CN102566646A

CN102566646A - Maximum power point tracking method under partial shade condition of photovoltaic system

Info

Publication number: CN102566646A
Application number: CN2012100111058A
Authority: CN
Inventors: 张渊明; 孙彦广; 张云贵
Original assignee: Automation Research and Design Institute of Metallurgical Industry
Current assignee: Automation Research and Design Institute of Metallurgical Industry
Priority date: 2012-01-13
Filing date: 2012-01-13
Publication date: 2012-07-11

Abstract

The invention discloses a maximum power point tracking method under a partial shade condition of a photovoltaic system, and belongs to the technical field of a photovoltaic power generation system. The maximum power point tracking method comprises the following steps of: searching a point close to a peak value of the maximum power point by using voltage scanning; and finally determining the position of the maximum power point using a variable step incremental conductance method by taking a voltage value of the point as an initial value of the incremental conductance method. According to the method, the problem of reduced system efficiency caused by incapability of tracking the maximum power point in the case that a battery panel in the photovoltaic system is under a shade condition is solved. The method has the advantages of simple structure and capability of effectively avoiding local extremum produced when the photovoltaic system gets into local shade.

Description

Maximum power point tracing method under the local shading condition of a kind of photovoltaic system

Technical field

The invention belongs to the photovoltaic generating system technical field, the maximum power point tracing method under the local shading condition of a kind of photovoltaic system particularly is provided.

Background technology

Existing maximum power point tracing method is based on photovoltaic cell output P-V characteristic (power-voltage characteristic) mostly and has that a peak value carries out.The most frequently used constant-voltage method, disturbance observation and conductance increment method will lose effectiveness fully under the condition that produces multimodal of shading.

The shortcoming of constant-voltage method: maximum power point voltage is fixed during beginning, when variations in peak, can't follow the tracks of.

The shortcoming of disturbance observation: carry out the change of perturbation direction based on the comparative result of output power; Because its disturbance step-length is fixed; When bigger change takes place in peak; Disturbance can't be crossed over two distances between the peak value, therefore can be absorbed in local maximum, can't trace into real maximum power point.

The shortcoming of conductance increment method and disturbance observation are similar, also are that the problem owing to step-length causes it to be absorbed in local extremum, even adopt the method for variable step, still can not effectively follow the tracks of unpredictable peak.

Shading to the part produces the situation of multi-peak, and the particle cluster algorithm of employing and fuzzy logic control method are also arranged.

The shortcoming of particle cluster algorithm: 1, convergence is unstable.2, parameter is difficult for selecting.3, the arithmetic speed to controller requires very high.

The shortcoming of fuzzy logic control: 1, control law is not easy to obtain.2, require very high to the controller arithmetic speed.

Summary of the invention

The object of the present invention is to provide the maximum power point tracing method under the local shading condition of a kind of photovoltaic system; Solve cell panel in the photovoltaic generating system and be under the shading condition when influencing (receive dust, trees, building even cloud layer etc.), can't track the problem that maximum power point causes system effectiveness to descend.

This method is simple in structure, the local extremum that produces in the time of can effectively avoiding being absorbed in the part and shade, and adopt the controller of general performance to meet the demands.

Processing step of the present invention is following:

1, at first adopting a scope is 0.1V-U _Set=(0.9～0.99) * Uoc (Uoc is the open-circuit voltage of photovoltaic array); Step-length is that the scanning voltage of 1V-6V scans photovoltaic array; Measure the output current I (k) of array; Calculate output power P (k)=U (k) * I (k), through comparing, the magnitude of voltage of performance number larger part is composed to U* then with P (k-1).Obtain the maximum a bit pairing scanning voltage U* of this sweep limit internal power behind the end of scan, this point is any point pairing voltage of maximum power point place on unimodal;

2, with the initial value of U* as variable step conductance increment method; Step-length with Δ U=0.1V-0.5V is carried out disturbance to U*, measures the output current I (k) of cell panel, calculates dU=U (k)-U (k-1); DI=I (k)-I (k-1), dP=U (k) * I (k)-U (k-1) * I (k-1).If dU is 0, judge the positive and negative of dI, greater than 0, then voltage increases Δ U, otherwise reduces Δ U as if dI; If dU is not 0, then judge the positive and negative of dI/dU+I/U, greater than zero, then voltage increases (dP/dU) * Δ U, otherwise reduces (dP/dU) * Δ U as if dI/dU+I/U.

The variable step conductance increment method that here adopts is in order to accelerate the tracking speed of conductance increment method; Add a coefficient d P/dU before giving step delta U, its mechanism of action is: when away from maximum power point, dP/dU is bigger for the P-V rate of curve; Step-length is also bigger, and search speed is faster; Along with maximum power point near, P-V slope of a curve dP/dU reduces, step-length also reduces gradually, search precision is higher;

Advantage of the present invention is following:

1, can avoid being absorbed in local extremum effectively through voltage scanning.

2, employing can be accelerated search speed and precision with scanning gained voltage as the variable step conductance increment method of initial value.

3, the scanning voltage of big step-length can be accelerated search speed equally.

4, also be suitable for for the homogeneous radiation condition, and search speed will be faster than classic method.

Description of drawings

Fig. 1 is the process flow diagram of algorithm of the present invention.Wherein: U (k) is a scanning voltage; Δ U is the scanning voltage step-length; I (k) is and the corresponding output current of U (k); P (k) is an output power; U _SetFor setting the scanning voltage maximal value; U* is the rreturn value after voltage scanning finishes; U is the voltage that whole algorithm finishes gained system maximum power point place, back.

Embodiment

With the 30kW photovoltaic array is example, and its open-circuit voltage is 820V, short-circuit current 49.3A, and the situation of shading is that irradiation is respectively 1000W/m2 and 500W/m2.

The concrete operating procedure of the inventive method is:

1, initial to array a step-length to be provided be Δ U=5V, and sweep limit is 0.1V-U _SetThe scanning voltage of=800V, the output current I (k) of measuring light photovoltaic array then;

2, calculate output power P (k)=U (k) * I (k),, the scanning voltage value of output power larger part is composed to U* through relatively big or small with output power P (k-1)=U (k-1) * I (k-1) of last time; U*=695V behind the end of scan;

3, with the U*=695V that the returns initial value as the conductance increment method, step-length adopts Δ U=0.5V.Measure the output current I (k) of cell panel, calculate dU=U (k)-U (k-1), dI=I (k)-I (k-1), dP=U (k) * I (k)-U (k-1) * I (k-1);

If 4 dU are 0, judge the positive and negative of dI, greater than 0, then voltage increases Δ U, otherwise reduces Δ U as if dI; If dU is not 0, then judge the positive and negative of dI/dU+I/U, greater than zero, then voltage increases (dP/dU) * Δ U, otherwise reduces (dP/dU) * Δ U as if dI/dU+I/U.

5, algorithm finishes, and returns U=688.6V, and the output power at this magnitude of voltage place is the peak power output under the current shading condition.

This method optimized operation condition is to combine local irradiation statistical conditions to set initial value and scanning step and scope; This method is also set up for the situation of homogeneous radiation, and its search speed and precision are better than existing searching method commonly used.

Claims

1. the maximum power point tracing method under the local shading condition of a photovoltaic system is characterized in that processing step is following:

(1) adopting a scope is 0.1V-U _Set=(0.9～0.99) * Uoc; Step-length is that the scanning voltage of 1V-6V scans photovoltaic array, measures the output current I (k) of array, calculates output power P (k)=U (k) * I (k); Through comparing, the magnitude of voltage of performance number larger part is composed to U* then with P (k-1); Obtain the maximum a bit pairing scanning voltage U* of this sweep limit internal power behind the end of scan, this point is any point pairing voltage of maximum power point place on unimodal, and described Uoc is the open-circuit voltage of photovoltaic array;

(2) with the initial value of U* as variable step conductance increment method; Step-length with Δ U=0.1V-0.5V is carried out disturbance to U*, measures the output current I (k) of cell panel, calculates dU=U (k)-U (k-1); DI=I (k)-I (k-1), dP=U (k) * I (k)-U (k-1) * I (k-1); If dU is 0, judge the positive and negative of dI, greater than 0, then voltage increases Δ U, otherwise reduces Δ U as if dI; If dU is not 0, then judge the positive and negative of dI/dU+I/U, greater than zero, then voltage increases (dP/dU) * Δ U, otherwise reduces (dP/dU) * Δ U as if dI/dU+I/U.