local search; When the voltage change dU≠0, if the conductance increment is equal to the instantaneous conductance, i.e. dI / dU=-I / U, then the local search will reach the maximum power point and terminate, indicating that the photovoltaic power generation system is already operating at the photovoltaic maximum power point. If the conductance increment is greater than the instantaneous conductance (i.e., dI / dU > -I / U), then the local search has not tracked the maximum power point, and the step size is reduced to perform the local search again. If the conductance increment is less than the instantaneous conductance (i.e., dI / dU < -I / U), then the local search has not tracked the maximum power point, and the step size is increased to perform the local search again.

7. The multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method according to claim 6, characterized in that, The decrease and increase of the step size are achieved through adaptive compensation calculation, and the adaptive step size calculation formula is as follows: ； in, The duty cycle at the current moment is the conductivity increment method. This represents the change in duty cycle. The duty cycle at the next moment is the conductivity increment method; The formula for calculating the change in duty cycle is: ； in, dP / dU is the current across the photovoltaic panel, dP / dU is the rate of change of the PU characteristic of the photovoltaic panel, and C is a constant.

8. The multi-peak MPPT method based on the improved PSFOA-GBS algorithm and the improved incremental conductance method according to claim 1, characterized in that, The restart condition for S3, which states that if the percentage change in power is greater than or equal to a set threshold, is as follows: ； in, The percentage change in power. The current sampling power, For maximum power, To set a threshold.

Images