阅读说明：本技术 一种局部阴影条件下最大功率点跟踪方法 (Maximum power point tracking method under local shadow condition ) 是由 梁智超 李梦达 郑旭彬 姚林萍 于 2021-06-29 设计创作，主要内容包括：本发明涉及一种局部阴影条件下最大功率点跟踪方法,包括：1)结合局部阴影下的光伏阵列电流-电压特性,构建光伏电池模型、升压电路模型和基于和声搜索算法的控制器模型；2)初始化自适应和声搜索算法的参数,并初始化和声记忆库,将寻优的初始电压值作为和声,生成新和声向量,进而获取更新电压；3)对和声记忆库中的和声向量进行更新；4)判断是否达到算法终止标准,若是,则输出最优电压,追踪到全局最大功率点,否则,继续生成新和声向量并更新和声记忆库,直至达到算法终止标准。与现有技术相比,本发明具有提高跟踪速度和精度,提升太阳能利用率等优点。(The invention relates to a maximum power point tracking method under a local shadow condition, which comprises the following steps: 1) establishing a photovoltaic cell model, a booster circuit model and a controller model based on a harmony search algorithm by combining the current-voltage characteristics of the photovoltaic array under the local shadow; 2) initializing parameters of a self-adaptive harmony search algorithm, initializing a harmony memory library, taking the optimized initial voltage value as harmony, generating a new harmony vector, and further acquiring an updated voltage; 3) updating the harmony vector in the harmony memory bank; 4) and judging whether the algorithm termination standard is met, if so, outputting the optimal voltage and tracking to the global maximum power point, otherwise, continuously generating a new harmony vector and updating the harmony memory base until the algorithm termination standard is met. Compared with the prior art, the invention has the advantages of improving the tracking speed and precision, improving the solar energy utilization rate and the like.)

1. A maximum power point tracking method under the condition of local shadow is characterized by comprising the following steps:

1) establishing a photovoltaic cell model, a booster circuit model and a controller model based on a harmony search algorithm by combining the current-voltage characteristics of the photovoltaic array under the local shadow;

2) initializing parameters of a self-adaptive harmony search algorithm, initializing a harmony memory library, taking the optimized initial voltage value as harmony, generating a new harmony vector, and further acquiring an updated voltage;

3) updating the harmony vector in the harmony memory bank;

4) and judging whether the algorithm termination standard is met, if so, outputting the optimal voltage and tracking to the global maximum power point, otherwise, continuously generating a new harmony vector and updating the harmony memory base until the algorithm termination standard is met.

2. The maximum power point tracking method under partial shadow condition of claim 1, wherein in step 2), the parameters of the adaptive harmony search algorithm include harmony memory bank size HMS, harmony memory bank consideration probability HMCR, pitch fine tuning probability PAR, pitch fine tuning amplitude BW, and maximum iteration number NI.

3. The maximum power point tracking method under the partial shadow condition according to claim 2, wherein the harmony memory bank is composed of HMS randomly generated harmony vectors, and the expression of the initialized harmony memory bank is as follows:

where HM is harmonyMemory bank, each sum sound { x¹，x²，L，x^HMSThe optimized initial voltage value is obtained; f (x) is a fitness function, and the target fitness function is f (x)ⁱ)＝P(U_pv)＝U_pv*I_pv，P(U_pv) For the current power, U_pvFor the present input voltage, I_pvIs the present input current.

4. The method of claim 2, wherein the new harmony vector is generated by ad hoc authoring based on a harmony memory base considering probability HMCR, pitch fine tuning probability PAR, and pitch fine tuning amplitude BW.

5. The maximum power point tracking method under the partial shadow condition according to claim 4, wherein the specific content for generating the new sum sound vector is as follows:

firstly, for the current input voltage U_pvAnd adjusting, wherein the updating formula is as follows:

in the formula, x^newTo update the voltage values stored in the harmonic waves; x is the number of^HMSFor the second HMS harmonic vector in HM, U_ocFor the open circuit voltage of the photovoltaic cell, r₁And rand (1) is in [0,1 ]]A random number generated in between;

then obtaining the tone fine tuning probability PAR and the tone fine tuning amplitude BW which are dynamically changed;

current input voltage U based on regulation_pvObtaining a harmony variable by the dynamically changed tone fine-tuning probability PAR and the tone fine-tuning amplitude BW;

and judging whether the obtained harmony variable needs to be adjusted in the next step, if so, acquiring new harmony, and otherwise, keeping the original harmony variable.

6. The maximum power point tracking method under partial shadow condition according to claim 5,

the dynamically varying expression for the pitch trimming probability PAR is:

wherein PAR (t) denotes the change in the probability PAR with non-linear decrease of the iterative pitch trimming, PAR_maxAnd PAR_minRespectively a maximum value and a minimum value of the tone fine-tuning probability, and t is the current iteration number;

the dynamically varying expression for the pitch trim amplitude BW is:

BW(t)＝(BW_max-BW_min)e^-t+BW_min

where BW (t) is the change in BW with a non-linear decrease in the iterative pitch trimming amplitude BW_maxAnd BW_minThe maximum amplitude of pitch trimming and the minimum amplitude of pitch trimming are respectively, and t is the current iteration number.

7. The maximum power point tracking method under the partial shadow condition according to claim 6, wherein if the obtained harmony variable is obtained from a harmony library, fine tuning is performed on the harmony variable to generate a new harmony variable; otherwise, no adjustment is made.

8. The maximum power point tracking method under the local shadow condition according to claim 7, wherein the expression of the new harmonic variable generated after the fine tuning is as follows:

in the formula, r₂Is [0,1 ]]A random number is generated.

9. The maximum power point tracking method under the partial shadow condition according to claim 7, wherein the specific contents of updating the harmony memory bank are as follows:

determining the worst sum sound x stored in the sum sound memory HM^worstFor new harmonic variables x^newEvaluating the fitness function if the fitness f (x)^new)＞f(x^worst) Then use the new harmony variable x^newReplacing worst sum sound x^worstAnd updating the harmony data to the harmony memory library HM, otherwise, performing the next harmony creation without any operation.

10. The maximum power point tracking method under the local shadow condition according to claim 7, wherein in the step 4), if the algorithm operation does not reach the maximum iteration number NI, the new harmony vector is continuously generated and the harmony memory base is updated until the algorithm termination criterion is reached; and if the operation of the algorithm reaches the maximum iteration number NI, outputting the optimal voltage and tracking to a global maximum power point.

Technical Field

The invention relates to the technical field of photovoltaic power generation, in particular to a maximum power point tracking method under a local shadow condition.

Background

With the development of society, energy and environmental problems are more and more prominent, and solar energy has a good application prospect as a renewable energy source. However, the photovoltaic conversion efficiency of the photovoltaic cell is low, the output power has a large relationship with the sunlight intensity and the ambient temperature, and the photovoltaic cell has obvious nonlinearity, and a Maximum Power Point Tracking (MPPT) circuit needs to be connected between the photovoltaic device and the load to fully exert the efficacy of the photovoltaic cell. The conventional MPPT control algorithm comprises a constant voltage tracking control method, a disturbance observation method, a conductance increment method and the like.

Under the condition of local shadow, a photovoltaic power generation P-V curve is changed in a multi-peak mode, a constant voltage method needs to work under the condition of specific illumination, the tracking precision is poor under the condition of local shadow, and the power loss is large; the maximum power tracking precision under a unimodal P-V curve is high by a disturbance observation method and a conductance incremental method, but the maximum power tracking precision under a unimodal P-V curve under a local shadow can only track a first power peak point, but cannot track a following power peak point, so that the situation of local optimization is caused, the maximum photovoltaic energy cannot be obtained, and the power generation efficiency is low; while some intelligent algorithms such as a particle swarm optimization algorithm, an artificial bee colony optimization algorithm and a random leaping algorithm can track the global optimum, the parameter setting is complex, and a certain probability is trapped in the local optimum.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a maximum power point tracking method under the condition of local shadow.

The purpose of the invention can be realized by the following technical scheme:

a maximum power point tracking method under the condition of local shadow comprises the following steps:

initializing parameters of a self-adaptive harmony search algorithm.

The algorithm parameter setting comprises the size HMS of the harmony memory bank, the consideration probability HMCR of the harmony memory bank and the pitch fine tuning probability PAR_max(Fine maximum probability) and PAR_min(fine minimum probability), Pitch Fine adjustment amplitude BW_max(Fine-tuning maximum amplitude) and BW_min(fine minimum amplitude), maximum number of iterations NI.

Preferably, the HMS takes 5, the probability HMCR is 0.3 considered by the acoustic memory bank, and the pitch fine-tuning probability PAR_max＝0.5，PAR_minTone fine tuning amplitude BW of 0.1_max＝1.4，BW_minThe maximum number of iterations NI is 0.1, 100.

Step two, initializing and memorizing the HMS with sound.

The Harmony memory bank (HM) is composed of HMs randomly generated Harmony vectors (solution vectors), represented by the following formula:

wherein each harmony { x¹，x²，L，x^HMSExpressing as the optimized initial voltage value; f (x) is a fitness function, and the target fitness function is expressed as f (x)ⁱ)＝P(U_pv)＝U_pv*I_pv。

And step three, generating new harmony sound.

The new harmony solution vector is created on an impromptu basis based on the harmony memory bank considering the probability HMCR, the pitch trimming probability PAR and the pitch trimming amplitude BW, i.e. the current voltage U is_pvAdjusting a certain step length, wherein the updating formula is as follows:

in the formula, x^newIndicating to update the voltage value stored in the harmonic wave; x is the number of^HMSRepresents the HMs-th harmonic vector in HM, where HMs is 5; u shape_ocAn open circuit voltage for the photovoltaic cell; r is₁And rand (1) is represented by [0,1 ]]A random number is generated.

The PAR dynamics equation is:

wherein PAR (t) denotes the change in the probability PAR with non-linear decrease of the iterative pitch trimming, PAR_maxAnd PAR_min0.5 and 0.1, respectively, t representing the current iteration number.

The BW dynamic change formula is:

BW(t)＝(BW_max-BW_min)e^-t+BW_min

where BW (t) represents the change in BW with a non-linear decrease in the iterative pitch trimming amplitude BW_maxAnd BW_min1.4 and 0.1, respectively, t representing the current iteration number.

Obtaining a harmony variable from the database, and if the harmony variable is obtained from the harmony library, fine tuning the harmony variable; otherwise, no adjustment is made; x is the number of^newThe update formula is:

in the formula, r₂Is [0,1 ]]A random number is generated.

Step four, updating and memorizing the sound database (HM).

In this step, the worst sum sound x stored in the HM is determined^worstFor newly generated harmony x^newPerforming fitness function evaluation if f (x)^new)＞f(x^worst) Then use x^newReplacement of x^worstAnd updating the harmony data to the HM, otherwise, performing the next harmony creation without any operation.

And step five, checking whether the algorithm is terminated.

When the termination criterion is not reached, namely the maximum optimization time (the maximum iteration time NI) is not reached, returning to the third step and the fourth step, and operating again until the termination criterion is reached; and finally, outputting the optimal voltage and tracking to a global maximum power point.

Compared with the prior art, the maximum power point tracking method under the local shadow condition at least has the following beneficial effects:

1) the method utilizes the characteristics of strong global searching capability and high convergence speed of the improved harmony search algorithm of the human behavior imitating mechanism to be applied to the photovoltaic power generation MPPT under the local shadow, avoids the trapping of local optimization, improves the tracking speed and precision and improves the solar energy utilization rate.

2) The harmony search algorithm is an optimization algorithm imitating human intelligence, and the optimization is completed by repeatedly adjusting solution variables in the memory library to enable function values to continuously converge along with the increase of iteration times; the algorithm concept is simple, the adjustable parameters are few, the realization is easy, the algorithm can effectively jump out of local optimum through the self-adaptive setting of the parameters PAR and BW, the tracking precision and the convergence speed are improved, and the global maximum power point is tracked finally.

Drawings

FIG. 1 is a diagram of a photovoltaic MPPT simulation model in an embodiment;

FIG. 2 is a diagram of a simulation model of a partially shaded photovoltaic cell in an embodiment;

FIG. 3 is a P-V graph partially shaded in an embodiment;

fig. 4 is a schematic flowchart of a maximum power point tracking method under a local shadow condition according to an embodiment of the present invention;

FIG. 5 is a diagram illustrating MPPT effects of the present invention using adaptive and acoustic search algorithms in an embodiment.

Detailed Description

The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.

Examples

The invention relates to a maximum power point tracking method under a local shadow condition, which is realized based on a self-adaptive harmony search algorithm. The principle of the MPPT controller model based on HS is as follows: the method comprises HS algorithm code, a voltage outer ring current inner ring comparison module and a duty ratio conversion module, wherein the Mppt algorithm obtains the current optimizing voltage V and the photovoltaic cell voltage V through iterative calculation_pvPerforming difference comparison to form a voltage outer ring, and adjusting the difference value with the current I of the photovoltaic cell after the difference value is subjected to PI adjustment_pvAnd performing difference to form a current inner loop, performing difference to obtain an amplification factor of 15, performing gain to obtain a value serving as a numerator, dividing the value by the denominator 50, and outputting d'. The voltage outer ring ensures that the output voltage is stable, and the current inner ring performs sampling feedback in a small range to ensure that the waveform is stable. Finally, duty ratio d of the sawtooth wave is adjusted through the compared voltage, and the on-off of the Mosfet device is controlled. Secondly, simulating the local shadow condition in a photovoltaic cell model, and respectively giving 1000W/m to three photovoltaic panels at the standard temperature of 25 DEG C²、600W/m²And 300W/m²Different illumination intensities; finally, the input quantity V is measured_pvAnd I_pvCalculating the current power P (V)_pv)＝V_pv*I_pvComparing the fitness function f (V) of P by the harmony search algorithm of MPPT controller_pv) Magnitude versus voltage V_pvPerforming iterative optimization adjustment to output the optimal occupying voltage V_bestTo realize obtaining the global maximum power P_max。

First, as shown in fig. 1, a photovoltaic MPPT simulation model is established by Matlab/Simulink, and a photovoltaic cell is formed by connecting three photovoltaic cell modules in series, and has an open-circuit voltage V_oc37.92V, short-circuit current I_sc8.62A, the parameters are set as shown in fig. 2, T1, T2, T3, and S1, 1000W/m²，S2＝600W/m²，S3＝300W/m²So as to simulate the situation of local shadow; then, by measuring V in the model_pv，I_pvAnd as an input quantity, inputting the input quantity into the MPPT algorithm module for iterative optimization, outputting a current voltage V, converting the current voltage V into a duty ratio through double closed-loop control to control the switching-on and switching-off of a Mosfet device, and performing optimal voltage regulation so as to perform maximum power point tracking.

The P-V curve of the photovoltaic model of the local shadow used by the invention is shown in figure 3, the ordinate is power P, the abscissa is voltage V, and three illumination intensities which represent the local shadow are 1000W/m²、600W/m²、300W/m²Under irradiationThe measured P-V curve is in a three-peak state, wherein the peak point of the maximum power point is 104.5W, and the corresponding optimal voltage is 21.01V.

The control block diagram of the method of the invention is shown in fig. 4, and the specific steps are as follows:

s1: algorithm parameters are initialized.

The algorithm parameters comprise the size HMS of the harmony memory base, namely the number of solution vectors; the maximum iteration number NI is 300; the harmony memory bank considers the probability HMCR, the pitch trimming probability PAR, and the pitch trimming amplitude BW.

S2: the HMS is initialized and acoustically memorized.

The Harmony memory bank (HM) is composed of HMs randomly generated Harmony vectors (solution vectors), represented by the following formula:

in HS (Harmony search) algorithm (harmony search algorithm), X1, X²,…X^HMSIs the size of the population; random harmony tone for each populationIs an n-dimensional vector; the dimension n in the photovoltaic MPPT algorithm is 1, so the HM formula in the photovoltaic model is as follows;

let x beⁱIn [0, U_oc]The maximum power point is usually 0.7V as can be known from experience_ocGenerated around 26V at 0.7V_ocNear set up HMS 5 initial harmony, the initial voltage value of each harmony is { x¹，x²，x³，x⁴，x⁵}＝[18，22，26，30，34](ii) a f (x) is a fitness function, and the target fitness function is expressed as f (x)ⁱ)＝P(V_pv)＝V_pv*I_pv。

S3: a new harmony sound is generated.

The new harmony solution vector is based on HMCR, PAR and BW for impulse creation, i.e. for the current voltage U_pvAdjusting a certain step length; the HMCR is 0.3, and the smaller HMCR is beneficial to the diversity of the population and the improvement of the global capability; if r₁<HMCR, randomly taking a harmony variable from the harmony memory library; otherwise, randomly generating a harmony variable from the solution space to take a value from the outside, so as to avoid falling into the local optimum, wherein the updating formula is as follows:

wherein r is₁Is at [0,1 ]]A random number is generated and compared with the HMCR.

For setting PAR, in the initial stage of HS algorithm search, smaller PAR is beneficial to the algorithm to quickly search a region with larger power, and in the later stage of HS algorithm search, larger PAR is beneficial to the algorithm to jump out a local extreme value, track to a global maximum power point, and set PAR_min＝0.1,PAR_max0.5, decrease PAR (t) with increasing number of iterations, t being the current number of iterations, t ∈ [0, NI]Dynamically changing according to the formula:

for BW setting, in the initial stage of algorithm search, the adoption of larger BW is beneficial to the algorithm to explore in a larger range, and graceful harmonic storage and comparison can be carried out on all power extreme points to strengthen the global property; in the later stage of algorithm search, the adoption of smaller BW is beneficial to the fine search in a small range of the algorithm, and the maximum power tracking precision is improved. BW (t) is set to decrease as the number of iterations increases_min＝0.5，BW_maxT is the current iteration number, t ∈ [0, NI ═ 3]Dynamically changing according to the formula: BW (t) ═ BW_max-BW_min)e^-t+BW_min。

From the above, a harmonic variable is obtained, if this isThe harmony variable is obtained from a harmony library, and the harmony variable needs to be subjected to fine adjustment; otherwise, no adjustment is made;the update formula is:

wherein r is₂Is [0,1 ]]Generates a random number if r₂<And PAR (t) adjusting the obtained harmonic variable according to the fine tuning bandwidth BW to obtain a new harmonic variable corresponding to a new voltage value Upv.

S4: update and sound memory library (HM).

In this step, the worst sum sound x stored in the HM is determined^worstFor newly generated harmony x^newIndicating the present voltage U_pvPerforming fitness function evaluation if f (x)^new)＞f(x^worst) Indicating that the power P obtained at the present voltage value is larger, then x is used^newReplacement of x^worstAnd updating the harmony data to the HM, otherwise, performing the next harmony creation without any operation.

S5: it is checked whether the algorithm is terminated.

When the termination criterion (i.e., the maximum number of optimizations) is not reached, the process returns to step S3 and step S4, and the operation is repeated until the termination criterion is reached, the optimal voltage is output, and the global maximum power point is tracked.

The MPPT effect graph based on the adaptive harmony search algorithm is shown in FIG. 5, and the optimal voltage V output at this time can be known by combining the P-V curve of FIG. 3_bestThe MPPT is tracked to the global maximum power point P under the condition of 21.01V_max104.5W, the effectiveness of the algorithm in terms of a multi-peak curve global search is demonstrated.

While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.