CN104793691A - Photovoltaic array whole situation MPPT method based on ant colony algorithm under partial shadow - Google Patents

Abstract

The invention discloses a photovoltaic array whole situation MPPT method based on an ant colony algorithm under a partial shadow. The method comprises the steps that 1 the voltage of the maximum power point in a whole situation acted by a photovoltaic array is researched by using the ant colony algorithm combining with a PI controller; 2 a Boost circuit connected to the photovoltaic array is controlled by the PI controller, and the photovoltaic array is operated on the optimal voltage searched in the first step; 3 the whole situation maximum power point of the photovoltaic array is traced by using a small-step-size disturbance observation method in the position of the optimal voltage. By means of the decimal ant colony algorithm, the maximum searching optimization of a voltage variate to power is generated, the output voltage of the array is controlled by the PI controller, and the anti-interference performance of a system is effectively improved; by means of the ant colony algorithm combining with the observation method, the problem of tracking the maximum power point of the photovoltaic array with multi-peak output features can be solved, the disadvantage that conventional algorithm tracking may fall into a local maximum power point is overcome, the real whole situation maximum power point can be found, and rapid and precise tracking is achieved.

Description

A kind of overall situation of the photovoltaic array under local shadow based on ant group algorithm MPPT method

Technical field

The invention belongs to technical field of photovoltaic power generation.

Background technology

Along with the minimizing of global energy and the continuous deterioration of environment, energy-conservationly to attract attention with environmental protection, photovoltaic power generation technology is renewable and more and more come into one's own because of its clean energy.Photovoltaic cell is a kind of typical non-linear power, and its output power is easily affected by the external environment, as light radiation intensity, battery temperature etc.For making full use of sun power, improving the conversion efficiency of photovoltaic cell, just need be controlled the output voltage of photovoltaic cell and electric current, ensure that photovoltaic cell always works in maximum power point (MPP).

For the maximal power tracing (MPPT) of photovoltaic cell, conventional method has fixed voltage method (CV), disturbance observation (P & O), conductance increment method (IC) etc.Wherein P & O method controls simply, is easy to realize, and engineering is applied comparatively wide, IC method tracking effect is better than P & O method, but realizes more complicated.These algorithms are followed the tracks of the maximum power point of photovoltaic array that single peak exports and are achieved good effect, but cloudy weather and near have buildings, trees block time, local shades situation may be there is in photovoltaic array, at this moment there is multimodal in its power out-put characteristic, the algorithm of above-mentioned routine is easily absorbed in local extremum, can not export real peak power.

Summary of the invention

Object of the present invention is solve the global maximum power point tracking problem of photovoltaic array under local shadow when be multimodal output characteristics, proposes a kind of overall MPPT method of photovoltaic array under local shadow based on ant group algorithm.

The present invention is achieved by the following technical solutions.

A kind of overall situation of the photovoltaic array under local shadow based on ant group algorithm MPPT method of the present invention, comprises the steps:

(1) ant group algorithm is utilized to work in the voltage at global maximum power point place in conjunction with PI controller search photovoltaic array;

(2) by the Boost circuit of PI controller control linkage photovoltaic array, make photovoltaic array stable operation at (1) optimum voltage place of seeking;

(3) global maximum power point of the disturbance observation tracking photovoltaic array of little step-length is adopted at optimum voltage place.

Further, described step (1) utilizes ant group algorithm in conjunction with the method for PI controller optimizing to be:

(1-1) each parameter of initialization algorithm, determines ant quantity n and decimal system voltage solution precision, and constructs decimal number optimizing map;

(1-2) ant group algorithm and decimal system optimizing map is utilized to generate each ant voltage solution;

(1-3) successively the reference value of each ant voltage solution as PI controller is compared with measured light photovoltaic array output voltage, obtain pulse-length modulation (PWM) Duty ratio control amount through PI controller;

(1-4) control Boost circuit by PWM Duty ratio control amount, make photovoltaic array output voltage stabilization at reference value place, and measure array output power corresponding to each voltage solution;

(1-5) array output power of more each ant gained and peak power record at present, upgrade the optimum voltage of current maximum power value and correspondence thereof, the pheromones in renewal decimal number optimizing map and expectation value; Judge whether to meet termination of iterations condition, if do not meet, get back to (1-2).

Further, in step (2), PI controller obtains by input voltage deviation the switching tube (IGBT) that PWM Duty ratio control amount controls Boost circuit, makes array output voltage stabilization.

Further, described step (3) adopts the method for the global maximum power point of little step-length disturbance observation accurate tracking photovoltaic array to be:

(3-1) a little step-length disturbance quantity Δ V is applied to the voltage reference value of PI controller;

(3-2) measure photovoltaic array output power P (k+1), and compare to obtain Δ P=P (k+1)-P (k) with P (k);

If (3-3) Δ P/ Δ V is just, then next step disturbance quantity Δ V is just, otherwise is then negative.

Feature of the present invention and beneficial effect:

1, adopt decimal system ant group algorithm formation voltage variable to the optimizing of power maximal value, controlled the output voltage of array by PI controller, the anti-interference of effective raising system;

2, ant group algorithm can solve the maximum power point of photovoltaic array tracking problem with multimodal output characteristics in conjunction with disturbance observation, overcome conventional algorithm and follow the tracks of the shortcoming that may be absorbed in local maximum power point, real global maximum power point can be found and realize following the tracks of fast and accurately.

Accompanying drawing explanation

Accompanying drawing 1 is system construction drawing of the present invention.

Accompanying drawing 2 is algorithm flow chart of the present invention.

Accompanying drawing 3 is decimal number optimizing map of the present invention.

Accompanying drawing 4 is the process flow diagram of the present invention's medium and small step-length disturbance observation.

Embodiment

Below in conjunction with accompanying drawing and principle of work, the specific embodiment of the present invention is described in detail.

As shown in Figure 1, the present invention is the maximum power point of photovoltaic array tracking based on ant group algorithm, it is main when photovoltaic array is under local shades, not mating because of each assembly illumination causes array output characteristics to be polymodal curve, adopt ant group algorithm formation voltage variable to search global maximum power point, and utilize PI controller that system works is stablized; Search terminates rear little step-length disturbance observation to its accurate tracking, and the global maximum power point realizing photovoltaic array under local shadow is followed the tracks of.

As shown in Figure 2, the control method of the embodiment of the present invention comprises the following steps:

1, each parameter alpha of initialization algorithm, β, ρ, Q, determines ant number m and decimal number figure place D (this routine m=3, D=3).

2, decimal system optimizing map according to Fig. 3, ant creeps generation pass in the drawings, obtains each ant voltage solution.In optimizing map, the longitudinal axis is everybody value of decimal number X, and transverse axis represents that decimal number X's is every successively, and this routine transverse axis comprises two integer-bit and a decimal place, and this figure also easily extensible becomes more high precision and multidimensional variable.The decision variable of the one dimension of path representation one shown in Fig. 3, its value is determined by following formula:

$X=\underset{d=1}{\mathrm{\Σ}}[n_d\×{10}^{(b-d)}]---\left(1\right)$

Wherein n _dbe the position that ant d walks, b is the integral part figure place of X, b=2, X=34.2 in Fig. 3.

3, in ant group algorithm, ant creeps working direction by the decision of formula (2) new probability formula, the node advanced according to roulette method choice. represent m ant goes to the j point of d layer probability from the i point of d-1 layer; τ _d(i, j) is the pheromones of point-to-point transmission, η _dj () is the expectation value of d layer j point, its value equals [10-abs (j _d-j _db)]/10, j _dbrepresent each decimal digit of current globally optimal solution; α, β are the significance level of pheromones and expectation value respectively.This rule can make ant tend to select pheromones higher and form the path of current optimum solution.

$P_d^m(i,j)=\frac{{\mathrm{\τ}}_d{(i,j)}^{\mathrm{\α}}\·{\mathrm{\η}}_d{\left(j\right)}^{\mathrm{\β}}}{\underset{j=0}{\overset{9}{\mathrm{\Σ}}}{\mathrm{\τ}}_d{(i,j)}^{\mathrm{\α}}\·{\mathrm{\η}}_d{\left(j\right)}^{\mathrm{\β}}}---\left(2\right)$

4, successively the voltage solution that each ant draws is flowed to PI controller, the photovoltaic array output power value corresponding to each voltage solution can be obtained according to the electric current and voltage of measuring photovoltaic array output.

5, the performance number of actual measurement is come lastest imformation element and at present peak power and optimum solution as ant group algorithm evaluation function f.Pheromone update formula is as follows:

${\mathrm{\τ}}_d^{k+1}(i,j)=(1-\mathrm{\ρ}){\mathrm{\τ}}_d^k(i,j)+\mathrm{\Δ}{\mathrm{\τ}}_d(i,j)---\left(3\right)$

Wherein, ρ is pheromones volatility coefficient, Δ τ _d(i, j) is the pheromones increment between d-1 layer i point to d layer j point, and pheromones increment is only for the path that the optimum ant of each iteration is passed by, and its expression formula is:

Δτ _d(i,j)＝Q·f _max(4)

Wherein, Q is a scale-up factor, f _maxthe i.e. power maximal value of current iteration.

After carrying out formula (3), for preventing pheromones difference between each section too large, the codomain of introduction-type (5) restricted information element, there is the phenomenon no longer spreading, stagnate too early in Restrainable algorithms.

${\mathrm{\&tau;}}_d^{k+1}(i,j)=\left\{\begin{array}{cc}{\mathrm{\&tau;}}_{\min },& \mathrm{if}{\mathrm{\&tau;}}_d^{k+1}(i,j)<{\mathrm{\&tau;}}_{\min }\\ {\mathrm{\&tau;}}_{\max },& \mathrm{if}{\mathrm{\&tau;}}_d^{k+1}(i,j)>{\mathrm{\&tau;}}_{\max }\end{array}\right.---\left(5\right)$

6, whether detection algorithm meets stop condition, when iterations is greater than certain number of times, is assumed to be K _minif peak power of seeking is not improved for a long time, then judge that ant group algorithm needs out of service, otherwise return the 2nd step and carry out next iteration.

7, the optimum voltage searched by ant group algorithm is as the input of PI controller, at this moment, the output voltage of photovoltaic array will be stabilized in sought optimum voltage place, this voltage is the voltage corresponding to the maximum power point of photovoltaic array, also may be the voltage near maximum power point, so the disturbance observation now calling little step-length carries out accurate tracking to maximum power point, concrete grammar as shown in Figure 4.

8, apply a little step-length disturbance quantity Δ V to voltage reference value, measure new power stage value, and compare with previous step performance number, obtain Δ P.

9, judge next step perturbation direction according to formula Δ P/ Δ V, if it is just, then next step disturbance quantity Δ V is just also, otherwise then Δ V is negative.

10, because the output characteristics of array when external environment changes also will change, ant group algorithm should be restarted and search for new global maximum power point.When large change occurs suddenly power stage, when namely formula (6) is set up, then think and again need search for peak power.Another when environmental change is slow, algorithm also needs the ability of following the tracks of peak power and slowly changing, so setting ant group algorithm also needs nature to restart in time, often through after a while, algorithm restarts the new maximum power point of search voluntarily.If meet above-mentioned condition, then get back to the 1st step, otherwise, then get back to the 8th step.

$\frac{|P_{\mathrm{new}}-P_{\mathrm{old}}|}{P_{\mathrm{old}}}>0.1---\left(6\right)$

In formula, P _newthe power newly detected, P _oldthen the power before disturbance.

Claims (4)

1., based on a photovoltaic array under local shadow overall situation MPPT method for ant group algorithm, it is characterized in that comprising the steps:

(1) ant group algorithm is utilized to work in the voltage at global maximum power point place in conjunction with PI controller search photovoltaic array;

(2) by the Boost circuit of PI controller control linkage photovoltaic array, make photovoltaic array stable operation at (1) optimum voltage place of seeking;

(3) global maximum power point of the disturbance observation tracking photovoltaic array of little step-length is adopted at optimum voltage place.

2. a kind of overall situation of the photovoltaic array under local shadow based on ant group algorithm MPPT method according to claim 1, is characterized in that described step (1) is:

(1-1) each parameter of initialization algorithm, determines ant quantity n and decimal system voltage solution precision, and constructs decimal number optimizing map;

(1-2) ant group algorithm and decimal system optimizing map is utilized to generate each ant voltage solution;

(1-3) successively the reference value of each ant voltage solution as PI controller is compared with measured light photovoltaic array output voltage, obtain PWM Duty ratio control amount through PI controller;

(1-4) control Boost circuit by PWM Duty ratio control amount, make photovoltaic array output voltage stabilization at reference value place, and measure array output power corresponding to each voltage solution;

(1-5) array output power of more each ant gained and peak power record at present, upgrade the optimum voltage of current maximum power value and correspondence thereof, the pheromones in renewal decimal number optimizing map and expectation value; Judge whether to meet termination of iterations condition, if do not meet, get back to (1-2).

3. a kind of overall situation of the photovoltaic array under local shadow based on ant group algorithm MPPT method according to claim 1, it is characterized in that described step (2) is: PI controller obtains by input voltage deviation the switching tube that PWM Duty ratio control amount controls Boost circuit, makes array output voltage stabilization.

4. a kind of overall situation of the photovoltaic array under local shadow based on ant group algorithm MPPT method according to claim 1, is characterized in that described step (3) is:

(3-1) a little step-length disturbance quantity Δ V is applied to the voltage reference value of PI controller;

(3-2) measure photovoltaic array output power P (k+1), and compare to obtain Δ P=P (k+1)-P (k) with P (k);

If (3-3) Δ P/ Δ V is just, then next step disturbance quantity Δ V is just, otherwise is then negative.