CN119808608A - A method and device for identifying partial shading of photovoltaic modules - Google Patents

Abstract

The application belongs to the technical field of photovoltaic power generation and machine learning, and relates to a method and a device for identifying partial shielding of a photovoltaic module, wherein an IV curve equation of each battery piece in the photovoltaic module is obtained, a circuit characteristic equation of the photovoltaic module is obtained based on the IV curve equation and a connection relation of all the battery pieces, irradiance proportions of different battery pieces in the photovoltaic module are changed to simulate different shielding states, a gray level map of the photovoltaic module in each shielding state is generated based on irradiance proportions of each battery piece in the photovoltaic module in each shielding state, an IV curve of the photovoltaic module in each shielding state is obtained based on the circuit characteristic equation, the IV curve and the gray level map of the photovoltaic module in each shielding state are respectively used as a sample and a label thereof, a training set is constructed, and a deep learning model is trained, so that a real-time gray level map is obtained based on real-time IV curves of the photovoltaic module, and the shielded battery pieces in the photovoltaic module are identified. The scheme provided by the application not only improves the detection efficiency, but also greatly reduces the detection cost.

Description

Photovoltaic module local shielding recognition method and device

Technical Field

The invention relates to the technical field of photovoltaic power generation and machine learning, in particular to a method and a device for identifying partial shielding of a photovoltaic module.

Background

The photovoltaic module in the photovoltaic power plant is connected by a plurality of battery pieces and constitutes, and when the battery piece was sheltered from by substances such as dust, fallen leaves, snow, the illumination intensity who reaches its surface reduces, leads to the battery piece to be in dark state, and the electric current that its produced can obviously be less than the battery piece that does not sheltered from, causes battery piece mismatch problem, influences photovoltaic module's generated energy. Meanwhile, the voltage of the battery piece which is not shielded can form reverse bias voltage at two ends of the battery piece which is shielded, so that the local temperature of the photovoltaic module is increased, a hot spot effect is generated, and even a fire disaster is caused. In addition, the mismatching of the battery pieces can accelerate the aging of the photovoltaic module and reduce the service life of the photovoltaic power station.

The method for detecting the partial shielding of the photovoltaic module in the prior art comprises the steps of monitoring the current and the voltage of the photovoltaic module in real time, comparing the current and the voltage with the voltage and the current in a fault state, judging whether the battery piece is shielded or not in the photovoltaic module, wherein the method can only detect the whole photovoltaic module, cannot accurately acquire the specific position of the shielded battery piece, and can only determine the shielding position by further checking the whole photovoltaic module. The other is to apply high-definition camera shooting or unmanned aerial vehicle to combine with an image processing algorithm to acquire an infrared thermal imaging image of the photovoltaic module, identify the infrared thermal imaging image and judge whether the infrared thermal imaging image has a local shielding problem, and the method can position the shielding position in the photovoltaic module, but for a large-area photovoltaic power station, a large number of high-definition cameras are required to be erected or unmanned aerial vehicle arrays are required to be used for detection due to the large number of battery pieces, so that the detection is long in time and high in energy consumption, and the maintenance cost of a large number of camera shooting devices is high.

In summary, the existing method for identifying the partial shielding of the photovoltaic module has the problems of high detection cost and low detection efficiency.

Disclosure of Invention

Therefore, the invention aims to solve the technical problems of high detection cost and low detection efficiency of the partial shielding identification method of the photovoltaic module in the prior art.

In order to solve the technical problems, the invention provides a method for identifying partial shielding of a photovoltaic module, which comprises the following steps:

acquiring an IV curve equation of each battery piece in the photovoltaic module, and acquiring a circuit characteristic equation of the photovoltaic module based on the IV curve equation and the connection relation of all the battery pieces in the photovoltaic module;

The irradiance proportion of different battery pieces in the photovoltaic module is changed to simulate different shielding states of the photovoltaic module;

Generating a gray scale image of the photovoltaic module in each shielding state based on irradiance proportion of each battery piece in the photovoltaic module in each shielding state, and acquiring an IV curve of the photovoltaic module in each shielding state based on a circuit characteristic equation;

And respectively taking the IV curve and the gray level map of the photovoltaic module in each shielding state as a sample and a label thereof, constructing a training set and training a deep learning model so as to obtain a real-time gray level map based on the real-time IV curve of the photovoltaic module, thereby identifying the shielded battery piece in the photovoltaic module.

Preferably, obtaining the IV curve equation for each cell in the photovoltaic module includes:

normalizing the electrical parameters of the photovoltaic module based on the number of the series battery pieces and the number of the parallel battery pieces in the photovoltaic module to obtain the electrical parameters of the single battery piece;

constructing a residual matrix based on the electrical parameters of the single battery piece and the equivalent port current equation of the double diodes of the battery piece and the constraint conditions of the equivalent port current equation;

Solving a residual matrix to obtain a first saturated dark current, a second saturated dark current, a series resistor and a parallel resistor of a double-diode equivalent model of a single battery piece;

Inputting the first saturated dark current, the second saturated dark current, the series resistor, the parallel resistor and the reverse breakdown voltage of the photovoltaic module into a reverse breakdown curve equation of each battery piece, and then fitting the reverse breakdown curve equation with the reverse breakdown curve of each battery piece in the photovoltaic module to obtain the reverse breakdown coefficient of each battery piece;

An IV curve equation for each cell is derived based on the reverse breakdown coefficient of each cell.

Preferably, the residual matrix is expressed as:

,

Wherein,Representing a residual matrix; Representing photo-generated current; Representing the short circuit current of a single cell; representing a first saturated dark current; Representing the series resistance; Representing the thermal voltage; Representing a second saturated dark current; representing parallel resistances; Represents the open circuit voltage of an individual cell; Representing the dual diode equivalent port current of the battery piece; representing the dual diode equivalent port voltage of the battery piece; representing the equivalent port power of the double diodes of the battery piece; representing the maximum power of the battery piece;

The reverse breakdown curve equation for the cell is expressed as:

,

Wherein,A reverse breakdown curve equation of the battery piece is represented;、、 respectively representing a first reverse breakdown coefficient, a second reverse breakdown coefficient and a third reverse breakdown coefficient of the battery piece; representing equivalent saturated dark current when the battery piece breaks down reversely; representing the reverse breakdown voltage of the photovoltaic module.

Preferably, generating the gray scale map of the photovoltaic module in each shielding state based on the irradiance ratio of each cell in the photovoltaic module in each shielding state includes:

acquiring light intensity data of each battery piece in each shielding state based on irradiance proportion of each battery piece in the photovoltaic module in each shielding state;

Generating a light intensity diagram of the photovoltaic module in each shielding state based on the light intensity data of all the battery pieces in each shielding state;

And converting the light intensity graph of the photovoltaic module in each shielding state to obtain the gray level graph of the photovoltaic module in each shielding state.

Preferably, generating the gray scale map of the photovoltaic module in each shielding state based on the irradiance ratio of each cell in the photovoltaic module in each shielding state includes:

Calculating the product of the irradiance proportion of each battery piece in the photovoltaic module in each shielding state and the preset irradiance proportion and the gray value mapping coefficient to obtain the gray value of each battery piece in each shielding state;

and filling the gray value of each battery piece under each shielding condition into a two-dimensional array based on the position of each battery piece in the photovoltaic module to obtain a gray image of the photovoltaic module under each shielding state.

Preferably, obtaining the IV curve of the photovoltaic module in each shielding state based on the circuit characteristic equation includes:

calculating the photo-generated current of each battery piece in each shielding state based on the irradiance proportion of each battery piece in the photovoltaic module in each shielding state;

Inputting the electrical parameters of the photovoltaic module and the photo-generated current of all the battery pieces in each shielding state into a circuit characteristic equation, changing the voltage value in the circuit characteristic equation, and calculating to obtain a current value corresponding to each voltage value;

and obtaining an IV curve of the photovoltaic module based on each voltage value and the corresponding current value.

Preferably, the deep learning model includes a variational self-encoder and an adaptive pooling layer.

Preferably, obtaining the real-time gray scale map based on the real-time IV curve of the photovoltaic module, so as to identify the blocked battery piece in the photovoltaic module includes:

Inputting the real-time IV curve into a variation self-encoder and outputting a predicted gray level diagram;

inputting the predicted gray level map into the adaptive pooling layer for up-sampling or down-sampling, and outputting a real-time gray level map;

And acquiring gray values of all pixel points in the real-time gray map, and identifying the blocked battery pieces based on the gray values of the pixel points corresponding to each battery piece.

Preferably, inputting the predicted gray map into the adaptive pooling layer for upsampling or downsampling comprises:

If the size of the predicted gray level map is smaller than the preset size, upsampling the predicted gray level map by using a bilinear interpolation method;

And if the size of the predicted gray level map is larger than the preset size, downsampling the predicted gray level map by using an average pooling method.

The invention also provides a device for identifying partial shielding of the photovoltaic module, which comprises the following steps:

The circuit characteristic equation acquisition module is used for acquiring an IV curve equation of each battery piece in the photovoltaic module, and acquiring a circuit characteristic equation of the photovoltaic module based on the IV curve equation and the connection relation of all the battery pieces in the photovoltaic module;

The shielding simulation module is used for simulating the shielded state of different battery pieces in the photovoltaic module by changing the irradiance proportion of the different battery pieces in the photovoltaic module;

The training data generation module is used for generating a gray level graph of the photovoltaic module in each shielding state based on irradiance proportion of each battery piece in the photovoltaic module in each shielding state, and acquiring an IV curve of the photovoltaic module in each shielding state based on a circuit characteristic equation;

The model training and shielding identification module is used for respectively taking an IV curve and a gray level diagram of the photovoltaic module in each shielding state as a sample and a label thereof, constructing a training set and training a deep learning model so as to obtain a real-time gray level diagram based on a real-time IV curve of the photovoltaic module, thereby identifying a shielded battery piece in the photovoltaic module.

The method for identifying the partial shielding of the photovoltaic module has the following beneficial effects:

Because the irradiance proportion is 1 when the cell is not shielded, the irradiance proportion gradually decreases along with the increase of the shielded area, so that the shielding condition of the cell in the photovoltaic module can be measured by the irradiance proportion of the cell. According to the method, the irradiance proportion of different battery pieces in the photovoltaic module is changed to simulate different shielding states of the photovoltaic module, and the change of the irradiance proportion can lead to the change of gray values and currents of the battery pieces, namely, the gray level diagram and the IV curve of the photovoltaic module in the different shielding states are different, so that the circuit characteristic equation of the photovoltaic module is obtained firstly based on the IV curve equation and the connection relation of the battery pieces, the gray level diagram in the different shielding states is generated based on irradiance proportion mapping of the battery pieces in the photovoltaic module in the different shielding states, meanwhile, the IV curve of the photovoltaic module in the different shielding states is obtained based on the circuit characteristic equation, a training set is built based on the IV curve and the gray level diagram in the different shielding states, and a deep learning model is trained, so that the correlation between the gray level diagram and the IV curve is learned, the real-time gray level diagram can be obtained based on the real-time IV curve of the photovoltaic module, and the shielded battery pieces can be identified. The method can identify the specific position of the shielded battery piece only by collecting the IV curve of the photovoltaic module during detection, does not need to collect and analyze the voltage and the current of each battery piece, does not need to rely on hardware equipment such as a camera, an unmanned aerial vehicle and the like to acquire the image of the whole photovoltaic module, improves the detection efficiency, greatly reduces the detection cost, and can be directly used for guiding a cleaning robot or manually checking the battery piece of the photovoltaic module based on the detection result obtained by the method, so that the power generation loss of a photovoltaic power station is reduced, and the service life is prolonged.

Drawings

In order that the invention may be more readily understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings, in which:

FIG. 1 is a flow chart of a method for identifying partial shielding of a photovoltaic module;

FIG. 2 is a schematic diagram of a deep learning model structure provided by the application;

FIG. 3 is a schematic diagram of a dual diode equivalent model of a photovoltaic module comprising a plurality of cells connected together and a single cell according to the present application;

fig. 4 is a schematic structural diagram of a photovoltaic module partial shielding recognition device provided by the application;

FIG. 5 is a schematic diagram of comparing the prediction result and the actual measurement result of the partial shielding recognition of the photovoltaic module with smaller shielding proportion by using the method provided by the application, wherein (a) in FIG. 5 is a schematic diagram of a first group of actual IV curve of the photovoltaic module with smaller shielding proportion, (b) in FIG. 5 is a schematic diagram of a second group of actual IV curve of the photovoltaic module with smaller shielding proportion, (c) in FIG. 5 is a schematic diagram of a third group of actual IV curve of the photovoltaic module with smaller shielding proportion, (d) in FIG. 5 is a gray level diagram of the prediction result of the partial shielding recognition of the photovoltaic module with smaller shielding proportion, e) in FIG. 5 is a gray level diagram of the prediction result of the partial shielding recognition of the photovoltaic module with smaller shielding proportion, f in FIG. 5 is a gray level diagram of the prediction result of the partial shielding recognition of the photovoltaic module with smaller shielding proportion, g in FIG. 5 is a gray level diagram of the partial shielding result of the photovoltaic module with smaller shielding proportion, h in FIG. 5 is a gray level diagram of the partial shielding result of the photovoltaic module with smaller shielding proportion, and i is a gray level diagram of the partial shielding result of the photovoltaic module with smaller shielding proportion;

Fig. 6 is a schematic diagram of comparing a prediction result and an actual measurement result of local shielding recognition of a plurality of groups of photovoltaic modules with larger shielding proportion by using the method provided by the application, wherein (a) in fig. 6 is a schematic diagram of a first group of actual IV curves of the photovoltaic modules with larger shielding proportion, (b) in fig. 6 is a schematic diagram of a second group of actual IV curves of the photovoltaic modules with larger shielding proportion, (c) in fig. 6 is a schematic diagram of a third group of actual IV curves of the photovoltaic modules with larger shielding proportion, (d) in fig. 6 is a gray level diagram of a prediction result of local shielding recognition of the photovoltaic modules with larger shielding proportion in the first group, (e) in fig. 6 is a gray level diagram of a prediction result of local shielding recognition of the photovoltaic modules with larger shielding proportion in the second group, f) in fig. 6 is a gray level diagram of a prediction result of local shielding recognition of the photovoltaic modules with larger shielding proportion in the third group, g in fig. 6 is a gray level diagram of local shielding result of the photovoltaic modules with larger shielding proportion in the first group, h in fig. 6 is a gray level diagram of local shielding result of the photovoltaic modules with larger shielding proportion in the third group, and i is a gray level diagram of local gray level diagram of the local shielding result of the photovoltaic modules with larger shielding result in the third group.

Detailed Description

The present invention will be further described with reference to the accompanying drawings and specific examples, which are not intended to be limiting, so that those skilled in the art will better understand the invention and practice it.

Referring to fig. 1, fig. 1 is a flowchart of a method for identifying partial shielding of a photovoltaic module, which specifically includes:

and S10, acquiring an IV curve equation of each battery piece in the photovoltaic module, and acquiring a circuit characteristic equation of the photovoltaic module based on the IV curve equation and the connection relation of all the battery pieces in the photovoltaic module.

Specifically, based on the IV curve equation of all the battery pieces and the series-parallel connection relation among the battery pieces, the circuit characteristic equation of the photovoltaic module can be obtained by combining kirchhoff's law and ohm's law, and further, based on the series-parallel connection relation among the photovoltaic modules in the photovoltaic power station and the circuit characteristic equation of each photovoltaic module, the circuit characteristic equation of the photovoltaic power station can be obtained.

S20, simulating different shielding states of the photovoltaic module by changing irradiance ratios of different battery pieces in the photovoltaic module.

Specifically, when the cell is not blocked, the irradiance ratio E is 1, and as the blocked area increases gradually, the irradiance ratio E decreases gradually, so that the blocking condition of the photovoltaic module can be represented by the irradiance ratio E of each cell. The photovoltaic module under different local shielding conditions can be simulated by setting irradiance proportion of the battery pieces at different positions.

For example, optical simulation software such as Solar Pro, PVsyst and COMSOL Multiphysics can be used for constructing a model of the photovoltaic module, and various shielding scenes can be simulated by setting irradiance ratios of different battery slices in the photovoltaic module.

S30, generating a gray scale map of the photovoltaic module in each shielding state based on irradiance proportion of each battery piece in the photovoltaic module in each shielding state, and acquiring an IV curve of the photovoltaic module in each shielding state based on a circuit characteristic equation.

Specifically, the smaller the cell gray value with larger irradiance ratio, the larger the cell gray value with smaller irradiance ratio, so that the irradiance ratio of all cells in the photovoltaic module in each shielding state can be mapped into the gray values of different pixels in the gray map.

Optionally, in some embodiments of the present application, a gray scale map may be randomly generated, and the gray scale map is reversely mapped to irradiance ratio of each battery piece in the photovoltaic module, so as to calculate IV curves of the photovoltaic module, so as to quickly establish a training set, which is worth noting that 10000 IV curve data can be calculated every 15 minutes in average.

And S40, respectively taking the IV curve and the gray level map of the photovoltaic module in each shielding state as a sample and a label thereof, constructing a training set and training a deep learning model so as to obtain a real-time gray level map based on the real-time IV curve of the photovoltaic module, thereby identifying the shielded battery piece in the photovoltaic module.

Alternatively, in some embodiments of the present application, a series of voltage values and current values of the battery sheet may be collected, the collected voltage values and current values may be plotted on a two-dimensional coordinate plane, and data points on the two-dimensional coordinate plane may be connected into a smooth curve by a data fitting method, so as to obtain an IV curve equation of the battery sheet.

However, there is a larger error in the method for drawing the IV curve and obtaining the IV curve equation by data collection, so in the embodiment of the present application, another method for obtaining the IV curve equation of the battery sheet is further provided, which specifically includes:

and 1, normalizing the electrical parameters of the photovoltaic module based on the number of the series battery pieces and the number of the parallel battery pieces in the photovoltaic module to obtain the electrical parameters of the single battery piece.

Specifically, the electrical parameter of the photovoltaic module is measured under standard test conditions (STANDARD TEST Condition, STC).

In some embodiments of the application, the electrical parameters include short circuit current, open circuit voltage, maximum power point voltage, and maximum power point current.

Specifically, the calculation formula of the electrical parameters of the individual battery cells is expressed as:

,

,

,

,

Wherein,Representing the short circuit current of a single cell; Representing a short circuit current of the photovoltaic module; representing the number of parallel battery pieces; Represents the open circuit voltage of an individual cell; representing an open circuit voltage of the photovoltaic module; representing the number of the battery pieces connected in series; representing the maximum power point current of a single battery piece; representing the maximum power point current of the photovoltaic module; Representing the maximum power point voltage of a single battery piece; representing the maximum power point voltage of the photovoltaic module.

And 2, constructing a residual matrix based on the electrical parameters of the single battery piece and the equivalent port current equation of the double diodes of the battery piece and the constraint conditions of the equivalent port current equation.

Specifically, the dual diode equivalent port current equation for the battery plate is:

,

Wherein,Representing the dual diode equivalent port current of the battery piece; representing the dual diode equivalent port voltage of the battery piece; Representing photo-generated current; representing a first saturated dark current; Representing the series resistance; Representing the thermal voltage; Representing a second saturated dark current; representing parallel resistances;、 Is a two-diode ideality factor,,。

The constraint conditions include:

1. short circuit constraint conditions:,;

2. Open circuit constraint conditions:,;

3、 Wherein, the method comprises the steps of,Representing the equivalent port power of the double diodes of the battery piece; representing the maximum power of the battery piece;

4、。

Further, the residual matrix is expressed as:

。

And step 3, solving the residual matrix to obtain a first saturated dark current, a second saturated dark current, a series resistor and a parallel resistor of a double-diode equivalent model of the single battery piece.

In a specific example of the present application, the residual matrix is solved by using a Powell method, and the first saturated dark current, the second saturated dark current, the series resistance and the parallel resistance obtained by solving are processed so that the orders of magnitude of the parameters are equal, specifically, a formula for processing the parameters is expressed as follows:。

And 4, inputting the first saturated dark current, the second saturated dark current, the series resistor, the parallel resistor and the reverse breakdown voltage of the photovoltaic module into a reverse breakdown curve equation of each battery piece, and then fitting the reverse breakdown curve equation with the reverse breakdown curve of each battery piece in the photovoltaic module to obtain the reverse breakdown coefficient of each battery piece.

Specifically, the reverse breakdown curve equation of the battery sheet is expressed as:

,

Wherein,A reverse breakdown curve equation of the battery piece is represented;、、 respectively representing a first reverse breakdown coefficient, a second reverse breakdown coefficient and a third reverse breakdown coefficient of the battery piece; representing equivalent saturated dark current when the battery piece breaks down reversely; representing the reverse breakdown voltage of the photovoltaic module.

And 5, obtaining an IV curve equation of each battery piece based on the reverse breakdown coefficient of each battery piece.

Further, in some embodiments of the present application, an optical simulation software may be directly used to create a three-dimensional model of the photovoltaic module, and light intensity data of each battery piece in different shielding states is obtained through simulation, and the light intensity data obtained through simulation is converted into a gray value, so as to obtain a gray map of the photovoltaic module in different local shielding conditions, which specifically includes:

acquiring light intensity data of each battery piece in each shielding state based on irradiance proportion of each battery piece in the photovoltaic module in each shielding state;

Generating a light intensity diagram of the photovoltaic module in each shielding state based on the light intensity data of all the battery pieces in each shielding state;

And converting the light intensity graph of the photovoltaic module in each shielding state to obtain the gray level graph of the photovoltaic module in each shielding state.

Alternatively, in other embodiments of the present application, the irradiance ratio of each cell in each shielding state may also be directly mapped to a gray value, so as to construct a gray map of the photovoltaic module in each shielding state based on the gray values of all the cells, which specifically includes:

Calculating the product of the irradiance proportion of each battery piece in the photovoltaic module in each shielding state and the preset irradiance proportion and the gray value mapping coefficient to obtain the gray value of each battery piece in each shielding state;

and filling the gray value of each battery piece under each shielding condition into a two-dimensional array based on the position of each battery piece in the photovoltaic module to obtain a gray image of the photovoltaic module under each shielding state.

For example, the gray value G is usually between 0 and 255, and the irradiance ratio E of the battery cells is usually between 0 and 1, so that the gray value of the pixel at the position of the battery cell with the irradiance ratio E of 1 may be set to 255, and the calculation formula of the gray value G of the pixel at each battery cell may beTherefore, the gray level graph of the photovoltaic module can be obtained based on the gray level values of the pixel points where all the battery pieces are located.

Further, obtaining IV curves for the photovoltaic module in each occlusion state based on the circuit characteristic equation includes:

calculating the photo-generated current of each battery piece in each shielding state based on the irradiance proportion of each battery piece in the photovoltaic module in each shielding state;

Inputting the electrical parameters of the photovoltaic module and the photo-generated current of all the battery pieces in each shielding state into a circuit characteristic equation, changing the voltage value in the circuit characteristic equation, and calculating to obtain a current value corresponding to each voltage value;

and obtaining an IV curve of the photovoltaic module based on each voltage value and the corresponding current value.

Further, in some embodiments of the application, as shown in FIG. 2, the deep learning model includes a variational self-encoder (Variance Auto Encoder, VAE) and an adaptive pooling layer.

Specifically, the application uses the transfer learning technology, namely, the pre-trained variation self-encoder is utilized to train the new training set, and the picture size in the new training set is possibly inconsistent with the picture size in the pre-training process, so that in order to ensure that the picture size output by the model on the new training set is consistent with the preset size, the application is connected with an adaptive pooling layer after the variation self-encoder, and the gray level image output by the variation self-encoder is up-sampled or down-sampled, so that the gray level image size output by the model on different data sets can be equal to the preset size, and the battery piece shielding recognition precision is improved. Meanwhile, the variation self-encoder is pre-trained, so that the model can achieve a high-precision recognition result only by training more than ten rounds on different training sets, and the training time of the model is shortened.

It should be noted that, in the model training process, the preset size refers to the size of the label in the training set, and in the model prediction process, the preset size is set based on the correspondence between the battery pieces in the photovoltaic module and the pixels in the gray scale image, for example, fig. 3 shows a 72 x 2 photovoltaic module provided by the present application, where each battery piece may be equivalently a dual diode equivalent model, each three pixels is preset to correspond to one battery piece, and the pixel position in the gray scale image corresponds to the relative position of the battery piece and the photovoltaic module, and when the photovoltaic module is identified by shielding, the preset size of the gray scale image is 216 x 432.

Further, in some embodiments of the present application, training the deep learning model in step S40 specifically includes:

And acquiring samples in the training set and labels of the samples, inputting the samples into a variable self-encoder, and outputting an initial prediction gray level map.

And inputting the initial predicted gray level map into the self-adaptive pooling layer, and up-sampling or down-sampling the initial predicted gray level map based on the size of the initial predicted gray level map and the size of the label, so as to output the target predicted gray level map.

And constructing a loss function based on the target prediction gray level graph and the label, and performing iterative training on the variation self-encoder and the self-adaptive pooling layer by utilizing samples in the training set until the value of the loss function converges or reaches the preset iteration times, so as to obtain a trained deep learning model.

In a specific example of the application, the preset iteration number is 10-30 during model training.

Further, in step S40, a real-time gray scale map is obtained based on the real-time IV curve of the photovoltaic module, so that the specific implementation manner of identifying the blocked battery piece in the photovoltaic module includes:

and inputting the real-time IV curve into a variation self-encoder and outputting a predicted gray level diagram.

And inputting the predicted gray level map into the adaptive pooling layer for up-sampling or down-sampling, and outputting the real-time gray level map.

And acquiring gray values of all pixel points in the real-time gray map, and identifying the blocked battery pieces based on the gray values of the pixel points corresponding to each battery piece.

Specifically, inputting the predicted gray map into the adaptive pooling layer for upsampling or downsampling comprises:

If the size of the predicted gray level map is smaller than the preset size, upsampling the predicted gray level map by using a bilinear interpolation method;

and if the size of the predicted gray level map is larger than the preset size, downsampling the predicted gray level map by using an average pooling method.

For example, assume that the predicted gray-scale map has a size ofThe preset size isIf (if),The predicted gray level map is up-sampled, specifically, the real-time gray level map obtained after up-sampling is expressed as:

,

Wherein,Representing a real-time gray scale obtained after up-sampling; representing the row and column values of pixel points in the real-time gray scale map; representing a predicted gray scale map; Representing a rank value of pixel points in the prediction gray level image; Representing the interpolation weights.

If it is,Downsampling the predicted gray scale map, specifically, the real-time gray scale map obtained after downsampling is expressed as:

,

Wherein,Representing a real-time gray scale obtained after downsampling; representing the row and column values of pixel points in the real-time gray scale map; Representing the size of the pooled core.

It should be noted that, the interpolation weight during up-sampling may be automatically calculated by an interpolation method, and the size of the pooling kernel during down-sampling may be automatically adjusted according to the scaling.

Based on the method for identifying the partial shielding of the photovoltaic module provided by the embodiment, the embodiment of the application also provides a device for identifying the partial shielding of the photovoltaic module, as shown in fig. 4, which specifically comprises the following steps:

The circuit characteristic equation obtaining module 10 is configured to obtain an IV curve equation of each battery piece in the photovoltaic module, and obtain a circuit characteristic equation of the photovoltaic module based on the IV curve equation and the connection relationship of all the battery pieces in the photovoltaic module;

The shielding simulation module 20 is used for simulating the shielded state of different battery pieces in the photovoltaic module by changing the irradiance proportion of the different battery pieces in the photovoltaic module;

the training data generating module 30 is configured to generate a gray scale map of the photovoltaic module in each shielding state based on irradiance proportions of each battery piece in the photovoltaic module in each shielding state, and obtain an IV curve of the photovoltaic module in each shielding state based on a circuit characteristic equation;

The model training and shielding identification module 40 is configured to take the IV curve and the gray level map of the photovoltaic module in each shielding state as a sample and a label thereof, construct a training set, and train the deep learning model, so as to obtain a real-time gray level map based on the real-time IV curve of the photovoltaic module, thereby identifying the shielded battery piece in the photovoltaic module.

As shown in fig. 5, a comparison between the prediction result and the actual measurement result of the partial shielding recognition of the photovoltaic modules with smaller shielding proportion by using the method provided by the application is shown, wherein (a) in fig. 5 is a graph showing the actual IV curve of the photovoltaic modules with smaller shielding proportion in the first group, (b) in fig. 5 is a graph showing the actual IV curve of the photovoltaic modules with smaller shielding proportion in the second group, (c) in fig. 5 is a graph showing the actual IV curve of the photovoltaic modules with smaller shielding proportion in the third group, (d) in fig. 5 is a graph showing the prediction result of the partial shielding recognition of the photovoltaic modules with smaller shielding proportion in the first group, and (e) in fig. 5 is a graph showing the prediction result of the partial shielding recognition of the photovoltaic modules with smaller shielding proportion in the second group, and (f) in fig. 5 is a graph showing the prediction result of the partial shielding recognition of the photovoltaic modules with smaller shielding proportion in the third group, and (h) in fig. 5 is a graph showing the partial measurement result of the photovoltaic modules with smaller shielding proportion in the third group.

As shown in fig. 6, a comparison between the prediction result and the actual measurement result of the partial shielding recognition of the photovoltaic modules with larger shielding proportion by using the method provided by the application is shown, wherein (a) in fig. 6 is a graph showing the actual IV curve of the photovoltaic modules with larger shielding proportion in the first group, (b) in fig. 6 is a graph showing the actual IV curve of the photovoltaic modules with larger shielding proportion in the second group, (c) in fig. 6 is a graph showing the actual IV curve of the photovoltaic modules with larger shielding proportion in the third group, (d) in fig. 6 is a graph showing the prediction result of the partial shielding recognition of the photovoltaic modules with larger shielding proportion in the first group, (e) in fig. 6 is a graph showing the prediction result of the partial shielding recognition of the photovoltaic modules with larger shielding proportion in the second group, and (f) in fig. 6 is a graph showing the prediction result of the partial shielding recognition of the photovoltaic modules with larger shielding proportion in the third group, and (h) in fig. 6 is a graph showing the partial shielding result of the photovoltaic modules with larger shielding proportion in the third group.

As can be seen from the multiple groups of comparison results shown in fig. 5 and 6, the method for identifying local shielding based on the IV curve of the photovoltaic module provided by the application has better identification performance on the photovoltaic modules with smaller shielding proportion and larger shielding proportion.

According to the method for identifying the local shielding of the photovoltaic module, when the deep learning model is trained, a large amount of training data can be generated in a short time by generating the gray level diagram and combining with the electrical simulation to quickly construct the training set, so that the labor cost of data acquisition and marking is reduced, when the trained deep learning model is utilized for detection, the shielding condition can be automatically analyzed through an algorithm only by acquiring IV curve data of the photovoltaic module, the hardware cost and the time cost are obviously reduced without additionally erecting a high-definition camera or an unmanned aerial vehicle array, meanwhile, the obtained shielding data can be directly used for guiding a cleaning robot or manual inspection, the power generation loss is reduced, the efficient and stable operation of battery pieces in the photovoltaic module is ensured, the service life of the photovoltaic module is prolonged, and the method has high practical value.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations and modifications of the present invention will be apparent to those of ordinary skill in the art in light of the foregoing description. It is not necessary here nor is it exhaustive of all embodiments. While still being apparent from variations or modifications that may be made by those skilled in the art are within the scope of the invention.

Claims (10)

Translated fromChinese

1.一种光伏组件局部遮挡识别方法，其特征在于，包括：1. A method for identifying partial shading of a photovoltaic module, comprising:获取光伏组件内每个电池片的IV曲线方程，基于光伏组件内所有电池片的IV曲线方程和连接关系，得到光伏组件的电路特性方程；Obtaining the IV curve equation of each cell in the photovoltaic module, and obtaining the circuit characteristic equation of the photovoltaic module based on the IV curve equations and connection relationships of all cells in the photovoltaic module;通过改变光伏组件内不同电池片的辐照度比例，以模拟光伏组件的不同遮挡状态；By changing the irradiance ratio of different cells in the photovoltaic module, different shading states of the photovoltaic module can be simulated;基于每种遮挡状态下光伏组件内各个电池片的辐照度比例，生成每种遮挡状态下光伏组件的灰度图，并基于电路特性方程获取每种遮挡状态下光伏组件的IV曲线；Based on the irradiance ratio of each cell in the photovoltaic module under each shading state, a grayscale image of the photovoltaic module under each shading state is generated, and the IV curve of the photovoltaic module under each shading state is obtained based on the circuit characteristic equation;将每种遮挡状态下光伏组件的IV曲线和灰度图分别作为样本及其标签，构建训练集并训练深度学习模型，以便基于光伏组件的实时IV曲线得到实时灰度图，从而识别光伏组件中被遮挡的电池片。The IV curve and grayscale image of the photovoltaic module under each shading state are used as samples and their labels respectively, and a training set is constructed and a deep learning model is trained so that a real-time grayscale image can be obtained based on the real-time IV curve of the photovoltaic module, thereby identifying the obscured cells in the photovoltaic module.2.根据权利要求1所述的光伏组件局部遮挡识别方法，其特征在于，获取光伏组件内每个电池片的IV曲线方程包括：2. The method for identifying partial shading of a photovoltaic module according to claim 1, wherein obtaining the IV curve equation of each cell in the photovoltaic module comprises:基于光伏组件内的串联电池片数量和并联电池片数量，对光伏组件的电学参数进行归一化，得到单个电池片的电学参数；Based on the number of series-connected cells and the number of parallel-connected cells in the photovoltaic module, the electrical parameters of the photovoltaic module are normalized to obtain the electrical parameters of a single cell;基于单个电池片的电学参数、电池片的双二极管等效端口电流方程及其约束条件，构建残差矩阵；Based on the electrical parameters of a single cell, the dual-diode equivalent port current equation of the cell and its constraints, a residual matrix is constructed;对残差矩阵求解，得到单个电池片的双二极管等效模型的第一饱和暗电流、第二饱和暗电流、串联电阻和并联电阻；Solving the residual matrix, obtaining the first saturated dark current, the second saturated dark current, the series resistance and the parallel resistance of the double-diode equivalent model of a single cell;将第一饱和暗电流、第二饱和暗电流、串联电阻、并联电阻和光伏组件的反向击穿电压输入电池片的反向击穿曲线方程后，分别与光伏组件内每个电池片的反向击穿曲线拟合，得到每个电池片的反向击穿系数；After inputting the first saturated dark current, the second saturated dark current, the series resistance, the parallel resistance and the reverse breakdown voltage of the photovoltaic module into the reverse breakdown curve equation of the cell, they are respectively fitted with the reverse breakdown curve of each cell in the photovoltaic module to obtain the reverse breakdown coefficient of each cell;基于每个电池片的反向击穿系数得到每个电池片的IV曲线方程。The IV curve equation of each battery cell is obtained based on the reverse breakdown coefficient of each battery cell.3.根据权利要求2所述的光伏组件局部遮挡识别方法，其特征在于，残差矩阵表示为：3. The photovoltaic module partial occlusion identification method according to claim 2, characterized in that the residual matrix is expressed as:， ,其中，表示残差矩阵；表示光生电流；表示单个电池片的短路电流；表示第一饱和暗电流；表示串联电阻；表示热电压；表示第二饱和暗电流；表示并联电阻；表示单个电池片的开路电压；表示电池片的双二极管等效端口电流；表示电池片的双二极管等效端口电压；表示电池片的双二极管等效端口功率；表示电池片的最大功率；in, represents the residual matrix; represents the photogenerated current; Indicates the short-circuit current of a single battery cell; represents the first saturation dark current; represents the series resistance; represents thermal voltage; represents the second saturation dark current; represents parallel resistance; Indicates the open circuit voltage of a single cell; Indicates the double-diode equivalent port current of the battery cell; Indicates the double diode equivalent port voltage of the battery cell; Indicates the dual-diode equivalent port power of the battery cell; Indicates the maximum power of the battery cell;电池片的反向击穿曲线方程表示为：The reverse breakdown curve equation of the battery cell is expressed as:， ,其中，表示电池片的反向击穿曲线方程；、、分别表示电池片的第一反向击穿系数、第二反向击穿系数和第三反向击穿系数；表示电池片反向击穿时的等效饱和暗电流；表示光伏组件的反向击穿电压。in, Represents the reverse breakdown curve equation of the battery cell; , , Respectively represent the first reverse breakdown coefficient, the second reverse breakdown coefficient and the third reverse breakdown coefficient of the battery cell; Indicates the equivalent saturation dark current when the battery cell is reversely broken down; Indicates the reverse breakdown voltage of a photovoltaic module.4.根据权利要求1所述的光伏组件局部遮挡识别方法，其特征在于，基于每种遮挡状态下光伏组件内各个电池片的辐照度比例，生成每种遮挡状态下光伏组件的灰度图包括：4. The method for identifying partial shading of a photovoltaic module according to claim 1, characterized in that generating a grayscale image of the photovoltaic module under each shading state based on the irradiance ratio of each cell in the photovoltaic module under each shading state comprises:基于每种遮挡状态下光伏组件内每个电池片的辐照度比例，获取每种遮挡状态下每个电池片的光强数据；Based on the irradiance ratio of each cell in the photovoltaic module under each shading state, the light intensity data of each cell under each shading state is obtained;基于每种遮挡状态下所有电池片的光强数据，生成每种遮挡状态下光伏组件的光强图；Based on the light intensity data of all cells in each shading state, a light intensity map of the photovoltaic module in each shading state is generated;对每种遮挡状态下光伏组件的光强图进行转换，得到每种遮挡状态下光伏组件的灰度图。The light intensity map of the photovoltaic module under each shading state is converted to obtain a grayscale map of the photovoltaic module under each shading state.5.根据权利要求1所述的光伏组件局部遮挡识别方法，其特征在于，基于每种遮挡状态下光伏组件内各个电池片的辐照度比例，生成每种遮挡状态下光伏组件的灰度图包括：5. The method for identifying partial shading of a photovoltaic assembly according to claim 1, characterized in that generating a grayscale image of the photovoltaic assembly under each shading state based on the irradiance ratio of each cell in the photovoltaic assembly under each shading state comprises:计算每种遮挡状态下光伏组件内每个电池片的辐照度比例，与预设的辐照度比例与灰度值映射系数的乘积，得到每种遮挡状态下每个电池片的灰度值；Calculate the irradiance ratio of each cell in the photovoltaic module under each shading state, and multiply the product of the preset irradiance ratio and the gray value mapping coefficient to obtain the gray value of each cell under each shading state;基于光伏组件内每个电池片的位置，将每种遮挡情况下每个电池片的灰度值填充至二维数组中，得到每种遮挡状态下光伏组件的灰度图。Based on the position of each cell in the photovoltaic module, the grayscale value of each cell in each shading condition is filled into a two-dimensional array to obtain a grayscale image of the photovoltaic module in each shading state.6.根据权利要求1所述的光伏组件局部遮挡识别方法，其特征在于，基于电路特性方程获取每种遮挡状态下光伏组件的IV曲线包括：6. The method for identifying partial shading of a photovoltaic module according to claim 1, wherein obtaining the IV curve of the photovoltaic module under each shading state based on the circuit characteristic equation comprises:基于每种遮挡状态下光伏组件内每个电池片的辐照度比例，计算每种遮挡状态下每个电池片的光生电流；Based on the irradiance ratio of each cell in the photovoltaic module under each shading state, the photocurrent of each cell under each shading state is calculated;将光伏组件的电学参数和每种遮挡状态下所有电池片的光生电流输入电路特性方程，改变电路特性方程中的电压值，计算得到每个电压值对应的电流值；The electrical parameters of the photovoltaic module and the photocurrent of all the cells under each shading state are input into the circuit characteristic equation, the voltage value in the circuit characteristic equation is changed, and the current value corresponding to each voltage value is calculated;基于每个电压值及其对应的电流值，得到光伏组件的IV曲线。Based on each voltage value and its corresponding current value, the IV curve of the photovoltaic module is obtained.7.根据权利要求1所述的光伏组件局部遮挡识别方法，其特征在于，深度学习模型包括变分自编码器和自适应池化层。7. The photovoltaic module partial occlusion recognition method according to claim 1 is characterized in that the deep learning model includes a variational autoencoder and an adaptive pooling layer.8.根据权利要求7所述的光伏组件局部遮挡识别方法，其特征在于，基于光伏组件的实时IV曲线得到实时灰度图，从而识别光伏组件中被遮挡的电池片包括：8. The method for identifying partial shading of a photovoltaic assembly according to claim 7, wherein obtaining a real-time grayscale image based on a real-time IV curve of the photovoltaic assembly, thereby identifying the shaded cells in the photovoltaic assembly comprises:将实时IV曲线输入变分自编码器，输出预测灰度图；Input the real-time IV curve into the variational autoencoder and output the predicted grayscale image;将预测灰度图输入自适应池化层进行上采样或下采样，输出实时灰度图；Input the predicted grayscale image into the adaptive pooling layer for upsampling or downsampling, and output a real-time grayscale image;获取实时灰度图中各个像素点的灰度值，基于每个电池片对应像素点的灰度值识别被遮挡的电池片。The grayscale value of each pixel in the real-time grayscale image is obtained, and the obscured battery cell is identified based on the grayscale value of the pixel corresponding to each battery cell.9.根据权利要求8所述的光伏组件局部遮挡识别方法，其特征在于，将预测灰度图输入自适应池化层进行上采样或下采样包括：9. The photovoltaic module partial occlusion recognition method according to claim 8, characterized in that inputting the predicted grayscale image into the adaptive pooling layer for upsampling or downsampling comprises:若预测灰度图的尺寸小于预设尺寸，则利用双线性插值法对预测灰度图进行上采样；If the size of the predicted grayscale image is smaller than the preset size, the predicted grayscale image is upsampled using bilinear interpolation;若预测灰度图的尺寸大于预设尺寸，则利用平均池化法对预测灰度图进行下采样。If the size of the predicted grayscale image is larger than the preset size, the predicted grayscale image is downsampled using the average pooling method.10.一种光伏组件局部遮挡识别装置，其特征在于，包括：10. A photovoltaic module partial shading identification device, comprising:电路特性方程获取模块，用于获取光伏组件内每个电池片的IV曲线方程，基于光伏组件内所有电池片的IV曲线方程和连接关系，得到光伏组件的电路特性方程；A circuit characteristic equation acquisition module is used to acquire the IV curve equation of each cell in the photovoltaic module, and obtain the circuit characteristic equation of the photovoltaic module based on the IV curve equations and connection relationships of all cells in the photovoltaic module;遮挡模拟模块，用于通过改变光伏组件内不同电池片的辐照度比例，以模拟光伏组件内不同电池片被遮挡状态；The shading simulation module is used to simulate the shading state of different cells in the photovoltaic module by changing the irradiance ratio of different cells in the photovoltaic module;训练数据生成模块，用于基于每种遮挡状态下光伏组件内各个电池片的辐照度比例，生成每种遮挡状态下光伏组件的灰度图，并基于电路特性方程获取每种遮挡状态下光伏组件的IV曲线；A training data generation module is used to generate a grayscale image of the photovoltaic module under each shading state based on the irradiance ratio of each cell in the photovoltaic module under each shading state, and obtain an IV curve of the photovoltaic module under each shading state based on a circuit characteristic equation;模型训练及遮挡识别模块，用于将每种遮挡状态下光伏组件的IV曲线和灰度图分别作为样本及其标签，构建训练集并训练深度学习模型，以便基于光伏组件的实时IV曲线得到实时灰度图，从而识别光伏组件中被遮挡的电池片。The model training and occlusion recognition module is used to take the IV curve and grayscale image of the photovoltaic module under each occlusion state as samples and their labels, build a training set and train a deep learning model, so as to obtain a real-time grayscale image based on the real-time IV curve of the photovoltaic module, thereby identifying the obscured cells in the photovoltaic module.