CN110571853A - wind-solar power generation MPPT control method and system based on radial basis function neural network - Google Patents

wind-solar power generation MPPT control method and system based on radial basis function neural network Download PDF

Info

Publication number
CN110571853A
CN110571853A CN201910836914.4A CN201910836914A CN110571853A CN 110571853 A CN110571853 A CN 110571853A CN 201910836914 A CN201910836914 A CN 201910836914A CN 110571853 A CN110571853 A CN 110571853A
Authority
CN
China
Prior art keywords
power generation
wind
generation device
output
boost circuit
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910836914.4A
Other languages
Chinese (zh)
Inventor
李自成
王志豪
王后能
曾丽
熊涛
刘健
文小玲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Institute of Technology
Original Assignee
Wuhan Institute of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan Institute of Technology filed Critical Wuhan Institute of Technology
Priority to CN201910836914.4A priority Critical patent/CN110571853A/en
Publication of CN110571853A publication Critical patent/CN110571853A/en
Pending legal-status Critical Current

Links

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/50—Photovoltaic [PV] energy
    • Y02E10/56—Power conversion systems, e.g. maximum power point trackers
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/76—Power conversion electric or electronic aspects

Abstract

The invention relates to a wind-solar power generation MPPT control method and a system based on a radial basis function neural network, wherein the wind-solar complementary power generation system comprises the following components: photovoltaic power generation subsystem and wind power generation system and direct current bus, photovoltaic power generation subsystem includes: the photovoltaic power generation system comprises a photovoltaic power generation module, a first current sensor, a first voltage sensor, a photovoltaic Boost circuit, a photovoltaic PWM control module and a first MPPT control module; the wind power generation subsystem includes: the wind power generation system comprises a wind driven generator, a second current sensor, a second voltage sensor, a wind power Boost circuit, a wind power PWM control module and a second MPPT control module; the technical scheme of the invention not only can realize the rapid tracking of the maximum power point of the output current of the photovoltaic power generation electronic system and the wind power generation electronic system in the wind-solar hybrid power generation system, but also reduces the current oscillation in the maximum power point tracking process, and improves the stability and the tracking efficiency of the maximum power point tracking process of the wind-solar hybrid power generation system.

Description

Wind-solar power generation MPPT control method and system based on radial basis function neural network
Technical Field
the invention relates to the field of power generation equipment control, in particular to a wind-solar power generation MPPT control method and system based on a radial basis function neural network.
background
Under the background of global energy crisis, solar energy and wind energy are mainly concerned as energy sources which are abundant in reserves, simple to obtain, clean and pollution-free. The wind-solar hybrid power generation system can ensure the stability of power supply and improve the utilization rate of wind energy and light energy. In order to improve the power generation efficiency of the wind-solar hybrid power generation system, the wind-solar hybrid power generation system is often controlled by adopting a maximum power point tracking technology. The existing wind-solar hybrid power generation system has the problems of unstable maximum power point tracking process and low tracking efficiency.
Disclosure of Invention
The wind-solar hybrid power generation system aims to solve the technical problems of the existing wind-solar hybrid power generation system. The invention provides an MPPT control system and method based on a radial basis function neural network.
The technical scheme for solving the technical problems is as follows:
in a first aspect, an embodiment of the invention provides a radial basis function neural network-based wind-solar power generation MPPT control method, which is used for controlling a wind power generator or/and a photovoltaic power generation module in a wind-solar hybrid power generation system, and is characterized in that the method comprises the following steps:
Step 1: acquiring the output voltage of a power generation device in real time, and judging whether the output voltage of the power generation device is greater than the threshold voltage U of the power generation device or notyif yes, executing the step 2, otherwise, repeating the step 1;
step 2: establishing an RBF neural network prediction model, and predicting the output voltage U of a Boost circuit corresponding to the maximum power point of the power generation device according to the RBF neural network prediction modelaAnd adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the power generation device to be Ua；
And step 3:Detecting the output current and the output voltage of the power generation device according to a preset period interval, and calculating the output power P of the power generation device in the current periodkAnd the output power P of the previous periodk-1And judging | Pk-Pk-1if the result is that | ═ 0 is true, if yes, the procedure goes to step 4, and if not, the procedure goes to step 5;
And 4, step 4: adjusting the duty ratio of the Boost circuit through the PWM control module, and adjusting the output voltage of the power generation device according to a preset condition;
And 5: judgment of | Pk-Pk-1|＜e1If yes, go to step 6, if no, go to step 7, wherein e1a first threshold value for a preset power;
Step 6: judgment of | Pk-Pk-1|＜e2If yes, entering step 7, otherwise, determining that the delta U is a disturbance voltage difference corresponding to a small step length, and entering step 8, wherein e2A preset power second threshold;
And 7: determining the delta U as a disturbance voltage difference corresponding to the large step length, and entering the step 8;
And 8: making the output voltage of the power generation device be Uk＝Uk-1plus or minus delta U, calculating the output power of the power generation device and returning to the step 3; and the delta U is the disturbance voltage difference corresponding to the small step length or the disturbance voltage difference corresponding to the large step length.
further, the step 2 specifically includes:
Step 21: obtaining an output voltage x of the power generation device1＝Uioutput current x2＝Iiand obtaining a first secondary parameter x of the power generation device3And a second secondary parameter x4Using the output voltage x of said power generating means1＝Uioutput current x2＝IiAnd the first secondary parameter x3and the second sub-parameter X4 obtains an input vector X ═ X (X) of the RBF neural network prediction model1,x2,x3,x4)；
Step 22: according to the formula y (k) ═ h1w1+h2w2+…+hiwiobtaining the output power y (k) of the power generation device, wherein H ═ H1,h2,…,hi]Predicting a radial basis vector of the model for the RBF neural network; w ═ W1,w2,…,wi]Ta network weight vector between a radial base layer and an output layer of the RBF neural network prediction model is obtained;
step 23: using formula Ua＝WXTTraining to form an RBF neural network model;
step 24: predicting the voltage at the maximum power point of the power generation device by using the RBF neural network model, and adjusting the duty ratio of the Boost circuit through a PWM control module so as to adjust the output voltage of the power generation device to the voltage at the maximum power point;
Step 25: detecting the voltage and current of the output of the power generation device; and calculating the actual power;
Step 26: and adjusting the duty ratio of the Boost circuit by sending a disturbance signal to a PWM control module so as to adjust the actual output voltage of the power generation device and perform maximum power point tracking of the power generation device.
further, the step 4 specifically includes:
Step 41: judge Uk＞Uk-1If yes, go to step 42; if not, go to step 43;
step 42: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of the power generation device isand returning to the step 3 after calculating the output power of the power generation device, wherein n is the step progressive rate;
Step 43: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of the power generation device isAnd calculating the output power of the power generation device and returning to the step 3, wherein n is the step progressive rate.
Further, the step 8 specifically includes the following steps:
Step 81, judging Pk＞Pk-1If yes, go to step 82, otherwise go to step 83;
Step 82, judge Uk＞Uk-1if the output voltage is U-delta U, adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the power generation device to be U-delta U; if not, adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage U of the power generation device to be U + delta U;
step 83, judge Uk＞Uk-1If the output voltage U is equal to U + delta U, the duty ratio of the Boost circuit is adjusted through a PWM control module, and the output voltage U of the power generation device is enabled to be U + delta U; if not, enabling the output voltage U of the power generation device to be U-delta U;
Further, the step of adjusting the duty ratio of the Boost circuit by the PWM control module to adjust the output voltage of the power generation device to the voltage at the maximum power point specifically includes:
setting an initial value of a duty ratio output by the PWM control module to the BOOST circuit;
Judging an equivalent resistance calculation formula of the Boost circuitWhether the duty ratio is established or not is judged, if yes, the power generation device is output to be at the maximum power point, and if not, the initial value of the duty ratio is updated and then judgment is continued; wherein D in the calculation formula of the equivalent resistance of the Boost circuit is the duty ratio of the Boost circuit, U represents the output voltage of the power generation device, I represents the output current of the power generation device, and V represents the output current of the power generation deviceBRepresenting the output voltage of the Boost circuit, IBRepresenting the output current, R, of the Boost circuitLExpressed as the internal resistance of the power plant.
Furthermore, the power generation device is a solar panel, the first sub-parameter is the solar radiation amount L, and the second sub-parameter is the temperature T of the photovoltaic solar panel.
further, the power generation device is a wind driven generator, the first secondary parameter is air humidity, and the second secondary parameter is air temperature.
In a second aspect, the invention provides a radial basis function neural network based wind-solar power generation MPPT control system controlled by the radial basis function neural network based wind-solar power generation MPPT control method, the system comprising: photovoltaic power generation subsystem, wind power generation subsystem and direct current bus, photovoltaic power generation subsystem includes: the photovoltaic power generation system comprises a photovoltaic power generation module, a first current sensor, a first voltage sensor, a photovoltaic Boost circuit, a photovoltaic PWM control module and a first MPPT control module; the first current sensor is used for collecting output current of the photovoltaic power generation module, the first voltage sensor is used for collecting output voltage of the photovoltaic power generation module, the input end of the photovoltaic BOOST circuit is electrically connected with the output end of the photovoltaic power generation module, the output end of the photovoltaic BOOST circuit is connected with the direct current bus, and the first MPPT control module is electrically connected with the wind power PWM control module, the second current sensor and the second voltage sensor; the wind power generation subsystem includes: the wind power generation system comprises wind power generation, a second current sensor, a second voltage sensor, a wind power Boost circuit, a wind power PWM control module and a second MPPT control module; the second current sensor is used for collecting the output current of a rectifier of the wind driven generator, the second voltage sensor is used for collecting the output voltage of the rectifier of the wind driven generator, the wind power BOOST electric input end is electrically connected with the output end of the rectifier of the wind driven generator, the output end of the BOOST circuit is connected with the direct current bus, and the second MPPT control module is electrically connected with the wind power PWM control module, the second current sensor and the second voltage sensor.
In the technical scheme of the wind-solar power generation MPPT control method based on the radial basis function neural network, a maximum power point prediction model is established by adopting the RBF neural network to predict the maximum power point of a photovoltaic power generation electronic system and the maximum power point of a wind power generation electronic system, and the duty ratio of a photovoltaic Boost circuit and the wind power Boost circuit is controlled to realize maximum power point tracking, so that the photovoltaic power generation electronic system and the wind power generation electronic system in the wind-solar power generation system can independently carry out maximum power point tracking, and the damage of a wind power generator and a photovoltaic power generation module caused by the mutual interference of the photovoltaic power generation electronic system and the wind power generation electronic system is prevented; the technical scheme of the invention not only can realize the rapid tracking of the maximum power point of the output current of the photovoltaic power generation electronic system and the wind power generation electronic system in the wind-solar hybrid power generation system, but also reduces the current oscillation in the maximum power point tracking process, and improves the stability and the tracking efficiency of the maximum power point tracking process of the wind-solar hybrid power generation system.
drawings
FIG. 1 is a schematic flow chart of a wind-solar power generation MPPT control method based on a radial basis function neural network according to an embodiment of the present invention;
FIG. 2 is a structural block diagram of a wind-solar power generation MPPT control system based on a radial basis function neural network according to an embodiment of the present invention;
FIG. 3 is a graph of voltage, current, power and battery power at the battery side of a wind-solar hybrid system according to a conventional disturbance observation method;
FIG. 4 is a graph of voltage, current, power and battery level on the battery side of the wind-solar hybrid system according to the present invention;
FIG. 5 is a graph comparing the output power of the system with the temperature change of the photovoltaic power generation system according to the present invention and the disturbance observation method;
FIG. 6 is a graph comparing the output power of the system for changing the solar illuminance of a photovoltaic power generation system according to the present invention and a disturbance observation method;
FIG. 7 is a graph comparing the output power of the wind speed variation system of the wind power generation system according to the present invention and the disturbance observation method.
The reference numbers illustrate: the photovoltaic power generation system comprises a photovoltaic power generation subsystem 1, a wind power generation subsystem 2, a direct current bus 3, a photovoltaic power generation module 11, a first current sensor 12, a first voltage sensor 13, a photovoltaic Boost circuit 14, a photovoltaic PWM control module 15, a first MPPT control module 16, a wind driven generator 21, a rectifier 211, a second current sensor 22, a second voltage sensor 23, a wind power Boost circuit 24, a wind power PWM control module 25 and a second MPPT control module 26.
Detailed Description
the principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
In the present embodiment, to solve the above-mentioned technical problem, as shown in fig. 2, the wind-solar hybrid power generation system includes: photovoltaic power generation subsystem (1) and wind power generation subsystem (2) and direct current bus (3), the photovoltaic power generation subsystem includes: the photovoltaic power generation system comprises a photovoltaic power generation module (11), a first current sensor (12), a first voltage sensor (13), a photovoltaic Boost circuit (14), a photovoltaic PWM control module (15) and a first MPPT control module (16); the first current sensor (12) is used for collecting output current of the photovoltaic power generation module (11), the first voltage sensor (13) is used for collecting output voltage of the photovoltaic power generation module (11), the input end of the photovoltaic BOOST circuit (14) is electrically connected with the output end of the photovoltaic power generation module (11), the output end of the photovoltaic BOOST circuit (14) is connected with the direct current bus (3), and the first MPPT control module (16) is electrically connected with the wind power PWM control module (15), the second current sensor (12) and the second voltage sensor (13); the wind power generation subsystem (2) comprises: the wind power generation system comprises a wind driven generator (21), a second current sensor (22), a second voltage sensor (23), a wind power Boost circuit (24), a wind power PWM control module (25) and a second MPPT control module (26); the second current sensor (22) is used for collecting the output current of a rectifier (211) of the wind driven generator (21), the second voltage sensor (23) is used for collecting the output voltage of the rectifier (211) of the wind driven generator (21), the input end of the wind power BOOST circuit (24) is electrically connected with the output end of the rectifier (211) of the wind driven generator (21), the output end of the BOOST circuit is connected with the direct current bus (3), and the second MPPT control module (26) is electrically connected with the wind power PWM control module (25), the second current sensor (22) and the second voltage sensor (23).
as shown in fig. 1, an embodiment of the present invention provides a radial basis function neural network-based wind-solar power generation MPPT control method, for controlling a photovoltaic power generation subsystem of a wind-solar hybrid power generation system shown in fig. 2, where the method includes:
Step 1: acquiring the output voltage of a solar cell panel in real time, and judging whether the output voltage of the solar cell panel is greater than the threshold voltage U of the solar cell panely(1)If yes, executing the step 2, otherwise, repeating the step 1;
Step 2: establishing an RBF neural network prediction model, predicting the output voltage of a Boost circuit corresponding to the maximum power point of the solar cell panel according to the RBF neural network prediction model, and adjusting the duty ratio of the Boost circuit through a PWM (pulse width modulation) control module to enable the output voltage of the solar cell panel to be Ua(1)；
and step 3: detecting the output current and the output voltage of the solar panel according to a preset period interval, and calculating the output power P of the solar panel in the current periodk(1)And the output power P of the previous periodk-1(1)And judging | Pk(1)-Pk-1(1)if the result is that | ═ 0 is true, if yes, the procedure goes to step 4, and if not, the procedure goes to step 5;
And 4, step 4: the duty ratio of the Boost circuit is adjusted through the PWM control module, and the output voltage of the solar panel is adjusted according to a preset condition;
And 5: judgment of | Pk(1)-Pk-1(1)|＜e1(1)If yes, go to step 6, if no, go to step 7, wherein e1(1)A first threshold value for a preset power;
Step 6: judgment of | Pk(1)-Pk-1(1)|＜e2(1)If yes, go to step 7, otherwise, determine Δ U(1)The disturbance voltage difference corresponding to the small step length and step 8 is entered, wherein e2(1)a preset power second threshold;
And 7: determining Δ U(1)the disturbance voltage difference corresponding to the large step size is obtained, and the step 8 is carried out;
And 8: making the output voltage of the solar cell panel be Uk(1)＝Uk-1(1)±ΔU(1)And after calculating the output power of the solar panel, returning to the step 3; wherein, Delta U(1)And the disturbance voltage difference is the disturbance voltage difference corresponding to the small step length or the disturbance voltage difference corresponding to the large step length.
Specifically, the step 2 specifically includes:
Step 21: obtaining an output voltage x of the solar panel1(1)＝Ui(1)output current x2(1)＝Ii(1)as main parameters, and obtaining the solar radiation x of the solar panel3(1)Temperature x of the solar cell panel4(1)obtaining an input vector X of the RBF neural network prediction model by using the two main parameters and the two auxiliary parameters(1)＝(x1(1),x2(1),x3(1),x4(1))；
Step 22: according to the formula y (k) ═ h1w1+h2w2+…+hiwiObtaining the output power y (k) of the solar cell panel, wherein H ═ H1,h2,…,hi]Is a radial basis vector of the RBF neural network; w ═ W1,w2,…,wi]TA network weight vector between a radial base layer and an output layer of the RBF neural network;
In this step, an input layer, a radial base layer, and an output layer are determined in order. The input layer comprises the output voltage x of the solar panel1(1)＝Ui(1)Output current x of solar cell panel2(1)＝Ii(1)The radial basis function is α ═ radbas (| | ω -p | | | b), p is the input vector of the neuron, in this embodiment, the input vector X (1))＝(x1(1),x2(1),x3(1),x4(1)) (ii) a Omega is the weight vector of the input; b is a threshold value. Performing learning and numerical processing on the radial base layer part; input Ui(1)Less than a critical set value Uy(1)Judging that the light is insufficient, not starting the tracking process of the maximum power point, and inputtingUi(1)greater than a predetermined threshold value Uy(1)And judging that the light is sufficient, and starting maximum power point tracking. By y (k) h1w1+h2w2+…+hiwiderived output power y (k), where W ═ W1,w2,…,wi]TFor the weight vector of the output layer of the radial base layer, the actual voltage U at the maximum power point is obtained according to the maximum value of the output power y (k)o(1)And when the estimated voltage is compared with the estimated voltage and is smaller than the error value, the requirement is met.
Step 23: using formula Ua＝WXTTraining to form an RBF neural network model;
Step 24: and predicting the voltage at the maximum power point of the power generation device by using the RBF neural network model, and adjusting the duty ratio of the Boost circuit by using a PWM (pulse width modulation) control module to adjust the output voltage of the power generation device to the voltage at the maximum power point, wherein the voltage at the maximum power point of the power generation device is the output voltage of the Boost circuit corresponding to the maximum power point of the power generation device.
Step 25: detecting the voltage and current of the output of the solar cell panel; and calculating the actual power Po(1)；
Step 26: and sending a disturbance signal to a PWM control module to adjust the actual output voltage of the solar panel and carry out maximum power point tracking of the solar panel.
Further, the step 4 specifically includes:
Step 41: judge Uk(1)＞Uk-1(1)If yes, go to step 42; if not, go to step 43;
Step 42: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of the solar panel isAnd calculating the output power of the solar panel and returning to the step 3, wherein n is the step progressive rate;
Step 43: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of the solar cell panel isAnd calculating the output power of the solar panel and returning to the step 3, wherein n is the step progressive rate.
Specifically, the step 8 specifically includes the following steps:
step 81, judge whether P isk(1)＞Pk-1(1)if yes, go to step 82, otherwise go to step 83;
Step 82, judge Uk(1)＞Uk-1(1)If the output voltage is U, adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the solar cell panel to be U(1)＝U(1)-ΔU(1)(ii) a If not, the duty ratio of the Boost circuit is adjusted through a PWM control module, so that the output voltage U of the solar cell panel is enabled(1)＝U(1)+ΔU(1)；
Step 83, determine if Uk(1)＞Uk-1(1)and adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the solar cell panel to be U(1)＝U(1)+ΔU(1)(ii) a Otherwise, make it be U(1)＝U(1)-ΔU(1)；
specifically, the step of adjusting the duty ratio of the Boost circuit by the PWM control module to adjust the output voltage of the power generation device to the voltage at the maximum power point specifically includes:
Setting an initial value of a duty ratio output by the PWM control module to the BOOST circuit;
Judging an equivalent resistance calculation formula of the Boost circuitWhether the duty ratio is established or not is judged, if yes, the solar panel is output to be at the maximum power point, and if not, the initial value of the duty ratio is updated and then judgment is continued;wherein D in the equivalent resistance calculation formula of the Boost circuit(1)is the duty cycle, U, of the BOOST circuit(1)Representing the output voltage of the solar panel, I(1)Represents the output current, V, of the solar panelBRepresenting the output voltage of the Boost circuit, I(1)representing the output current, R, of the Boost circuitL(1)expressed as the solar panel internal resistance.
in the technical scheme of the wind-solar power generation MPPT control method based on the radial basis function neural network, a maximum power point prediction model is established by adopting the RBF neural network to predict the maximum power point of a photovoltaic power generation electronic system and the maximum power point of a wind power generation electronic system, and the duty ratio of a photovoltaic Boost circuit and the wind power Boost circuit is controlled to realize maximum power point tracking, so that the photovoltaic power generation electronic system and the wind power generation electronic system in the wind-solar power generation system can independently carry out maximum power point tracking, and the damage of a wind power generator and a photovoltaic power generation module caused by the mutual interference of the photovoltaic power generation electronic system and the wind power generation electronic system is prevented; the technical scheme of the invention not only can realize the rapid tracking of the maximum power point of the output current of the photovoltaic power generation electronic system and the wind power generation electronic system in the wind-solar hybrid power generation system, but also reduces the current oscillation in the maximum power point tracking process, and improves the stability and the tracking efficiency of the maximum power point tracking process of the wind-solar hybrid power generation system.
As shown in fig. 2, an embodiment of the present invention provides a wind-solar power generation MPPT control method based on a radial basis function neural network, for controlling a wind power generation subsystem, where the method includes:
Step 1: acquiring the output voltage of a rectifier of a wind driven generator in real time, and judging whether the output voltage of the rectifier of the wind driven generator is greater than the threshold voltage U of the rectifier of the wind driven generatory(2)if yes, executing the step 2, otherwise, repeating the step 1;
Step 2: establishing an RBF neural network prediction model, and predicting the output electricity of a Boost circuit corresponding to the maximum power point of a rectifier of the wind driven generator according to the RBF neural network prediction modeland regulating the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of a rectifier of the wind driven generator to be Ua(2)；
And step 3: detecting the output current and the output voltage of the rectifier of the wind driven generator according to a preset period interval, and calculating the output power P of the rectifier of the wind driven generator in the current periodk(2)And the output power P of the previous periodk-1(2)and judging | Pk(2)-Pk-1(2)If the result is that | ═ 0 is true, if yes, the procedure goes to step 4, and if not, the procedure goes to step 5;
And 4, step 4: the duty ratio of the Boost circuit is adjusted through the PWM control module, and the output voltage of a rectifier of the wind driven generator is adjusted according to a preset condition;
And 5: judgment of | Pk(2)-Pk-1(2)|＜e1(2)If yes, go to step 6, if no, go to step 7, wherein e1(2)A first threshold value for a preset power;
step 6: judgment of | Pk(2)-Pk-1(2)|＜e2(2)If yes, go to step 7, otherwise, determine Δ U(2)The disturbance voltage difference corresponding to the small step length and step 8 is entered, wherein e2(2)A preset power second threshold;
and 7: determining Δ U(2)The disturbance voltage difference corresponding to the large step size is obtained, and the step 8 is carried out;
And 8: order post
The output voltage of the rectifier of the wind driven generator is Uk(2)＝Uk-1(2)±ΔU(2)And after calculating the output power of the rectifier of the wind driven generator, returning to the step 3; wherein, Delta U(2)And the disturbance voltage difference is the disturbance voltage difference corresponding to the small step length or the disturbance voltage difference corresponding to the large step length.
Specifically, the step 2 specifically includes:
Step 21: obtaining an output voltage x of a rectifier of the wind turbine1(2)＝Ui(2)Output current x2(2)＝Ii(2)As a main parameter, and obtaining the solar radiation x of the rectifier of the wind power generator3(2)temperature x of a rectifier of said wind turbine4(2)Obtaining an input vector X of the RBF neural network prediction model by using the two main parameters and the two auxiliary parameters(2)＝(x1(2),x2(2),x3(2),x4(2))；
step 22: according to the formula y (k) ═ h1w1+h2w2+…+hiwiObtaining the output power y (k) of the rectifier of the wind driven generator, wherein H ═ H1,h2,…,hi]is a radial basis vector of the RBF neural network; input vector X(2)＝(x1(2),x2(2),x3(2),x4(2))；W＝[w1,w2,…,wi]TA network weight vector between a radial base layer and an output layer of the RBF neural network;
In this step, an input layer, a radial base layer, and an output layer are determined. The input layer comprises an open circuit voltage Ui(2)Open circuit current Ii(2)the radial base layer function is alpha ═ radbas (| | | omega-p | | | b), p is an input vector of a neuron, omega is an input weight vector, b is a threshold, and learning and numerical processing are performed on the radial base layer part; input Ui(2)Less than a critical set value Uy(2)Judging that the wind energy is insufficient, not starting the tracking process of the maximum power point, and inputting Ui(2)greater than a predetermined threshold value Uy(2)and judging that the wind energy is sufficient, and starting maximum power point tracking. By y (k)(2)＝h1w1+h2w2+…+hiwiderived output power y (k)(2)wherein W ═ W1,w2,…,wi]TFor the radial base layer output layer weight vector, according to the output power y (k)(2)the maximum value of (d) yields the actual voltage U at the maximum power pointo(2)And when the estimated voltage is compared with the estimated voltage and is smaller than the error value, the requirement is met.
Step 23: using sample data composed of two said main parameters and two said sub-parameters according to formula：Ua(2)＝WXTTraining to form an RBF neural network model;
Step 24: predicting the voltage at the maximum power point of the power generation device by using the RBF neural network model, and adjusting the duty ratio of the Boost circuit through a PWM control module so as to adjust the output voltage of the power generation device to the voltage at the maximum power point; the "voltage at the maximum power point of the power generation device" is the aforementioned "output voltage of the Boost circuit corresponding to the maximum power point of the power generation device".
Step 25: detecting a voltage U of an output of a rectifier of the wind turbineo(2)And current Io(2)(ii) a And calculating the actual power Po(2)；
step 26: and sending a disturbance signal to a PWM control module to regulate the actual output voltage of the rectifier of the wind driven generator and carry out maximum power point tracking on the rectifier of the wind driven generator.
Further, the step 4 specifically includes:
Step 41: judge Uk(2)＞Uk-1(2)If yes, go to step 42; if not, go to step 43;
Step 42: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of a rectifier of the wind driven generator iscalculating the output power of a rectifier of the wind driven generator and returning to the step 3, wherein n is the step progressive rate;
Step 43: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of a rectifier of the wind driven generator isand calculating the output power of the rectifier of the wind driven generator and returning to the step 3, wherein n is the step progressive rate.
Specifically, the step 8 specifically includes the following steps:
Step 81, judge whether P isk(2)＞Pk-1(2)If yes, go to step 82, otherwise go to step 83;
Step 82, judge Uk(2)＞Uk-1(2)If the output voltage is U, adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of a rectifier of the wind driven generator to be U(2)＝U(2)-ΔU(2)(ii) a If not, the duty ratio of the Boost circuit is adjusted through a PWM control module, so that the output voltage U of the rectifier of the wind driven generator(2)＝U(2)+ΔU(2)；
Step 83, determine if Uk(2)＞Uk-1(2)And adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of a rectifier of the wind driven generator to be U(2)＝U(2)+ΔU(2)(ii) a Otherwise, make it be U(2)＝U(2)-ΔU(2)；
specifically, the step of adjusting the duty ratio of the Boost circuit by the PWM control module to adjust the output voltage of the power generation device to the voltage at the maximum power point specifically includes:
Setting an initial value of a duty ratio output by the PWM control module to the BOOST circuit;
judging an equivalent resistance calculation formula of the Boost circuitWhether the duty ratio is established or not is judged, if yes, the rectifier of the wind driven generator is output to be at the maximum power point, and if not, the initial value of the duty ratio is updated and then judgment is continued; wherein D in the equivalent resistance calculation formula of the Boost circuit(2)Is the duty cycle, U, of the BOOST circuit(2)representing the rectifier output voltage, I, of said wind generator(2)Representing the output current, V, of the rectifier of said wind generatorB(2)Representing the output voltage of the Boost circuit, IB(2)Representing the output current, R, of the Boost circuitL(2)Is expressed as the windThe rectifier internal resistance of the force generator.
In the technical scheme of the wind-solar power generation MPPT control method based on the radial basis function neural network, a maximum power point prediction model is established by adopting the RBF neural network to predict the maximum power point of a photovoltaic power generation electronic system and the maximum power point of a wind power generation electronic system, and the duty ratio of a photovoltaic Boost circuit and the wind power Boost circuit is controlled to realize maximum power point tracking, so that the photovoltaic power generation electronic system and the wind power generation electronic system in the wind-solar power generation system can independently carry out maximum power point tracking, and the damage of a wind power generator and a photovoltaic power generation module caused by the mutual interference of the photovoltaic power generation electronic system and the wind power generation electronic system is prevented; the technical scheme of the invention not only can realize the rapid tracking of the maximum power point of the output current of the photovoltaic power generation electronic system and the wind power generation electronic system in the wind-solar hybrid power generation system, but also reduces the current oscillation in the maximum power point tracking process, and improves the stability and the tracking efficiency of the maximum power point tracking process of the wind-solar hybrid power generation system.
Referring to fig. 3 to 7, fig. 3 is a graph illustrating voltage, current, power and battery capacity at the battery side of a wind-solar hybrid system according to a conventional disturbance observation method. FIG. 4 is a graph of voltage, current, power and battery level on the battery side of the wind-solar hybrid system according to the present invention. FIG. 5 is a graph comparing the output power of the system with the temperature variation of the photovoltaic power generation system according to the present invention and the disturbance observation method. Fig. 6 is a comparison graph of the output power of the system for changing the solar illuminance of the photovoltaic power generation system according to the invention and the disturbance observation method. FIG. 7 is a graph comparing the output power of the wind speed variation system of the wind power generation system according to the present invention and the disturbance observation method. The output voltage corresponding to the maximum power point can be predicted through a model of the RBF neural network, the duty ratio corresponding to the maximum power point output can be predicted when the maximum power point tracking is started, and the problems of low tracking speed and maximum power point oscillation caused by the fact that a disturbance observation method selects the step length are solved. The RBF neural network model directly predicts the output voltage at the maximum power point, reduces the tracking step so as to reduce the tracking time, increases the asymptotic rate so as to reduce the disturbance step length so as to reduce the power oscillation at the maximum power point.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (8)

1. a wind-solar power generation MPPT control method based on a radial basis function neural network is used for controlling a wind driven generator or/and a photovoltaic power generation module in a wind-solar hybrid power generation system, and is characterized by comprising the following steps:
Step 1: acquiring the output voltage of a power generation device in real time, and judging whether the output voltage of the power generation device is greater than the threshold voltage U of the power generation device or notyIf yes, executing the step 2, otherwise, repeating the step 1;
step 2: establishing an RBF neural network prediction model, and predicting the output voltage U of a Boost circuit corresponding to the maximum power point of the power generation device according to the RBF neural network prediction modelaand adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the power generation device to be Ua；
And step 3: detecting the output current and the output voltage of the power generation device according to a preset period interval, and calculating the output power P of the power generation device in the current periodkAnd the output power P of the previous periodk-1And judging | Pk-Pk-1If the result is that | ═ 0 is true, if yes, the procedure goes to step 4, and if not, the procedure goes to step 5;
and 4, step 4: adjusting the duty ratio of the Boost circuit through the PWM control module, and adjusting the output voltage of the power generation device according to a preset condition;
And 5: judgment of | Pk-Pk-1|＜e1If yes, go to step 6, if no, go to step 7, wherein e1A first threshold value for a preset power;
Step 6: judgment of | Pk-Pk-1|＜e2If yes, go to step 7, if not, determine Δ U as the disturbance voltage difference corresponding to the small step length,And proceeding to step 8, wherein e2A preset power second threshold;
And 7: determining the delta U as a disturbance voltage difference corresponding to the large step length, and entering the step 8;
And 8: making the output voltage of the power generation device be Uk＝Uk-1plus or minus delta U, calculating the output power of the power generation device and returning to the step 3; and the delta U is the disturbance voltage difference corresponding to the small step length or the disturbance voltage difference corresponding to the large step length.
2. The radial basis function neural network-based wind-solar power generation MPPT control method according to claim 1, characterized in that the step 2 specifically comprises:
Step 21: obtaining an output voltage x of the power generation device1＝Uioutput current x2＝IiAnd obtaining a first secondary parameter x of the power generation device3And a second secondary parameter x4Using the output voltage x of said power generating means1＝Uioutput current x2＝Iiand the first secondary parameter x3And the second secondary parameter x4Obtaining an input vector X ═ X (X) of the RBF neural network prediction model1,x2,x3,x4)；
step 22: according to the formula y (k) ═ h1w1+h2w2+…+hiwiObtaining the output power y (k) of the power generation device, wherein H ═ H1,h2,…,hi]Predicting a radial basis vector of the model for the RBF neural network; w ═ W1,w2,…,wi]TA network weight vector between a radial base layer and an output layer of the RBF neural network prediction model is obtained;
step 23: using formula Ua＝WXTTraining to form an RBF neural network model;
Step 24: predicting the voltage at the maximum power point of the power generation device by using the RBF neural network model, and adjusting the duty ratio of the Boost circuit through a PWM control module so as to adjust the output voltage of the power generation device to the voltage at the maximum power point;
step 25: detecting the voltage and current of the output of the power generation device; and calculating the actual power;
Step 26: and adjusting the duty ratio of the Boost circuit by sending a disturbance signal to a PWM control module so as to adjust the actual output voltage of the power generation device and perform maximum power point tracking of the power generation device.
3. the radial basis function neural network-based wind-solar power generation MPPT control method according to claim 1, characterized in that the step 4 specifically comprises:
Step 41: judge Uk＞Uk-1If yes, go to step 42; if not, go to step 43;
step 42: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of the power generation device isAnd returning to the step 3 after calculating the output power of the power generation device, wherein n is the step progressive rate;
step 43: the duty ratio of the Boost circuit is adjusted through the PWM control module, so that the output voltage of the power generation device isAnd calculating the output power of the power generation device and returning to the step 3, wherein n is the step progressive rate.
4. the radial basis function neural network-based wind-solar power generation MPPT control method of claim 1,
The method is characterized in that the step 8 specifically comprises the following steps:
Step 81, judging Pk＞Pk-1if yes, go to step 82, otherwise go to step 83;
Step 82, judge Uk＞Uk-1If the output voltage is U-delta U, adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the power generation device to be U-delta U; if not, adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the power generation device to be U + delta U;
step 83, judge Uk>Uk-1If the output voltage is U + delta U, adjusting the duty ratio of the Boost circuit through a PWM control module to enable the output voltage of the power generation device to be U + delta U; and if not, setting the output voltage of the power generation device 15 to be U-delta U.
5. The wind-solar power generation MPPT control method based on the radial basis function neural network as claimed in claim 2, wherein the step of adjusting the duty ratio of the Boost circuit through the PWM control module to adjust the output voltage of the power generation device to the voltage at the maximum power point specifically comprises:
Setting an initial value of a duty ratio output by the PWM control module to the BOOST circuit;
Judging an equivalent resistance calculation formula of the Boost circuitWhether the duty ratio is established or not is judged, if yes, the power generation device is output to be at the maximum power point, and if not, the initial value of the duty ratio is updated and then judgment is continued; wherein D in the calculation formula of the equivalent resistance of the Boost circuit is the duty ratio of the Boost circuit, U represents the output voltage of the power generation device, I represents the output current of the power generation device, and V represents the output current of the power generation deviceBrepresenting the output voltage of the Boost circuit, IBRepresenting the output current, R, of the Boost circuitLRepresenting the internal resistance of the power generation device.
6. The radial basis function neural network-based wind-solar power generation MPPT control method according to claim 2, characterized in that the power generation device is a solar panel, the first sub-parameter is a solar radiation quantity L, and the second sub-parameter is a temperature T of the photovoltaic solar panel.
7. The radial basis function neural network-based wind-solar power generation MPPT control method of claim 2, wherein the power generation device is a wind power generator, the first sub-parameter is air humidity, and the second sub-parameter is air temperature.
8. a radial basis function neural network based wind-solar power generation MPPT control system controlled by the method of any one of claims 1 to 7, the system comprising: photovoltaic power generation subsystem (1), wind power generation subsystem (2) and direct current generating line (3), photovoltaic power generation subsystem includes: the photovoltaic power generation system comprises a photovoltaic power generation module (11), a first current sensor (12), a first voltage sensor (13), a photovoltaic Boost circuit (14), a photovoltaic PWM control module (15) and a first MPPT control module (16); the first current sensor (12) is used for collecting output current of the photovoltaic power generation module (11), the first voltage sensor (13) is used for collecting output voltage of the photovoltaic power generation module (11), the input end of the photovoltaic BOOST circuit (14) is electrically connected with the output end of the photovoltaic power generation module (11), the output end of the photovoltaic BOOST circuit (14) is connected with the direct current bus (3), and the first MPPT control module (16) is electrically connected with the wind power PWM control module (15), the second current sensor (12) and the second voltage sensor (13); the wind power generation subsystem (2) comprises: the wind power generation system comprises a wind driven generator (21), a second current sensor (22), a second voltage sensor (23), a wind power Boost circuit (24), a wind power PWM control module (25) and a second MPPT control module (26); the second current sensor (22) is used for collecting the output current of a rectifier (211) of the wind driven generator (21), the second voltage sensor (23) is used for collecting the output voltage of the rectifier (211) of the wind driven generator (21), the input end of the wind power BOOST circuit (24) is electrically connected with the output end of the rectifier (211) of the wind driven generator (21), the output end of the BOOST circuit is connected with the direct current bus (3), and the second MPPT control module (26) is electrically connected with the wind power PWM control module (25), the second current sensor (22) and the second voltage sensor (23).
CN201910836914.4A 2019-09-05 2019-09-05 wind-solar power generation MPPT control method and system based on radial basis function neural network Pending CN110571853A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910836914.4A CN110571853A (en) 2019-09-05 2019-09-05 wind-solar power generation MPPT control method and system based on radial basis function neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910836914.4A CN110571853A (en) 2019-09-05 2019-09-05 wind-solar power generation MPPT control method and system based on radial basis function neural network

Publications (1)

Publication Number Publication Date
CN110571853A true CN110571853A (en) 2019-12-13

Family

ID=68778068

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910836914.4A Pending CN110571853A (en) 2019-09-05 2019-09-05 wind-solar power generation MPPT control method and system based on radial basis function neural network

Country Status (1)

Country Link
CN (1) CN110571853A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112711292A (en) * 2021-03-29 2021-04-27 深圳黑晶光电技术有限公司 Photovoltaic module maximum power tracking method, system and storage medium
CN113364082A (en) * 2021-05-26 2021-09-07 国网江西省电力有限公司供电服务管理中心 Efficient MPPT control system for cooperative collection of multiple micro-energies

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102231560A (en) * 2011-06-24 2011-11-02 南开大学 Intelligent distributed maximum generation power point tracking coordination control system
CN108491026A (en) * 2018-05-04 2018-09-04 武汉工程大学 The progressive variable step MPPT system and methods of ladder
CN108614612A (en) * 2018-04-24 2018-10-02 青岛高校信息产业股份有限公司 Solar-energy photo-voltaic cell maximum power tracing method and system
CN109634350A (en) * 2019-01-31 2019-04-16 福州大学 A method of it adjusting duty ratio based on neural network and realizes maximal power tracing

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102231560A (en) * 2011-06-24 2011-11-02 南开大学 Intelligent distributed maximum generation power point tracking coordination control system
CN108614612A (en) * 2018-04-24 2018-10-02 青岛高校信息产业股份有限公司 Solar-energy photo-voltaic cell maximum power tracing method and system
CN108491026A (en) * 2018-05-04 2018-09-04 武汉工程大学 The progressive variable step MPPT system and methods of ladder
CN109634350A (en) * 2019-01-31 2019-04-16 福州大学 A method of it adjusting duty ratio based on neural network and realizes maximal power tracing

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
崔岩等: ""基于神经网络应用的光伏阵列最大功率点跟踪"", 《系统仿真技术》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112711292A (en) * 2021-03-29 2021-04-27 深圳黑晶光电技术有限公司 Photovoltaic module maximum power tracking method, system and storage medium
CN112711292B (en) * 2021-03-29 2021-07-09 深圳黑晶光电技术有限公司 Photovoltaic module maximum power tracking method, system and storage medium
CN113364082A (en) * 2021-05-26 2021-09-07 国网江西省电力有限公司供电服务管理中心 Efficient MPPT control system for cooperative collection of multiple micro-energies

Similar Documents

Publication Publication Date Title
CN109888834B (en) Photovoltaic power generation system based on improved MPPT method and storage battery self-balancing quick charging coupling control
Reisi et al. Classification and comparison of maximum power point tracking techniques for photovoltaic system: A review
Ngan et al. A study of maximum power point tracking algorithms for stand-alone photovoltaic systems
CN101783621B (en) Global maximum power point tracking method of photovoltaic generating system and system device
CN106125817B (en) A kind of photovoltaic MPPT methods based on photovoltaic cell four parameter model
CN110571853A (en) wind-solar power generation MPPT control method and system based on radial basis function neural network
CN104871388A (en) Control apparatus, method, and program, and natural energy generation apparatus provided with control apparatus, method, and program
Anu Shalini et al. Hybrid power generation forecasting using CNN based BILSTM method for renewable energy systems
CN103138347B (en) Charge control method of photovoltaic charge controller
CN106787866A (en) The method that photovoltaic system extends maximum power tracing run time
Wongsathan et al. Optimal hybrid neuro-fuzzy based controller using MOGA for photovoltaic (PV) battery charging system
CN114069774B (en) Photovoltaic energy storage system for tracking maximum power generated by photovoltaic cell and photovoltaic power generation system based on system
Ali et al. Evaluating the performance and efficiency of MPPT algorithm for PV systems
Eydi et al. A novel strategy of maximum power point tracking for photovoltaic panels based on fuzzy logic algorithm
CN109217357A (en) A kind of grid-connected photovoltaic system MPPT method based on Markov model
Hong et al. A model-free control strategy for battery energy storage with an application to power accommodation
Shen et al. Neural network predictive control for vanadium redox flow battery
Rami Reddy et al. Chimp optimized fractional-order tilt integral derivative controller (NFO-TIDC) assisted battery for EV charging
CN105870964A (en) Microgrid-based distributed photovoltaic power generation boost (Boost) control method
Mahjoub et al. Control of Hybrid Renewable Energy System Based on MPPT Strategy Technique
Kadhem et al. Improve the energy efficiency of PV systems by installing a soft switching boost converter with MPPT control
CN111193267A (en) Extra-high voltage alternating current and direct current voltage stability control method based on cloud computing
Wu et al. Health Monitoring for Capacitors in Superbuck Converter through LSTM Neural Network
Masenge et al. Modeling and control of solar PV with battery energy storage for rural electrification
KR102243645B1 (en) Battery charge control method and battery charge control system using new renewable energy

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20191213