Photovoltaic maximum power point tracking method under local shadow, related device and system

Abstract

The invention discloses a photovoltaic maximum power point tracking method, related device and system under local shadow, and the method comprises the steps: obtaining a photovoltaic MPPT power generation system model which comprises a photovoltaic array equivalent model and a Boost circuit model; obtaining P-V output characteristics of the photovoltaic array equivalent model under a non-uniform illumination condition; based on a Boost circuit model, setting an initial population number, a duty ratio and an iteration number in combination with a grey wolf algorithm so as to obtain the fitness of grey wolf individuals; the grey wolves in the grey wolf population comprise alpha, beta, gamma and omega wolves; assigning the positions of the first N fitness to alpha, beta and gamma wolves according to a descending order, and updating the gray wolf population position and coefficient vectors A and C; and under the condition that the fitness calculated by the grey wolf algorithm is the optimal fitness, adjusting the duty ratio of the Boost circuit model to obtain the maximum power output point of the photovoltaic power generation system model. By means of the method, the local optimal solution can be jumped out, and the maximum power output point of the photovoltaic power generation system model can be tracked.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photovoltaic power generation, and more particularly to a photovoltaic maximum power point tracking method under local shadow, and related devices and systems. BACKGROUND

[0002] With the increasing depletion of traditional fossil energy, it has become an inevitable trend to develop renewable energy. Solar energy is the most ideal renewable energy because it is clean, non-polluting, and has unlimited reserves. In solar power generation technology, how to reduce costs and improve power generation efficiency is a core problem. Since the output characteristics of a solar cell are affected by external conditions such as light and temperature, implementing maximum power point tracking (MPPT) is one of the key technologies to improve the overall efficiency of the system.

[0003] The control target of MPPT is to control the switch tube of the Boost circuit model to match the impedance of the equivalent model of the front-stage photovoltaic array, so that the photovoltaic MPPT power generation system always outputs power at the maximum power, thereby improving the photoelectric conversion efficiency and power generation stability of the system.

[0004] Under local shadow conditions, the output power curve of a photovoltaic power generation system presents a multi-peak state. Traditional MPPT methods such as the perturbation and observation method and the incremental conductance method are prone to being trapped in local peak points and unable to track the global maximum power. Therefore, swarm intelligence optimization algorithms with global search capability, such as particle swarm, firefly, and grey wolf optimization algorithms, have become a research hotspot for solving the MPPT problem. However, these algorithms still have problems such as low solution accuracy, slow convergence speed, and the possibility of being trapped in local extrema, which need to be improved. SUMMARY

[0005] Based on the above technical problems, the present application provides a photovoltaic maximum power point tracking method under local shadow, related devices and systems to solve the problems of low solution accuracy and easy trapping into local optimum of the traditional grey wolf algorithm.

[0006] The first aspect of the present application provides a photovoltaic maximum power point tracking method under local shadow, which can include the following steps:

[0007] Obtaining a photovoltaic MPPT power generation system model, wherein the photovoltaic MPPT power generation system model includes a photovoltaic array equivalent model and a Boost circuit model;

[0008] Obtaining the P-V output characteristics of the photovoltaic array equivalent model under non-uniform light conditions, wherein the P-V output characteristics contain multiple different local maximum power output points;

[0009] Based on the Boost circuit model, the initial population number, duty cycle and iteration number are set in combination with the grey wolf algorithm to obtain the fitness of the grey wolf individual, wherein the fitness of the grey wolf individual is used to represent the output power under the control of each duty cycle; the grey wolves in the grey wolf population include a, β, γ and ω wolves;

[0010] The positions of the first N fitness values are respectively assigned to the a, β and γ wolves in descending order, and the grey wolf population position and coefficient vector A and C are updated, wherein the vector coefficient A is related to the convergence factor a, and the convergence factor The vector coefficient C = 2 / 3(r3+a); e is the base of natural logarithm, k is the current iteration number, n is the maximum iteration number; r3 is a random vector in the interval [0, 1];

[0011] In the case that the fitness calculated by the grey wolf algorithm is the optimal fitness, the duty cycle of the Boost circuit model is adjusted to obtain the maximum power output point of the photovoltaic power generation system model.

[0012] In this way, in the process of calculating the maximum power output point of the photovoltaic power generation system model by using the grey wolf algorithm, the coefficient vectors A and C are fully considered, wherein the vector coefficient A is related to the convergence factor a, and the convergence factor The problem of low solution accuracy and easy falling into local optimum of the traditional grey wolf algorithm can be solved, and the local optimal solution is successfully jumped out and the maximum power output point of the photovoltaic power generation system model is successfully tracked.

[0013] The second aspect of the present application provides a local shadow photovoltaic maximum power point tracking device, comprising:

[0014] A first acquisition unit is configured to acquire a photovoltaic MPPT power generation system model, wherein the photovoltaic MPPT power generation system model comprises a photovoltaic array equivalent model and a Boost circuit model;

[0015] A second acquisition unit is configured to acquire P-V output characteristics of the photovoltaic array equivalent model under uneven illumination conditions, wherein the P-V output characteristics contain multiple different local maximum power output points;

[0016] A third acquisition unit is configured to acquire the fitness of the grey wolf individual based on the Boost circuit model in combination with the grey wolf algorithm setting the initial population number, duty cycle and iteration number, wherein the fitness of the grey wolf individual is used to represent the output power under the control of each duty cycle; the grey wolves in the grey wolf population include a, β, γ and ω wolves;

[0017] The processing unit is configured to assign positions of the top N fitness values to the alpha, beta and gamma wolves in descending order, respectively, and update the gray wolf population position and the coefficient vector A and C, wherein the vector coefficient A is related to a convergence factor a, and the convergence factor a is related to the maximum iteration number n and the current iteration number k. The vector coefficient C is 2 / 3(r3+a), e is the base of natural logarithm, k is the current iteration number, and n is the maximum iteration number; and r3 is a random vector in the interval [0, 1].

[0018] The fourth acquisition unit is configured to adjust the duty cycle of the Boost circuit model to obtain the maximum power output point of the photovoltaic power generation system model when the fitness value calculated by the gray wolf algorithm is the optimal fitness value.

[0019] The third aspect of the present application provides an MPPT controller, which comprises a processor and a memory, the memory is configured to store program instructions and / or data, and the processor is configured to call the program instructions stored in the memory to execute the above method.

[0020] The fourth aspect of the present application provides a photovoltaic maximum power point tracking system, which comprises the above MPPT controller.

[0021] The fifth aspect of the present application provides a computer readable storage medium, which comprises a computer program, and the computer program causes the processor to execute the above method when the computer program is run by the processor.

[0022] The sixth aspect of the present application provides a computer program product, and the computer program causes the processor to execute the above method when the computer program is run by the processor.

[0023] In some implementations, the duty cycle of the Boost circuit model is adjusted according to the following formula:

[0024]

[0025] In the formula, D α (k), D β (k), D γ (k) are learning rates of the alpha, beta and gamma wolves in the kth iteration, respectively; D α (k), D β (k), D γ (k) reflects the amplitude of the change of the duty cycle and is a variable step size coefficient of the dynamic adjustment of the duty cycle; D α (k+1), D β (k+1), D γ (k+1) are learning rates of the alpha, beta and gamma wolves in the k+1th iteration, respectively.

[0026] In some implementations, the duty cycle of the Boost circuit model is adjusted according to the following formula:

[0027]

[0028] Where D α (k), D β (k), D γ The coefficients before the (k) term are the learning rates of wolf ω on wolf α, β, and γ in the kth iteration; D α (k), D β (k), D γ (k) reflects the amplitude of duty cycle change and is the variable step coefficient for dynamic duty cycle adjustment; D α (k+1), D β (k+1), D γ The coefficients before the (k+1) term are the learning rates of the ω wolf on the α, β, and γ wolves in the k+1th iteration. In some implementations, the vector coefficient A is calculated according to the following formula:

[0029] A=2a·r1-a

[0030] Where a represents the convergence factor; r1 is a random vector in the interval [0,1].

[0031] In some implementations, the photovoltaic array equivalent model includes:

[0032]

[0033]

[0034] Where, I SC is the short-circuit current, U OC is the open circuit voltage, K I is the temperature variation coefficient of current (generally 0.0017A / ℃), S is the light intensity, and the standard light intensity S ref 1000W / m 2 , E g is the band gap energy, T is the ambient temperature (°C), q is the charge (1.6021×10 -19 C), k is the Boltzmann constant (1.38×10-23J / K), V T is the equivalent diode thermal voltage, S ref and T ref are the reference light and reference temperature, respectively.

[0035] In some implementations, the Boost circuit model includes:

[0036]

[0037] wherein U PV , I PV are the output voltage and output current of the photovoltaic cell respectively, U O is the output voltage of the Boost circuit, L is the energy storage freewheeling inductance, C1 and C2 are energy storage capacitors, I L is the current flowing through the inductance, d k is the control signal for realizing impedance matching. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of the provided drawings.

[0039] Figure 1a A photovoltaic maximum power point tracking method under partial shadow is provided in the embodiments of the present application;

[0040] Figure 1b A photovoltaic cell diode mathematical model is provided in the embodiments of the present application;

[0041] Figure 1c A Boost circuit topology diagram is provided in the embodiments of the present application;

[0042] Figure 2 A P-V output characteristic comparison diagram of a photovoltaic array under uniform illumination and partial shadow is provided in the embodiments of the present application;

[0043] Figure 3a A flowchart of an improved grey wolf algorithm is provided in the embodiments of the present application;

[0044] Figure 3b A diagram of the convergence factor of a traditional grey wolf algorithm and the convergence factor a of the improved grey wolf algorithm provided in the embodiments of the present application changing with the iteration number is provided;

[0045] Figure 4a A principle block diagram of the maximum power point tracking of the improved grey wolf algorithm provided in the embodiments of the present application is provided;

[0046] Figure 4b A photovoltaic array structure diagram is provided in the embodiments of the present application;

[0047] Figure 4c A photovoltaic array P-V curve is provided in the embodiments of the present application;

[0048] Figure 4dThe time-voltage waveform diagram of the photovoltaic array under the improved grey wolf algorithm is provided in the embodiments of the present application;

[0049] Figure 4e The time-voltage waveform diagram of the photovoltaic array under the traditional P&O algorithm is provided in the embodiments of the present application;

[0050] Figure 4f The power-voltage waveform diagram of the photovoltaic array under the improved grey wolf algorithm is provided in the embodiments of the present application;

[0051] Figure 4g The power-voltage waveform diagram of the photovoltaic array under the traditional P&O algorithm is provided in the embodiments of the present application;

[0052] Figure 5 The structure schematic diagram of the photovoltaic maximum power point tracking device under local shadow is provided in the embodiments of the present application;

[0053] Figure 6 The structure schematic diagram of the MPPT controller is provided in the embodiments of the present application. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0055] In the prior art, under the condition of local shadow, the output power curve of the photovoltaic power generation system presents a multi-peak state. The traditional MPPT methods such as the perturbation and observation method and the conductance increment method are easy to fall into local peak points and cannot track the global maximum power. Therefore, the swarm intelligence optimization algorithm with global search ability, such as the particle swarm, firefly, grey wolf optimization and the like, has become a research hotspot for solving the MPPT problem. However, the applicant found in the research that the particle swarm algorithm can realize multi-peak tracking, but has the problem of being sensitive to system parameters, and when the environmental temperature and the light intensity change sharply, the dynamic response speed is slow. Therefore, in order to realize efficient and stable photovoltaic power generation, a new type of MPPT control method capable of accurately and stably tracking the global maximum power point is urgently needed.

[0056] In view of the above problems, the improved nonlinear grey wolf algorithm for photovoltaic MPPT power generation system is provided, aiming at the maximum power point tracking problem of the photovoltaic MPPT power generation system under local shading conditions, an improved nonlinear grey wolf algorithm is provided, the convergence factor a of the grey wolf algorithm is optimized, the linear decreasing convergence factor of the traditional grey wolf algorithm is changed into the nonlinear decreasing convergence factor, so that the speed and steady precision of the maximum power point tracking are increased. The control coefficient (that is, the vector coefficient C) is redesigned, so that the grey wolf individual is more likely to approach the prey, and the global search ability of the algorithm is improved. Meanwhile, the duty cycle control of the Boost circuit is combined with the improved nonlinear grey wolf algorithm, the weight coefficient of the duty cycle of the Boost circuit is dynamically adjusted, the local search and the global search are balanced, so that the speed and the control precision of the maximum power output of the system under local shading conditions are improved.

[0057] Reference Figure 1a A photovoltaic maximum power point tracking method under local shadow provided by the embodiment of the application can include but is not limited to the following steps:

[0058] In step S101, a photovoltaic MPPT power generation system model is acquired, wherein the photovoltaic MPPT power generation system model includes a photovoltaic array equivalent model and a Boost circuit model.

[0059] The photovoltaic array equivalent model (as shown in the figure) is composed of a plurality of photovoltaic cells connected in series and in parallel, is equivalent to a photovoltaic panel in actual engineering application, converts light signals into electrical signals as an input end. The Boost circuit is used as a rear DC-DC impedance matching circuit. Figure 1b

[0060] In the embodiment of the application, the equivalent model of a single photovoltaic cell is as follows:

[0061]

[0062] In the formula, I pv is a photovoltaic panel generated photoelectric current, I d is a current flowing through the equivalent diode, R s is an equivalent series resistance, R p is an equivalent parallel resistance, I sh is a current flowing through R p , alpha is a diode ideal factor (generally 1-2), V T is an equivalent diode thermal voltage, I0 is an equivalent diode reverse saturation current, U and I are output voltage and output current of the photovoltaic cell model.

[0063] Further, the simplified photovoltaic cell model and the determined part parameters are as follows:

[0064] ​

[0065] wherein I SC is the short-circuit current, U OC is the open-circuit voltage, K I is the current temperature coefficient (generally taken as 0.0017 A / ℃), S is the light intensity, and the standard light intensity S ref is 1000 W / m 2 , E g is the band gap energy, T is the ambient temperature (℃), q is the charge quantity (1.6021 x 10 -19 C), k is the Boltzmann constant (1.38 x 10 -23 J / K), V T is the equivalent diode thermal voltage, S ref and T ref are the reference light and the reference temperature, respectively.

[0066] In the embodiments of the present application, the Boost circuit model (as shown in Figure 1c ) comprises:

[0067]

[0068] wherein U PV , I PV are the output voltage and the output current of the photovoltaic cell, respectively, U O is the output voltage of the Boost circuit, L is the energy storage freewheeling inductance, C1 and C2 are both energy storage capacitors, I L is the current flowing through the inductance, d k is the control signal used to achieve impedance matching.

[0069] In step S102, the P-V output characteristic of the equivalent model of the photovoltaic array under non-uniform light conditions is obtained, wherein the P-V output characteristic contains multiple different local maximum power output points.

[0070] In the embodiments of the present application, impedance matching with the photovoltaic array is achieved by controlling the duty cycle of the Boost circuit, so as to calculate the maximum power output of the photovoltaic MPPT power generation system. The output power of the photovoltaic MPPT power generation system is:

[0071]

[0072] wherein R is the equivalent internal resistance of the photovoltaic array, and R L is the actual impedance of the Boost circuit.

[0073] The derivative of the output power of the photovoltaic MPPPT power generation system is:

[0074]

[0075] According to the maximum power transmission law, let dP L / dR L = 0, only when R = R L , the output power is maximum.

[0076] By adjusting the duty cycle to change the equivalent impedance of the Boost circuit, as follows:

[0077] R' = R L (1-d) 2

[0078] In the formula, R' is the equivalent impedance of the Boost circuit, d is the duty cycle of the switching device in the Boost circuit, and R L is the actual impedance of the Boost circuit.

[0079] The P-V output characteristics of the equivalent model of the photovoltaic array under the local shading condition present a multi-peak state (such as Figure 2 ), that is, the P-V curve has multiple peak values, and the traditional MPPT algorithm such as the perturbation and observation method and the conductance increment method is easily trapped in a local maximum value, and cannot accurately track the global maximum value, so that the photovoltaic MPPT power generation system cannot guarantee maximum power output.

[0080] In step S103, based on the Boost circuit model, the initial population number, the duty cycle and the iteration number are set in combination with the grey wolf algorithm to obtain the fitness of the grey wolf individual, wherein the fitness of the grey wolf individual is used to represent the output power under each duty cycle control; the grey wolves in the grey wolf population include alpha, beta, gamma and omega wolves.

[0081] The grey wolf algorithm simulates the leadership structure and hunting mechanism of the grey wolf group in the natural world. The model assumes that the grey wolf group is at the top of the food chain and is gregarious. The grey wolves in the wolf group are divided into four categories: alpha, beta, gamma and omega, to simulate the leadership structure of the wolf group. The alpha wolf is called the alpha wolf group, and the second level of the grey wolf hierarchy is the beta wolf group. The beta wolf group is a subordinate wolf group that helps the alpha wolf group make decisions to carry out group activities. The lowest-ranking grey wolf is the omega wolf group. The omega wolf group plays a replacement role. If the grey wolf is neither an alpha wolf, a beta wolf, nor an omega wolf, it is called a gamma wolf. The gamma wolf must obey the alpha wolf and the beta wolf. Young wolves, sentries, old wolves, predators and caretakers belong to this category. Specifically, the implementation process of the grey wolf algorithm proposed in the present application is as shown in Figure 3a .

[0082] In step S104, the positions of the first N fitness values are respectively assigned to the alpha, beta and gamma wolves in descending order, and the grey wolf population position and the coefficient vector A and C are updated, wherein the vector coefficient A is related to the convergence factor a, and the convergence factor Vector coefficient C = 2 / 3(r3+a); e is the base of natural logarithm, k is the current iteration number, n is the maximum iteration number; r3 is a random vector in the interval [0, 1].

[0083] The gray wolf hunting process can be divided into three processes: tracking prey, surrounding prey, and attacking prey. The gray wolf algorithm is used to determine the global optimal solution, and the most suitable solution is designated as alpha, the second and third best solutions are named beta and gamma, and the remaining candidate solutions are designated as omega. The hunting process, i.e., optimization, is guided by alpha, beta, and gamma, and omega follows the three wolves. In order to mathematically simulate the process of tracking to surrounding of gray wolves, the model is as follows:

[0084] D = |C·X p (t)-X p (t)|

[0085] X(t+1) = X p (t)-A·D

[0086] In the formula, t is the iteration number, D is the distance between the gray wolf and the prey, and since there are three gray wolves, D is in vector form. D, A, and C are all coefficient vectors, and C is also called the control coefficient. X p (t) represents the position vector of the prey, and X(t+1) is the position vector of the gray wolf.

[0087] Where, the vectors A and C can be calculated by the following formula:

[0088] A = 2a·r1-a

[0089] C = 2·r2

[0090] In the formula, a is called the convergence factor; r1 and r2 are random vectors in the interval [0, 1]. r1 is used to calculate the moving step length of the wolf, and its randomness makes the moving step length of each wolf different in each iteration, increasing the diversity of the search. Through the randomness of r1, the wolf pack can search in a large range and avoid converging to a local optimal solution too early; r2 is used to calculate the position update vector of the wolf, and its randomness makes the direction of each wolf learning from the alpha (alpha, beta, gamma) different in each iteration, increasing the randomness of the search. Through the randomness of r2, the wolf pack can learn from the alpha from different angles, improving the exploration ability of the algorithm.

[0091] To approach the prey in the mathematical simulation, the value of a must be reduced. At the same time, the fluctuation range of A is also reduced with the reduction of a, in other words, A is a random value in the interval [-a, a]. When the random value of A is in [-1, 1], the next position of the searcher can be anywhere between its current position and the position of the prey, and when |A| < 1, the wolf pack is forced to approach the prey. Gray wolves mainly search according to the position of the dominant wolf, and they separate from each other to find prey and aggregate to attack the prey. In order to establish a mathematical analysis model, A with a random value greater than 1 or less than -1 is used to force the searcher to separate from the prey, so that the gray wolf algorithm searches globally. When |A| ≥ 1 forces the gray wolf to deviate from the prey, it is hoped to find a more suitable prey.

[0092] The convergence factor a of the gray wolf algorithm is linearly reduced from 2 to 0 in the iteration process, but the algorithm is not linear in the iteration process. Therefore, the linear decreasing strategy of the convergence factor cannot fully reflect the superiority of the algorithm, and therefore a non-linear decreasing method is applied to the convergence factor of the improved non-linear gray wolf algorithm. When the algorithm gradually approaches the global optimal solution, the convergence factor is close to 2, at this time, increasing the iteration step length of the convergence factor can increase the fluctuation range of A, thereby speeding up the speed of the gray wolf tracking to the prey, that is, reaching the global optimal point. When the algorithm is near the global optimal solution, the iteration step length of the convergence factor is reduced, which can reduce the fluctuation range of A, avoid missing the global optimal solution, and thus improve the accuracy of the gray wolf tracking to the prey, that is, tracking the global optimal point. Moreover, the convergence factor presents non-linearity, which can ensure that in the P-V output multi-peak curve of the equivalent model of the photovoltaic array, it is avoided to fall into a local optimum, thereby enhancing the ability to find a global optimum and improving the convergence speed.

[0093] Based on this, in the embodiments of the present application, an improved convergence factor is proposed, which is specifically as follows:

[0094]

[0095] In the formula, e is the base of the natural logarithm, k is the current iteration number, and n is the maximum iteration number. a is the convergence factor of the gray wolf algorithm, which is essentially a dynamic behavior change simulating the hunting process of the wolf pack from a large-scale search to a precise encirclement, and is a dynamic regulator of the hunting behavior of the wolf pack. Through the control of the search step length, the balance of global and local search, and the strengthening of group cooperation, a smooth transition from large-scale exploration to precise encirclement is finally realized. The iteration number k and the maximum iteration number n are important parameters for controlling the search process and termination condition of the algorithm, which are used to limit the time and steps of the wolf pack in the hunting process, and ensure that the algorithm can efficiently find the optimal solution within a limited iteration.

[0096] The convergence factor a is a dynamic behavior change simulating the hunting process of the wolf pack from a large-scale search to a precise encirclement, and is a dynamic regulator of the hunting behavior of the wolf pack. Through the control of the search step length, the balance of global and local search, and the strengthening of group cooperation, a smooth transition from large-scale exploration to precise encirclement is finally realized. The iteration number k and the maximum iteration number n are important parameters for controlling the search process and termination condition of the algorithm, which are used to limit the time and steps of the wolf pack in the hunting process, and ensure that the algorithm can efficiently find the optimal solution within a limited iteration. It is explained that when k = 0, a = 2; when k = n, that is, the maximum number of iterations, a = 0. It can be known that a satisfies a nonlinear change from the beginning to the end of the grey wolf algorithm. Specifically, the convergence factor of the traditional grey wolf algorithm and the improved convergence factor a of the present application change with the number of iterations as shown in the graph. Figure 3b

[0097] Applicants found in research that the design of the control coefficient C of the traditional grey wolf algorithm would limit the global search ability of MPPT, and there would be a situation that the global maximum power point could not be tracked. Therefore, the control coefficient C is improved, which directly affects the difficulty of the grey wolf individual approaching the prey, and then affects the global search ability of the algorithm. In the embodiments of the present application, an improved vector coefficient C is proposed:

[0098] C = 2 / 3(r3 + a)

[0099] In the formula, r3 is a random vector in the interval [0, 1], which represents the strength of environmental random disturbance, r3 ≈ 0 simulates a favorable environment (such as tailwind, clear vision) that the wolf pack trusts the judgment of the leader wolf (tends to development); r3 ≈ 1 simulates an unfavorable environment (such as headwind, visual obstruction) that the wolf pack autonomously expands the search (tends to exploration). r3-a constitutes a coupling factor of environmental disturbance and hunting stage, increases randomness, and realizes dynamic decision-making.

[0100] Generally, the control coefficient C is a random vector in [0, 2], which provides a random weight to the prey, which helps to improve the randomness of the grey wolf algorithm in the entire optimization process, avoids the grey wolf algorithm falling into local optimum, and makes the global search more comprehensive. When the control coefficient C is large, the difficulty of the grey wolf approaching the prey increases; when the control coefficient C is small, the speed of the grey wolf approaching the prey can be accelerated. Therefore, C is designed to be a combination of r3 and a, which can consider both random variables at the same time, increase the randomness of the grey wolf algorithm search, specifically, C = 2 / 3(r3 + a), in the formula, a is a quantity in the range of 0-2, r3 is a random quantity in the range of 0-1, so C = 2 / 3(r3 + a) is a random quantity in the range of 0-2, which is consistent with the value range of C = 2r2 in the original grey wolf algorithm, where r2 is a random quantity in the range of 0-1.

[0101] Step S105, in the case that the fitness calculated by the grey wolf algorithm is the optimal fitness, the duty cycle of the Boost circuit model is adjusted to obtain the maximum power output point of the photovoltaic power generation system model.

[0102] ​When the algorithm searches for the global maximum power point, the vector D will automatically approach to 0. Therefore, it is necessary to combine the duty ratio control on the basis of this algorithm, that is, when at the maximum power point, the output duty ratio remains constant to reduce the output harmonic content of the steady-state system, and reduce the power loss caused by system oscillation, based on the mathematical model of the simulated gray wolf tracking to the surrounding, the improved nonlinear gray wolf algorithm is applied to the control of the duty ratio d:

[0103] D(k+1)=D(k)-A·D

[0104] In the formula, k is the iteration number, D(k)=[D α (k),D β (k),D γ (k)] and D(k+1)=[D α (k+1),D β (k+1),D γ (k+1)].

[0105] In the embodiment of the application, the duty ratio of the Boost circuit model can be adjusted through the traditional Boost circuit model duty ratio calculation formula, and the formula is as follows:

[0106]

[0107] However, the applicant found in the research that the traditional Boost circuit model duty ratio calculation formula is used to adjust the duty ratio of the Boost circuit model, and this implementation cannot well balance the local search and the global search. Based on this, the applicant dynamically adjusts the weight of the duty ratio, and proposes an improved Boost circuit model duty ratio calculation formula, and the formula is as follows:

[0108]

[0109]

[0110] In the formula, D α (k), D β (k), D γ (k) are respectively the learning rates of the alpha, beta and gamma wolves in the kth iteration; D α (k), D β (k), D γ (k) reflect the amplitude of the change of the duty ratio, and are variable step coefficients of the dynamic adjustment of the duty ratio; D α (k+1), D β (k+1), D γ (k+1) are respectively the learning rates of the alpha, beta and gamma wolves in the k+1th iteration.

[0111] Specifically, the duty cycle is an equivalent load operating point regulator of a photovoltaic MPPT power generation system, and the impedance matching between the photovoltaic panel and the load is achieved by dynamically adjusting the equivalent load impedance by changing the duty cycle, so that the solar panel works at the maximum power point. The value of the duty cycle and D β , D γ The essence of the decision is to simulate the wolf pack to form a group optimal action plan by synthesizing the three-level leadership information through democratic centralism, and the weights of the three-level leadership are adjusted dynamically. Figure 4a The principle block diagram of the maximum power point tracking based on the improved gray wolf algorithm is shown.

[0112] In the embodiments of the present application, the applicable condition of the improved nonlinear gray wolf algorithm is:

[0113] P(d(k+1))>P(d(k))

[0114] In the formula, P is the output power of the photovoltaic MPPT power generation system. For example, in the attached Figure 1b , P max is the actual maximum power output point, and G max is the maximum power output point obtained by the improved gray wolf algorithm.

[0115] In this way, in the process of calculating the maximum power output point of the photovoltaic power generation system model by using the gray wolf algorithm, the coefficient vectors A and C are fully considered, wherein the vector coefficient A is related to the convergence factor a, and the convergence factor can solve the problems of low solution accuracy and easy to fall into local optimum of the traditional gray wolf algorithm, and successfully jump out of the local optimal solution and track the maximum power output point of the photovoltaic power generation system model.

[0116] In order to better understand the technical solutions described in the present application, the following will be described in conjunction with specific examples to illustrate how the present application solves the problem of low solution accuracy and easy to fall into local optimum of the traditional gray wolf algorithm. As Figure 4a The principle block diagram of the maximum power point tracking based on the improved gray wolf algorithm is shown, wherein the leftmost side is a photovoltaic array, C1=10μF, L=1.1478mH, C2=470μF, and the rightmost side is a load end.

[0117] Further, the photovoltaic array is shown in Figure 4b , and four photovoltaic panels are connected in series. The local shading is simulated, and the light intensities of the four panels are 500w / m 2 , 800w / m 2 , 1000w / m 2 , and 1000w / m 2 , and the working temperature is 25℃.

[0118] The P-V curve of this photovoltaic array is shown inFigure 4c The maximum output power of the photovoltaic array is 636W, and there are two local maximum output power points, one of which is 566W.

[0119] In other experimental conditions, the control variable is improved gray wolf algorithm and traditional P&O algorithm, Figure 4d The time-voltage waveform diagram of the output of the photovoltaic array under the improved gray wolf algorithm is shown in the figure. Figure 4e The time-voltage waveform diagram of the output of the photovoltaic array under the traditional P&O algorithm is shown in the figure.

[0120] Further, Figure 4f The power-voltage waveform diagram of the output of the photovoltaic array under the improved gray wolf algorithm is shown in the figure. Figure 4g The power-voltage waveform diagram of the output of the photovoltaic array under the traditional P&O algorithm is shown in the figure.

[0121] According to the T-V curve and P-V curve of the two algorithms, the traditional algorithm falls into Figure 4c the local optimal power point (133, 566), while the improved gray wolf algorithm does not fall into the local optimal power point (133, 566) after finding it, but continues to explore until the global maximum power point (92, 636) is found. After finding the maximum power point, the curve of the improved gray wolf algorithm converges better than the P&O algorithm, which can further reduce the difficulty of subsequent processing.

[0122] It should be noted that for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the disclosure is not limited by the order of the described actions, because according to the disclosure, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily necessary for the disclosure.

[0123] It should be further noted that although Figure 1a each step in the flowchart is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless explicitly stated in this document, the execution of these steps has no strict order limitation, and these steps can be executed in other order. Moreover, Figure 1a At least part of the steps in may include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or other steps. Sub-steps or stages.

[0124] In order to better implement the above method of the embodiment of the present application, the embodiment of the present application also describes the above method. Figure 1a The method embodiment described is a schematic diagram of a photovoltaic maximum power point tracking device under partial shadow under the same inventive concept. The following is a detailed description with reference to the accompanying drawings:

[0125] like Figure 5 As shown, the photovoltaic maximum power point tracking device 50 under partial shadow may include:

[0126] A first acquisition unit 500 is configured to acquire a photovoltaic MPPT power generation system model, wherein the photovoltaic MPPT power generation system model includes a photovoltaic array equivalent model and a Boost circuit model;

[0127] A second acquiring unit 501 is configured to acquire PV output characteristics of the photovoltaic array equivalent model under non-uniform illumination conditions, wherein the PV output characteristics include a plurality of different local maximum power output points;

[0128] A third acquisition unit 502 is configured to set the initial population size, duty cycle, and number of iterations based on the Boost circuit model and the gray wolf algorithm to obtain the fitness of individual gray wolves, wherein the fitness of the individual gray wolf is used to represent the output power under various duty cycle controls; the gray wolves in the gray wolf population include α, β, γ, and ω wolves;

[0129] The processing unit 503 is used to assign the positions of the first N fitness values ​​to the α, β, and γ wolves in descending order, and update the position of the gray wolf population and the coefficient vectors A and C, wherein the vector coefficient A is related to the convergence factor a, and the convergence factor Vector coefficient C = 2 / 3 (r3 + a); e is the base of the natural logarithm, k is the current number of iterations, n is the maximum number of iterations; r3 is a random vector in the interval [0, 1];

[0130] The fourth obtaining unit 504 is configured to adjust the duty cycle of the Boost circuit model when the fitness calculated by the grey wolf algorithm is the optimal fitness, so as to obtain the maximum power output point of the photovoltaic power generation system model.

[0131] In a possible implementation, the fourth obtaining unit 504 is specifically configured to:

[0132] The duty cycle of the Boost circuit model is adjusted according to the following formula:

[0133]

[0134] Where D α (k), D β (k), Dγ The coefficients before the terms (k) are learning rates of the wolf omega to the three wolves alpha, beta and gamma in the kth iteration respectively. α (k), D β (k), D γ (k) reflects the amplitude of the change of the duty cycle, and is a variable step size coefficient of the dynamic adjustment of the duty cycle. α (k+1), D β (k+1), D γ The coefficients before the terms (k+1) are learning rates of the wolf omega to the three wolves alpha, beta and gamma in the k+1th iteration respectively.

[0135] In a possible implementation, the fourth acquisition unit 504 is specifically configured to:

[0136] The duty cycle of the Boost circuit model is adjusted according to the following formula:

[0137]

[0138] In the formula, D α (k), D β (k), D γ The coefficients before the terms (k) are learning rates of the wolf omega to the three wolves alpha, beta and gamma in the kth iteration respectively. α (k), D β (k), D γ (k) reflects the amplitude of the change of the duty cycle, and is a variable step size coefficient of the dynamic adjustment of the duty cycle. α (k+1), D β (k+1), D γ The coefficients before the terms (k+1) are learning rates of the wolf omega to the three wolves alpha, beta and gamma in the k+1th iteration respectively.

[0139] In a possible implementation, the processing unit 503 is specifically configured to:

[0140] The vector coefficient A is calculated according to the following formula:

[0141] A = 2a·r1-a

[0142] In the formula, a represents a convergence factor; and r1 is a random vector in the interval [0, 1].

[0143] In a possible implementation, the photovoltaic array equivalent model comprises:

[0144]

[0145] In the formula, I SC is a short-circuit current, U OC is an open-circuit voltage, and K Iis the current temperature coefficient (generally taken as 0.0017A / ℃), S is the light intensity, and S ref is 1000W / m 2 g is the band gap energy, T is the ambient temperature (℃), q is the charge quantity (1.6021x10 -19 C), k is the Boltzmann constant (1.38x10-23J / K), V T is the equivalent diode thermal voltage, S ref and T ref are the reference light and the reference temperature, respectively.

[0146] In a possible implementation, the Boost circuit model comprises:

[0147]

[0148] wherein U PV , I PV are the output voltage and the output current of the photovoltaic cell, respectively, U O is the output voltage of the Boost circuit, L is the energy storage freewheeling inductance, C1 and C2 are both energy storage capacitors, I L is the current flowing through the inductance, d k is the control signal to achieve impedance matching.

[0149] It can be understood that the functions of the functional units of the partial-shading photovoltaic maximum power point tracking device 50 in the embodiment can be implemented according to the method in the method embodiment described above, and the specific implementation process can refer to the related description of the method embodiment described above, which will not be described here. Figure 1a

[0150] In order to better implement the above-mentioned scheme of the embodiment of the present application, the present application also provides an MPPT controller, which will be described in detail below with reference to the accompanying drawings:

[0151] As Figure 6 ​​The structure schematic diagram of the MPPT controller provided by the embodiment of the application is shown. The MPPT controller 60 can include a processor 601, a memory 604 and a communication module 605, which can be connected with each other through a bus 606. The memory 604 can be a high-speed random access memory (RAM) memory, or a non-volatile memory such as at least one disk memory. The memory 604 can also be at least one storage system located away from the aforementioned processor 601. The memory 604 is used for storing application program codes, which can include an operating system, a network communication module, a user interface module and a data processing program. The communication module 605 is used for information interaction with external devices. The processor 601 is configured to call the program codes and perform the following steps:

[0152] Obtaining a photovoltaic MPPT power generation system model, wherein the photovoltaic MPPT power generation system model includes a photovoltaic array equivalent model and a Boost circuit model;

[0153] Obtaining a P-V output characteristic of the photovoltaic array equivalent model under a non-uniform illumination condition, wherein the P-V output characteristic contains a plurality of different local maximum power output points;

[0154] Based on the Boost circuit model, initializing the population number, the duty ratio and the iteration number by using the grey wolf algorithm to obtain the fitness of a grey wolf individual, wherein the fitness of the grey wolf individual is used to represent the output power under each duty ratio control. The grey wolves in the grey wolf population include alpha wolves, beta wolves, gamma wolves and omega wolves.

[0155] Assigning the positions of the first N fitnesses to the alpha wolves, the beta wolves and the gamma wolves in descending order, respectively, and updating the position and coefficient vectors A and C of the grey wolf population, wherein the vector coefficient A is related to a convergence factor a, and the convergence factor a is related to the current iteration number k and the maximum iteration number n. The vector coefficient C = 2 / 3(r3+a); e is the base of natural logarithm, k is the current iteration number, n is the maximum iteration number; r3 is a random vector in the interval [0, 1];

[0156] In the case that the fitness calculated by the grey wolf algorithm is the optimal fitness, adjusting the duty ratio of the Boost circuit model to obtain the maximum power output point of the photovoltaic power generation system model.

[0157] The processor 601 adjusts the duty ratio of the Boost circuit model according to the following formula:

[0158]

[0159]

[0160] wherein D α (k), D β (k), D γ (k) are learning rates of the wolves ωk to the three wolves α, β, γ in the kth iteration, respectively, and D α (k), D β (k), D γ (k) reflects the amplitude of the change of the duty cycle and is a variable step size coefficient of the dynamic adjustment of the duty cycle; D α (k+1), D β (k+1), D γ (k+1) are learning rates of the wolves ωk to the three wolves α, β, γ in the k+1th iteration, respectively.

[0161] The processor 601 adjusts the duty cycle of the Boost circuit model according to the following formula:

[0162]

[0163] wherein D α (k), D β (k), D γ (k) are learning rates of the wolves ωk to the three wolves α, β, γ in the kth iteration, respectively, and D α (k), D β (k), D γ (k) reflects the amplitude of the change of the duty cycle and is a variable step size coefficient of the dynamic adjustment of the duty cycle; D α (k+1), D β (k+1), D γ (k+1) are learning rates of the wolves ωk to the three wolves α, β, γ in the k+1th iteration, respectively.

[0164] The processor 601 calculates the vector coefficient A according to the following formula:

[0165] A = 2a·r1-a

[0166] wherein a represents a convergence factor; and r1 is a random vector in the interval [0, 1].

[0167] The photovoltaic array equivalent model comprises:

[0168]

[0169] wherein I SC is a short-circuit current, U OC is an open-circuit voltage, and K Iis the temperature variation coefficient of current (generally 0.0017A / ℃), S is the light intensity, and the standard light intensity S ref 1000W / m 2 , E g is the band gap energy, T is the ambient temperature (°C), q is the charge (1.6021×10 -19 C), k is the Boltzmann constant (1.38×10-23J / K), V T is the equivalent diode thermal voltage, S ref and T ref are the reference light and reference temperature, respectively.

[0170] Wherein, the Boost circuit model includes:

[0171]

[0172] Where U PV , I PV are the output voltage and output current of the photovoltaic cell, U O is the output voltage of the Boost circuit, L is the energy storage freewheeling inductor, C1 and C2 are both energy storage capacitors, I L is the current flowing through the inductor, d k It is a control signal used to achieve impedance matching.

[0173] It should be noted that the execution steps of the processor in the MPPT controller 60 in the embodiment of the present application can refer to the above-mentioned method embodiments. Figure 1a The specific implementation method in the embodiment will not be described here in detail.

[0174] An embodiment of the present application also provides a photovoltaic maximum power point tracking system, including the above-mentioned MPPT controller.

[0175] The present application also provides another computer storage medium for storing the above Figure 1a The computer software instructions used by the server include a program designed to execute the above method embodiment. By executing the stored program, photovoltaic maximum power point tracking under partial shadow can be completed.

[0176] The present application also provides a computer storage medium for storing the above Figure 1a The computer software instructions used by the terminal shown include a program designed to execute the above method embodiment. By executing the stored program, photovoltaic maximum power point tracking under partial shadow can be completed.

[0177] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, various elements are implemented in hardware, firmware, or software, or combinations thereof. In embodiments implemented in software, the functions can be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media include both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media can be any available media that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired computer program code means in the form of instructions or data structures and that can be accessed by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or other

[0178] The present application is described in reference to the drawings, which are as follows:

[0179] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0180] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0181] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A photovoltaic maximum power point tracking method under partial shadow, characterized in that: The method comprises: Obtaining a photovoltaic MPPT power generation system model, wherein the photovoltaic MPPT power generation system model includes a photovoltaic array equivalent model and a Boost circuit model; Obtaining PV output characteristics of the photovoltaic array equivalent model under uneven illumination conditions, wherein the PV output characteristics include a plurality of different local maximum power output points; Based on the Boost circuit model, the initialization population size, duty cycle and number of iterations are set in combination with the gray wolf algorithm to obtain the fitness of the gray wolf individuals, wherein the fitness of the gray wolf individuals is used to characterize the output power under the control of each duty cycle; the gray wolves in the gray wolf population include α, β, γ, and ω wolves; Assign the first N fitness positions to the α, β, and γ wolves in descending order, and update the gray wolf population position and coefficient vectors A and C. The vector coefficient A is related to the convergence factor a. Vector coefficient C = 2 / 3 (r3 + a); e is the base of the natural logarithm, k is the current number of iterations, n is the maximum number of iterations; r3 is a random vector in the interval [0, 1]; When the fitness calculated by the grey wolf algorithm is the optimal fitness, the duty cycle of the Boost circuit model is adjusted to obtain the maximum power output point of the photovoltaic power generation system model.

2. The method according to claim 1, characterized in that The duty cycle of the Boost circuit model is adjusted according to the following formula: Where D α (k), D β (k), D γ The coefficients before the (k) term are the learning rates of the wolf ω on the three wolves α, β, and γ in the kth iteration, respectively. α (k), D β (k), D γ (k) reflects the amplitude of duty cycle change and is the variable step coefficient for dynamic duty cycle adjustment; D α (k+1), D β (k+1), D γ The coefficients before the (k+1) term are the learning rates of wolf ω on wolves α, β, and γ in the k+1th iteration.

3. The method according to claim 1, characterized in that The duty cycle of the Boost circuit model is adjusted according to the following formula: Where D α (k), D β (k), D γ The coefficients before the (k) term are the learning rates of wolf ω on wolf α, β, and γ in the kth iteration; D α (k), D β (k), D γ (k) reflects the amplitude of duty cycle change and is the variable step coefficient for dynamic duty cycle adjustment; D α (k+1), D β (k+1), D γ The coefficients before the (k+1) term are the learning rates of wolf ω on wolves α, β, and γ in the k+1th iteration.

4. The method according to claim 1, wherein The vector coefficient A is calculated according to the following formula: A=2a·r1-a Where a represents the convergence factor; r1 is a random vector in the interval [0,1].

5. The method according to any one of claims 1 to 4, characterized in that The photovoltaic array equivalent model includes: Where, I SC is the short-circuit current, U OC is the open circuit voltage, K I is the temperature variation coefficient of current (generally 0.0017A / ℃), S is the light intensity, and the standard light intensity S ref 1000W / m 2 , E g is the band gap energy, T is the ambient temperature (°C), q is the charge (1.6021×10 -19 C), k is the Boltzmann constant (1.38×10-23J / K), V T is the equivalent diode thermal voltage, S ref and T ref are the reference light and reference temperature, respectively.

6. The method according to any one of claims 1 to 4, characterized in that The Boost circuit model includes: Where U PV , I PV are the output voltage and output current of the photovoltaic cell, U O is the output voltage of the Boost circuit, L is the energy storage freewheeling inductor, C1 and C2 are both energy storage capacitors, I L is the current flowing through the inductor, d k It is a control signal used to achieve impedance matching.

7. A photovoltaic maximum power point tracking device under partial shadow, characterized in that: include: A first acquisition unit is configured to acquire a photovoltaic MPPT power generation system model, wherein the photovoltaic MPPT power generation system model includes a photovoltaic array equivalent model and a Boost circuit model; a second acquiring unit, configured to acquire a PV output characteristic of the photovoltaic array equivalent model under uneven illumination conditions, wherein the PV output characteristic includes a plurality of different local maximum power output points; A third acquisition unit is configured to set the initialization population size, duty cycle, and number of iterations based on the Boost circuit model and in combination with the gray wolf algorithm to obtain the fitness of individual gray wolves, wherein the fitness of the individual gray wolf is used to characterize the output power under the control of each duty cycle; the gray wolves in the gray wolf population include α, β, γ, and ω wolves; The processing unit is used to assign the positions of the first N fitness values ​​to the α, β, and γ wolves in descending order, and update the position of the gray wolf population and the coefficient vectors A and C, wherein the vector coefficient A is related to the convergence factor a, and the convergence factor Vector coefficient C = 2 / 3 (r3 + a); e is the base of the natural logarithm, k is the current number of iterations, n is the maximum number of iterations; r3 is a random vector in the interval [0, 1]; The fourth acquisition unit is used to adjust the duty cycle of the Boost circuit model when the fitness calculated by the gray wolf algorithm is the optimal fitness, so as to obtain the maximum power output point of the photovoltaic power generation system model.

8. An MPPT controller, characterized in that: The MPPT controller includes: a processor and a memory, the memory is used to store program instructions and / or data, and the processor is used to call the program instructions stored in the memory to execute the steps of the photovoltaic maximum power point tracking method under partial shadow according to any one of claims 1 to 6.

9. A photovoltaic maximum power point tracking system, characterized in that: Including the MPPT controller described in claim 8.

10. A computer-readable storage medium, characterized in that The method comprises a computer program, which, when executed by a processor, causes the processor to perform the method according to any one of claims 1 to 6.

Images