Photovoltaic MPPT control method based on mayfly naiad fusion algorithm
Through the mayfly beetle whisker fusion algorithm, the adaptability and accuracy problems of the photovoltaic MPPT control algorithm in local shading peak scenarios are solved, and efficient maximum power point tracking of the photovoltaic system under complex working conditions is achieved, thereby improving the overall performance of the system.
Patent Information
- Application Number
- CN202510962530.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-11
- Publication Date
- 2025-10-17
AI Technical Summary
The existing photovoltaic MPPT control algorithm has poor adaptability in local shading peak scenarios, low search accuracy and slow dynamic response speed, resulting in long initial search time and poor output power stability under complex working conditions, making it difficult to accurately capture the maximum power point.
The mayfly and longhorn beetle whisker fusion algorithm is adopted. By cross-coupling the parameters of the longhorn beetle whisker algorithm and the mayfly algorithm, the longhorn beetle whisker sensor and the mayfly individual speed update are introduced, and combined with the dynamic update mating pool mechanism, an adaptive MPPT control strategy is constructed to achieve efficient maximum power point tracking of the photovoltaic system under complex working conditions.
It improves the maximum power point tracking accuracy and speed of the photovoltaic system under complex working conditions, shortens the tracking time, and improves the overall efficiency and stability of the photovoltaic system.
Smart Images
Figure CN120803200A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to the technical field of photovoltaic power generation, and in particular to a photovoltaic MPPT control method based on a caddisfly-moth antenna fusion algorithm. BACKGROUND
[0002] In a photovoltaic system, in order to achieve maximum power output, the performance of a photovoltaic module, the influence of temperature, the power factor and pollution and many other factors need to be comprehensively considered, and the duty cycle is adjusted reasonably so that the system can operate in the best state under various working conditions, which is photovoltaic MPPT control. In the prior art, a tracking algorithm based on a fixed mathematical model or a rule (such as a grey wolf algorithm, a butterfly algorithm and the like) is often used to adjust the duty cycle, and the moth antenna algorithm is one of them, but it has poor adaptability in the local shading peak value scene, low search precision and slow dynamic response speed, which leads to long search time in the early stage under complex working conditions, poor output power stability and difficulty in accurately capturing the maximum power point. Therefore, it is urgent to improve the tracking algorithm based on the fixed mathematical model or the rule, overcome such problems, and give a photovoltaic MPPT control scheme with higher search efficiency and more accurate results. SUMMARY
[0003] The application discloses a photovoltaic MPPT control method based on a caddisfly-moth antenna fusion algorithm, which is used for solving the problems in the background art.
[0004] TECHNICAL SCHEME
[0005] The application discloses a photovoltaic MPPT control method based on a caddisfly-moth antenna fusion algorithm, which comprises the following steps:
[0006] S1 initializes a moth population and a caddisfly population, and the moth individual is a photovoltaic system voltage;
[0007] S2 appends a moth individual of a caddisfly algorithm with a moth antenna sensor, introduces a male individual speed update of the caddisfly algorithm into moth population position update by using gradient information update of the moth antenna, and adopts a dynamic update mating pool mechanism to form a caddisfly-moth antenna fusion algorithm;
[0008] S3 obtains output power of the photovoltaic system under control of each duty cycle as fitness of the moth individual of the caddisfly-moth antenna fusion algorithm;
[0009] S4 carries out fitness calculation and carries out maximum number of iterations, records optimal individual and outputs corresponding voltage, stabilizes photovoltaic system at optimal voltage position, realizes photovoltaic maximum power point search MPPT function.
[0010] Further, S1 specific steps are as follows:
[0011] Initialize the number of beetles in two-dimensional space N, select random values in [0,1] to generate the initial position of the beetle population, the beetle individual represents the voltage in the photovoltaic system, the mayfly population is initialized, the number of male and female mayflies is M, the initial position is consistent with the beetle, the number of male and female mayflies M and the number of beetle individuals N satisfy M=2N, and the mayfly is the beetle whisker reset step.
[0012] Further, the mayfly individual of the mayfly algorithm of S2 is added with a beetle whisker sensor, and the specific steps are as follows:
[0013] The concept of beetle whisker is introduced, the number of beetle individuals in the population in two-dimensional space is N, the initial optimal position of the initialized group is introduced, the beetle whisker is generated, and a random vector in the direction of the beetle whisker is created.
[0014] Tactile perception optimization: each mayfly individual is added with a beetle whisker sensor, the left and right whisker position coordinates (x l , x r ) of each beetle are calculated, the power difference Δf=f(x l )-f(x r ) is calculated, and the gradient difference directly determines the splitting direction of the mayfly population; when Δf>0, the population moves to the right side of the whisker direction; otherwise, it moves to the left side of the whisker, realizes searching along the direction of the fastest power rise, adjusts the search range through the dynamic parameter update formula, and updates the positions of the left and right whiskers.
[0015] Further, the male individual speed update of the mayfly algorithm of S2 introduces the gradient information of the beetle whisker, updates the position of the beetle population, and introduces the formula as follows:
[0016]
[0017] In the formula, w is the inertia weight, c is the learning factor, gbest is the optimal position of the group, Xi is the current position sign is the sign function, wherein, The power gradient difference of the left and right whiskers is used to calculate the search direction; when Δf>0, sign[Δf]=1, which guides the male mayfly to move to the right side of the whisker; otherwise, it moves to the left side of the whisker; the inertia weight w and the coefficient s of the gradient difference term satisfy w
[0018] Further, the mayfly-moth antenna fusion algorithm adopts a dynamic update mating pool mechanism, dynamically resets the mayfly mating driving moth step, and the search step s of the moth antenna is directly determined by the disturbance intensity sigma, that is, s=0.5 sigma, when the mayfly mating produces offspring, the attenuation of sigma will synchronously reduce the search step of the moth antenna, and the sensing range of the moth antenna guides the mating direction of the mayfly, and the dynamic sensing radius l of the moth antenna determines the search density of the mayfly population: when l is large, the offspring generated by mayfly mating has a wide distribution range and covers the global search space; when l is small, the offspring position is gathered to the current optimal position of the moth antenna, and the local search is strengthened.
[0019] Further, the S4 specific process is as follows:
[0020] It is judged whether the result of the fitness calculation meets the termination condition of the maximum iteration number, if yes, the iteration is stopped and the corresponding voltage is output, and the photovoltaic system is stabilized at the optimal voltage value; the optimal voltage value is input into the PI regulator, and the output duty cycle of the PI regulator acts on the BOOST circuit and the moth population initialization range at the same time, forming a closed loop parameter update; if not, return to the S3 step to continue iteration.
[0021] Further, the PI regulator adjustment formula is as follows:
[0022]
[0023] Wherein e(t)=V PV (t)-V ref , V ref is the reference voltage of the photovoltaic system.
[0024] Beneficial effects:
[0025] 1. The mayfly-moth antenna fusion algorithm is obtained by attaching a moth antenna sensor to the mayfly individual of the mayfly algorithm, introducing the mayfly algorithm male individual speed update into the moth antenna gradient information update moth population position, effectively avoiding the dilemma of the result falling into a local extreme value, setting the output power as the fitness function of the moth antenna, and the mayfly-moth antenna fusion algorithm can carry out optimization work according to the real-time change of the output power of the photovoltaic system, and each will directly affect the fitness value, and positively affect the accuracy of the maximum power point tracking.
[0026] 2. The mayfly-moth antenna fusion algorithm adopts a dynamic update mating pool mechanism, based on a double-parameter cooperative attenuation mechanism, so that the mayfly-moth antenna fusion algorithm can quickly cover the global search space in the early stage, and focus on local fine optimization in the later stage, introduces an adaptive step formula, solves the problem that the traditional fixed step cannot balance the convergence speed and accuracy in maximum power point tracking, and avoids the tracking oscillation or slow convergence phenomenon caused by the fixed step.
[0027] 3. The application adds a PI controller, through the algorithm to search for the maximum theoretical voltage, through the PI regulator can quickly actual voltage, converted to duty cycle transmission DCDC circuit IGBT, improve the speed of maximum power tracking. BRIEF DESCRIPTION OF DRAWINGS
[0028] Figure 1 The specific method flowchart of the application is shown in the figure;
[0029] Figure 2 The 5x1 model diagram of the photovoltaic array in the embodiment of the application is shown in the figure;
[0030] Figure 3 The simulink principle diagram of the photovoltaic system in the embodiment of the application is shown in the figure;
[0031] Figure 4 The power, voltage and current output curve diagram of the photovoltaic system in the embodiment of the application is shown in the figure;
[0032] Figure 5 The multi-peak output diagram of the photovoltaic array in the embodiment of the application is shown in the figure. DETAILED DESCRIPTION
[0033] In order to make the purpose, technical scheme and advantages of the embodiment of the application clearer, the technical scheme in the embodiment of the application will be described clearly and completely below in combination with the drawings in the embodiment of the application. Obviously, the described embodiment is a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.
[0034] As shown in the figure, Figure 1 The application discloses a photovoltaic MPPT control method based on a mayfly dragonfly whisker fusion algorithm, and the method steps are as follows:
[0035] S1 initializes the mayfly and dragonfly population, and the mayfly individual is the voltage of the photovoltaic system;
[0036] The number of mayfly individuals is initialized in a two-dimensional space, random values are selected in [0, 1] to generate the initial positions of the mayfly population, the mayfly individual represents the voltage in the photovoltaic system, the dragonfly population is initialized, there are M female and male dragonflies, the initial positions are consistent with the mayfly, M and the number of mayfly individuals N satisfy M = 2N, and the search space coverage is ensured;
[0037] S2 appends the mayfly individual of the mayfly algorithm with a mayfly whisker sensor, introduces the speed update of the male individual of the mayfly algorithm into the mayfly whisker gradient information update mayfly population position, and adopts a dynamic update mating pool mechanism to constitute the mayfly mayfly whisker fusion algorithm;
[0038] The concept of beetle swarm is introduced: assuming that the number of beetle individuals in the population in two-dimensional space is N, the initial optimal position of the initialized population is introduced, the beetle antenna is generated, and the random vector of the direction of the beetle antenna is created. The direction formula can be expressed as:
[0039]
[0040] where rand() is a random function.
[0041] Tactile perception optimization: each mayfly individual is attached with a beetle antenna sensor, the coordinates (xl, xr) of the left and right antenna positions of each beetle are calculated, the power difference Δf = f(xl)-f(xr) is calculated, and the gradient difference directly determines the splitting direction of the mayfly population. When Δf>0, the population moves to the right side of the antenna direction; otherwise, it moves to the left side of the antenna, achieving the goal of "searching in the direction of the fastest power rise". The search range is adjusted by the dynamic parameter update formula to update the positions of the left and right antennas, and the formula is:
[0042]
[0043] where xl and xr are the position coordinates of the right and left antennas at the tth iteration, respectively; xt is the centroid coordinate at the tth iteration; l is the dynamic perception radius, and λ is the attenuation coefficient. s is the left and right antenna length update factor of the beetle individual in the beetle swarm. b is generated by a random function rand(1, D) to ensure smooth transition from global exploration in the early stage to local fine search in the later stage.
[0044] A dynamic update of the mating pool mechanism is adopted. In the traditional mayfly mating mechanism, the disturbance intensity σ of the offspring generation is mostly a fixed value or a linear decay, which cannot cooperate with the beetle antenna search strategy. The designed algorithm resets the beetle step length dynamically driven by mayfly mating, and the search step length s of the beetle antenna is directly determined by σ, i.e. s = 0.5σ. When mayfly mating produces offspring, the attenuation of σ will simultaneously reduce the search step length of the beetle antenna, achieving "wide step global exploration in the early stage and fine step local fine adjustment in the later stage".
[0045] The offspring generation formula of mayfly algorithm is as follows:
[0046] X child = δX male + (1-δ)X female + σbrand
[0047] δ ∈ [0.4, 0.6] is a genetic factor, and σ is the mayfly disturbance intensity.
[0048] where σ0 is the initial disturbance intensity, λ is the attenuation coefficient, which realizes the coordinated attenuation of disturbance amplitude and sensing range, strengthens the intensive search ability, t is the current iteration number, and T is the maximum iteration number. The mayfly's sense of direction is guided by the mayfly's sense of direction, and the mayfly's dynamic sensing radius l determines the search density of the mayfly population: when l is large, the offspring produced by mayfly mating have a wide distribution range, covering the global search space; when l is small, the offspring position is gathered to the current optimal position of the mayfly, and the local search is strengthened.
[0049] The mayfly algorithm is fused, and the mayfly gradient information is introduced into the speed update of the male individual:
[0050]
[0051] wherein w is the inertia weight, c is the learning factor, gbest is the global optimal position, Xi is the current position, and sign is the sign function. Among them, the power gradient difference Δf of the mayfly's left and right antennae is embedded in the speed update term through the sign function sign[], and the specific effect is as follows: direction strengthening: when Δf>0, sign[Δf]=1, guiding the male mayfly to move to the right side of the antenna (high power area); otherwise, move to the left side of the antenna, avoiding blind diffusion of the population. Weight distribution: the inertia weight w and the coefficient s of the gradient difference term need to satisfy w<s (w=0.7, s=1.2), to ensure the dominant role of the gradient information in the speed update, and to improve the dynamic response speed. The gradient information directly affects the speed update of the male mayfly, guiding it to move in the direction of higher power, and optimizing the direction of population splitting.
[0052] S3 obtains the output power of the photovoltaic system under each duty cycle control as the mayfly antenna fusion algorithm of the mayfly antenna;
[0053] S4 performs fitness calculation and maximum number of iterations, records the optimal individual and outputs the corresponding voltage:
[0054] f=P PV =U PV ×I PV
[0055] wherein P PV is the output power of the photovoltaic system, U PV is the output voltage of the photovoltaic array, and I PV is the output current of the photovoltaic array.
[0056] The photovoltaic system is stabilized at the optimal voltage position, realizing the MPPT function of searching for the maximum power point of photovoltaic.
[0057] The specific process is as follows: it is judged whether the result of the fitness calculation meets the termination condition of the maximum iteration number, if yes, the iteration is stopped and the corresponding voltage is output, and the photovoltaic system is stabilized at the optimal voltage value; the optimal voltage value is input into the PI regulator, the output duty cycle of which acts on the BOOST circuit and the beetle population initialization range at the same time, forming a closed loop parameter update. If not, return to step S3 and continue iteration.
[0058] This embodiment is based on the above method, and a model diagram is built in Simulink, as shown in Figure 2 , which is a photovoltaic array model diagram: the entire photovoltaic partial shadow multi-peak MPPT control system is divided into a photovoltaic array, a BOOST circuit, an MPPT controller and a PWM generator, as shown in Figure 3 .
[0059] As shown in Figure 4 , the photovoltaic system output power, output current and output voltage diagram obtained by Simulink simulation analysis of the present application is shown. It can be seen that the improved fusion algorithm has good MPPT tracking effect and keeps the photovoltaic module power output stable. Therefore, it can be determined that the caddisfly beetle antenna fusion algorithm applied to the MPPT method can improve the tracking speed of the maximum power tracking of the photovoltaic system. The photovoltaic array multi-peak output diagram is shown in Figure 5 .
[0060] As shown in Table 1, the data of the embodiment of the present application and the prior art are compared,
[0061] Table 1
[0062]
[0063]
[0064] In summary, the present application has the following advantages compared with the prior art in the technical scheme:
[0065] The conventional and part of the new photovoltaic partial shadow multi-peak MPPT control technology cannot accurately track the maximum power point of photovoltaic power generation output, is easy to fall into a local optimal solution, and leads to algorithm failure. The caddisfly beetle antenna fusion algorithm proposed in the present application improves the accuracy of global search by introducing a reverse learning function, avoids the problem of falling into a local optimal solution, improves the accuracy of tracking the maximum power point, and fits the light change of the real situation. The improved algorithm not only maintains the global search ability, but also improves the convergence speed, shortens the time of tracking the maximum power point, and improves the overall efficiency of the photovoltaic system. The technical scheme is not only suitable for photovoltaic partial shadow multi-peak MPPT control, but also can be applied to other fields requiring global optimization, has good universality and practical value.
[0066] In general, the technical scheme of the present application solves the shortcomings of the traditional MPPT algorithm, improves the efficiency and adaptability of the photovoltaic power generation system, and has important theoretical significance and application prospect.
[0067] The above description of the embodiments enables one skilled in the art to make or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art. The general principles defined herein can be applied to other embodiments without departing from the spirit or scope of the present application. Thus, the present application should not be limited to the embodiments shown herein but should be given the broadest scope consistent with the principles and novel features disclosed.
Claims
1. A photovoltaic MPPT control method based on the fusion algorithm of the mayfly longhorn beetle, characterized in that: The method comprises the following steps: S1 initializes the population of longhorn beetles and mayflies, and the longhorn beetle individuals are the voltage of the photovoltaic system; S2 adds a beetle whisker sensor to the mayfly individual of the mayfly algorithm, introduces the speed update of the male individual of the mayfly algorithm into the beetle whisker gradient information to update the beetle population position, and adopts a dynamic update mating pool mechanism to form a mayfly and beetle whisker fusion algorithm; S3 obtains the output power of the photovoltaic system under various duty cycle controls as the individual fitness of the longhorn beetle in the fusion algorithm of the mayfly longhorn beetle whiskers; S4 calculates the fitness and performs the maximum number of iterations, records the individual with the best fitness and outputs the corresponding voltage, stabilizes the photovoltaic system at the optimal voltage position, and realizes the photovoltaic maximum power point search MPPT function.
2. The photovoltaic MPPT control method based on the mayfly longhorn beetle whisker fusion algorithm according to claim 1, characterized in that: The specific steps of S1 are as follows: Initialize the number of longicorn individuals N in two-dimensional space, and select random values in [0,1] to generate the initial position of the longicorn population. The longicorn individuals represent the voltage in the photovoltaic system. Initialize the mayfly population with M male and M female mayflies. The initial position is consistent with that of the longicorn. The number of male and female mayflies M and the number of longicorn individuals N satisfy M = 2N. If the mayfly is a longicorn, the step size must be reset.
3. The photovoltaic MPPT control method based on the fusion algorithm of the mayfly beetle whiskers according to claim 1 is characterized in that: The details of the additional beetle whisker sensor for each mayfly in the mayfly algorithm described in S2 are as follows: The concept of beetle whiskers is introduced. The number of beetle individuals in a population in two-dimensional space is N. The initial optimal position of the population is introduced to initialize the beetle whiskers, and a random vector of the beetle whisker direction is created. Tactile perception optimization: Each mayfly individual is equipped with a beetle whisker sensor to calculate the position coordinates of the left and right whiskers of each beetle (x l 、x r ), calculate the power difference Δf=f(x l )-f(x r ), the gradient difference directly determines the division direction of the mayfly population; when Δf>0, the population moves toward the right whisker; otherwise, it moves toward the left whisker, realizing the search along the direction with the fastest power increase. The search range is adjusted through the dynamic parameter update formula to update the positions of the left and right whiskers.
4. The photovoltaic MPPT control method based on the fusion algorithm of the mayfly beetle whiskers according to claim 3 is characterized in that: The male individual speed update of the mayfly algorithm described in S2 introduces the gradient information of the longicorn beetle whiskers to update the position of the longicorn beetle population. The formula introduced is as follows: Where w is the inertia weight, c is the learning factor, gbest is the optimal position of the group, Xi is the current position, and sign is the sign function, where represents the power gradient difference between the left and right whiskers of the longicorn beetle, which is used to calculate the search direction. When Δf>0, sign[Δf]=1, guiding the male mayfly to accelerate toward the right whisker; otherwise, it moves toward the left whisker. The inertia weight w and the coefficient s of the gradient difference term satisfy w<s, w=0.7, s=1.
2.
5. The photovoltaic MPPT control method based on the fusion algorithm of the mayfly beetle whiskers according to claim 1 is characterized in that: The mayfly and longhorn beetle whisker fusion algorithm described in S2 adopts a dynamic mating pool update mechanism to dynamically reset the step size of the longhorn beetle driven by mayfly mating. The search step size s of the longhorn beetle whiskers is directly determined by the disturbance intensity σ, that is, s = 0.5σ. When the mayfly mates and produces offspring, the attenuation of σ will synchronously reduce the search step size of the longhorn beetle whiskers. The perception range of the longhorn beetle whiskers guides the mating direction of the mayfly. The dynamic perception radius l of the longhorn beetle whiskers determines the search density of the mayfly population: when l is large, the offspring produced by the mayfly mating are distributed over a wide range, covering the global search space; when l is reduced, the offspring position gathers toward the current optimal position of the longhorn beetle whiskers, strengthening the local search.
6. The photovoltaic MPPT control method based on the mayfly beetle whisker fusion algorithm according to claim 1, characterized in that: The specific process of S4 is as follows: Determine whether the result of the fitness calculation meets the termination condition of the maximum number of iterations. If so, stop the iteration and output the corresponding voltage to stabilize the photovoltaic system at the optimal voltage value; input the optimal voltage value into the PI regulator, and its output duty cycle acts on the BOOST circuit and the longicorn population initialization range at the same time to form a closed-loop parameter update; if not, return to step S3 to continue iteration.
7. The photovoltaic MPPT control method based on the fusion algorithm of the mayfly beetle whiskers according to claim 6, characterized in that: The PI regulator adjustment formula is as follows: where e(t) = V PV (t)-V ref , V ref It is the reference voltage of the photovoltaic system.