CN109753102B

CN109753102B - Improved photovoltaic fuzzy MPPT control method

Info

Publication number: CN109753102B
Application number: CN201711075039.XA
Authority: CN
Inventors: 游国栋; 严宇; 苏虹霖; 徐涛; 王军; 沈延新
Original assignee: Tianjin University of Science and Technology
Current assignee: Tianjin University of Science and Technology
Priority date: 2017-11-01
Filing date: 2017-11-01
Publication date: 2021-08-03
Anticipated expiration: 2037-11-01
Also published as: CN109753102A

Abstract

The invention relates to an improved photovoltaic fuzzy MPPT control method, which is mainly technically characterized by comprising the following steps: scanning, storing, disturbing and observing are adopted to obtain the optimal power. Under initial conditions or variable weather, the method can search and store the maximum power value of the photovoltaic system in a large range. The preset value represents an acceptable difference between the determined maximum power and the operating power as determined by the controller rules. If the difference between the determined maximum power and the operating power is greater than a preset value, the duty cycle is increased; otherwise, MPPT based on fuzzy logic is applied. In this manner, the algorithm may ensure that the MPPT does not get stuck in the local MPP and quickly reverts to the new global MPP. The invention combines the maximum power point tracking method based on fuzzy logic with scanning storage. The controller has quick convergence, can avoid the fluctuation of the working point near the MPP, greatly improves the control effect of the MPPT, and enhances the dynamic response of the system.

Description

Improved photovoltaic fuzzy MPPT control method

Technical Field

The invention relates to a method for tracking the maximum power of solar energy, in particular to an improved maximum power point tracking method controlled by photovoltaic fuzzy logic.

Background

With the rapid development of industry, the demand of human beings for energy is continuously increased, and the trend of the world has been to search for new energy to replace the conventional energy resources. Photovoltaic energy, one of the available alternative energy sources, is one of the most promising renewable energy sources. The photovoltaic energy source is clean and simple in design. At present, both household and industry have an optimistic attitude on photovoltaic power generation, and it is hoped that photovoltaic energy can be fully utilized, and the photovoltaic energy can be widely applied to various aspects of life and production of people. However, the high installation cost and the low photovoltaic efficiency are two main disadvantages of the photovoltaic system, and the low photovoltaic efficiency is mainly that the output characteristic has strong nonlinear characteristics, and the output current and the maximum output power of the solar panel can greatly change along with the difference of the light intensity and the ambient temperature.

The traditional algorithms comprise a fixed voltage method, a disturbance observation method, an admittance increment method, an optimal gradient method, a hysteresis comparison method, a neuron network control method, a fuzzy logic control method and the like. There are also some novel control strategies proposed and applied in practice in the academic world at home and abroad. The students of Pentao and Dingkun of river and sea university propose an improved global maximum power point control algorithm based on a disturbance observation method and an admittance incremental method, and the method has global improvement on the aspects of tracking rapidity and system stability. The Wenzand learner of Harbin's university of Physician uses the minimum quadratic multiplication to perform data fitting, and designs a three-point least square method combining fixed step length and variable step length to perform system tracking, so as to obtain the maximum output power. The Miyatake M scholars employ a direct search algorithm to search for the maximum power point, and to ensure that a global MPP is found, the initial point must be carefully selected, otherwise the controller may be trapped in a local MPP. To reduce the impact of local shadows, the Nguyen D scholars propose adaptive solar photovoltaic arrays, and control the fixed part connecting the solar adaptive library and the photovoltaic array switch matrix by using a model-based control algorithm. Likewise, the dynamic array reconstruction algorithm may improve the output of the photovoltaic system under local shadow conditions. The university of Velasco-quasdaa G inserts a controllable switch matrix between the photovoltaic power generation system and the central inverter to electrically reconnect the photovoltaic modules. In these methods, MPP can be obtained by using a conventional MPPT algorithm, but their power levels are complex and the cost is high. Therefore, research on an MPPT control method with low cost and high efficiency is required.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide an improved photovoltaic fuzzy MPPT control method which is reasonable in design and has good steady-state and dynamic performance.

The technical problem to be solved by the invention is realized by adopting the following technical scheme:

an improved photovoltaic fuzzy MPPT control method is characterized by comprising the following steps:

step 1, the photovoltaic system needs to be sampled before obtaining the optimal MPP, so the system is initialized by utilizing the maximum duty ratio.

Step 2, scanning a P-U curve and storing the global MPP; a large perturbation is preset to increase the search range.

And 3, obtaining the global MPP through disturbance and observation.

And 4, fuzzifying the input quantity and the output quantity.

The input and output of the fuzzy logic controller are as follows:

ΔP＝P(k)-P(k-1)

ΔI＝I(k)-I(k-1)

ΔP_M＝P_M(k)-P(k)

ΔD＝D(k)-D(k-1)

wherein, the delta P and the delta I are respectively the change of the output power and the current of the photovoltaic array, and the delta P_MGlobal MPP (P) for storage_M) Δ D is the duty cycle change from the current power. Δ P and Δ I are divided into four fuzzy subsets: positive large (PB), Positive Small (PS), negative large (NB), and Negative Small (NS). Delta P_MTwo fuzzy subsets are divided: PB and PS. Δ D is divided into six fuzzy subsets: PB, median (PM), PS, NB, Negative Median (NM), and NS. Therefore, the algorithm requires 32 control rules, which are based on the PO algorithm rules and are adjusted with reference power.

And 5, converting the fuzzy quantity obtained by the reasoning into a clear quantity.

These fuzzy combinations are deblurred using the Mamdani method with max-min. The formula is as follows:

where Δ D is the fuzzy control output, D_iIs the center of the maximum-minimum range of the output membership functions.

The invention has the advantages and positive effects that:

1. unlike the conventional Perturbation and Observation (PO) method for tracking MPP to obtain the optimal operating power of the photovoltaic system, the improved pv fuzzy MPPT method adopts scanning, storage, perturbation and observation to obtain the optimal power. Under initial conditions or variable weather conditions, the method can search a large range to scan and store the maximum power value of the photovoltaic system. The preset value represents an acceptable difference between the determined maximum power and the operating power as determined by the controller rules. If the difference between the determined maximum power and the operating power is greater than a preset value, the duty cycle is increased; otherwise, MPPT based on fuzzy logic is applied. In this manner, the algorithm may ensure that the MPPT does not get stuck in the local MPP and quickly reverts to the new global MPP.

2. The improved photovoltaic fuzzy logic MPPT combines the MPPT based on fuzzy logic with a scanning storage system, so that the dynamic response of the system is improved; after the algorithm is applied, the working point is fixed near the MPP, and the control effect of the MPPT is greatly improved; even under variable weather conditions, the MPPT can scan and track the global MPP in a short time, and the controller has high convergence speed and high system tracking performance.

Drawings

FIG. 1 is a modified photovoltaic fuzzy MPPT flow diagram and controller;

FIG. 2 is a graph of the shape of the membership function of the input and output and the fuzzy subset partition;

Detailed Description

The embodiments of the invention will be described in further detail below with reference to the accompanying drawings:

an improved photovoltaic fuzzy MPPT control method is different from a traditional Perturbation and Observation (PO) method for tracking MPP to obtain the optimal operating power of a photovoltaic system, and the improved photovoltaic fuzzy MPPT method adopts scanning, storage, perturbation and observation to obtain the optimal power. Under initial conditions or variable weather conditions, the method can search a large range to scan and store the maximum power value of the photovoltaic system. The preset value represents an acceptable difference between the determined maximum power and the operating power as determined by the controller rules. If the difference between the determined maximum power and the operating power is greater than a preset value, the duty cycle is increased; otherwise, MPPT based on fuzzy logic is applied. In this manner, the algorithm may ensure that the MPPT does not get stuck in the local MPP and quickly reverts to the new global MPP. FIG. 1a is a flow chart in which D is the duty cycle, P_MTo global MPP, Δ P_MIs a constant that identifies the allowable difference between the global MPP and the operating power point.

The algorithm employs three scanning and storage methods: the photovoltaic system needs to be sampled before obtaining the optimal MPP, so the system is initialized with the maximum duty cycle. The method comprises the steps of scanning a P-U curve to search and store the global MPP; the duty cycle is increased by a fixed step. Meanwhile, scanning a P-U curve and storing the global MPP; a large perturbation is preset to increase the search range. Unlike the two methods described above, this method is through perturbation and observation to obtain the global MPP. Through the three methods, the system can be ensured to find and store the global MPP, and the time for identifying the global MPP by the three methods is different. In addition, whenever global MPP is found, it must return to a minimum value as long as the duty cycle exceeds a maximum threshold.

The improved fuzzy logic based MPPT algorithm (fig. 1b), using a scan and store program, can quickly locate the global MPP. The input and output of the fuzzy logic controller are as follows:

ΔP＝P(k)-P(k-1)

ΔI＝I(k)-I(k-1)

ΔP_M＝P_M(k)-P(k)

ΔD＝D(k)-D(k-1) (1)

wherein, the delta P and the delta I are respectively the change of the output power and the current of the photovoltaic array, and the delta P_MΔ D is the duty cycle change, which is the difference between the stored global mpp (pm) and the current power. Δ P and Δ I are divided into four fuzzy subsets: positive large (PB), Positive Small (PS), negative large (NB), and Negative Small (NS). Delta P_MTwo fuzzy subsets are divided: PB and PS. Δ D is divided into six fuzzy subsets: PB, median (PM), PS, NB, Negative Median (NM), and NS. Therefore, the algorithm requires 32 control rules, which are based on the PO algorithm rules and are adjusted with reference power. These fuzzy combinations are operated on using the Mamdani method with max-min. The core algorithm during defuzzification is to convert the duty ratio of the fuzzy subset into a real number, and the formula is as follows:

where Δ D is the fuzzy control output, D_iIs the center of the maximum-minimum range of the output membership functions. The shapes of the membership functions and fuzzy subset partitions of the inputs and outputs are shown in fig. 2.

It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.

Claims

1. An improved photovoltaic fuzzy MPPT control method is characterized in that under an initial condition or a variable weather condition, the maximum power value of a photovoltaic system is searched and scanned in a large range and stored, and if the difference between the determined maximum power and the operation power is larger than a preset value, the duty ratio is increased; otherwise, applying MPPT based on fuzzy logic; the preset value represents an acceptable difference between the determined maximum power and the operating power determined by the controller rules; the method comprises the following steps:

step 1, a photovoltaic system needs to be sampled before obtaining the optimal MPP, so that the system is initialized by utilizing the maximum duty ratio;

step 2, scanning a P-U curve and storing the global MPP; increasing a duty ratio with a fixed step length, scanning a P-U curve at the same time, and storing the global MPP;

step 3, presetting a large disturbance to increase the search range, and obtaining the global MPP through disturbance and observation;

step 4, fuzzification of input quantity and output quantity;

wherein, the input and output of the fuzzy logic controller are as follows:

ΔP＝P(k)-P(k-1)

ΔI＝I(k)-I(k-1)

ΔP_M＝P_M(k)-P(k)

ΔD＝D(k)-D(k-1)

in the formula, Δ P and Δ I are respectively the output power of the photovoltaic arrayChange in current, Δ P_MGlobal MPP (P) for storage_M) From the difference in current power, Δ D is the duty cycle change, and Δ P and Δ I are divided into four fuzzy subsets: positive Big (PB), Positive Small (PS), Negative Big (NB) and Negative Small (NS), Δ P_MTwo fuzzy subsets are divided: PB, PS, Δ D are divided into six fuzzy subsets: PB, median (PM), PS, NB, Negative Median (NM), and NS; therefore, the algorithm needs 32 control rules which are based on the PO algorithm rule and adjusted along with the reference power;

step 5, converting the fuzzy quantity obtained by the reasoning into a clear quantity;

wherein the deblurring operation is performed on these fuzzy combinations using the Mamdani method with the maximum-minimum, and the formula is as follows:

where Δ D is the fuzzy control output and Di is the center of the maximum-minimum range of the output membership function.