Photovoltaic panel fault diagnosis method based on multi-source data fusion

Abstract

The invention relates to the technical field of photovoltaic panel fault diagnosis, and discloses a photovoltaic panel fault diagnosis method based on multi-source data fusion, and the method comprises the steps: collecting and marking data, collecting the data from an operation monitoring system and a field detection instrument of a photovoltaic panel, and carrying out the data expansion, and training the feature extraction model, the feature dimension reduction model and the classifier model in sequence, and finally inputting the collected data into the models for diagnosis. According to the method, the sample quality is enhanced by collecting and labeling photovoltaic panel operation data, performing data expansion and de-entanglement processing, feature extraction is performed by adopting a neural network based on adaptive fluctuation optimization, feature dimension reduction is realized by combining a self-encoding neural network of manifold mapping learning, and the classification precision is enhanced through a high-order neural network. According to the method, the accuracy and generalization ability of fault diagnosis are remarkably improved, the stability and efficiency of model training are optimized, a complex data mode is effectively dealt with, and stable operation of the photovoltaic panel in a variable environment is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic panel fault diagnosis, and in particular to a photovoltaic panel fault diagnosis method based on multi-source data fusion. Background Art

[0002] With the rapid development of renewable energy, the stability and efficiency of photovoltaic power generation systems have become important research topics. As the core component of photovoltaic power generation systems, the reliability of photovoltaic panels directly affects the operation of the entire power generation system. However, photovoltaic panels are easily affected by various factors during operation, such as environmental factors and material aging, which may lead to reduced efficiency or failure. Therefore, accurate and rapid diagnosis of the operating status and potential failures of photovoltaic panels is of great significance to ensure the stable operation of the system and improve power generation efficiency.

[0003] The Chinese invention patent with application number CN202011063756.2 proposes a method for diagnosing DC side faults in household photovoltaic systems, which involves the field of power grid operation and maintenance. Most of the DC side fault diagnosis of household photovoltaic systems is based on experimental conditions, with too high requirements on the dimension of data, blindly pursuing accuracy, and ignoring the applicable scenarios. The present invention first finds the three normal users closest to the faulty user, calculates the photocurrent value according to the single diode model formula of the photovoltaic module, and then calculates the power loss time series of the faulty user within a single day, and then uses the least squares method to fit the power loss. Combined with the output characteristics of the photovoltaic array, the fault diagnosis result is obtained according to the fitting result. This technical solution only relies on the readings and installation information of the smart meter installed by the user to complete the analysis and fault diagnosis of household photovoltaic status in a region, effectively reducing the losses caused by photovoltaic failures that cannot be detected by users.

[0004] The Chinese invention patent with application number CN202410580406.5 proposes a photovoltaic module fault detection method, which belongs to the field of photovoltaic power generation, including: establishing a photovoltaic module database; preprocessing the photovoltaic module database, the data preprocessing includes normalization, singular value algorithm denoising, phase space reconstruction and time series reconstruction; configuring the evaluation function of the photovoltaic module to be monitored; optimizing the evaluation function parameters with the preprocessed photovoltaic module database, and completing the configuration of the evaluation fitness function according to the optimization result; generating the first fault monitoring result and the second fault monitoring result, performing two-way trusted authentication, and reporting the fault anomaly according to the two-way trusted authentication result. This application solves the technical problem that the existing photovoltaic module fault detection relies on the photovoltaic module operating parameters and lacks trusted authentication, resulting in low fault detection accuracy. The technical effect of combining data analysis with image recognition through a two-way trusted authentication mechanism to improve the accuracy of fault detection is achieved.

[0005] Based on this, the existing technology has the following deficiencies: conventional data expansion methods may not be able to effectively handle the dependencies between data features, resulting in low quality of generated samples, affecting the generalization ability of the model; the neural network feature extraction methods in the existing technology may have oscillation and local minimum problems during the parameter update process, reducing the training efficiency and model stability; the traditional feature dimensionality reduction method may not be able to retain enough key information during the dimensionality reduction process, affecting the accuracy and reliability of the model in practical applications; the existing classifier model may face gradient explosion or training oscillation problems during the weight update process, limiting the training speed and the final classification accuracy. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the deficiencies in the prior art, the present invention provides a photovoltaic panel fault diagnosis method based on multi-source data fusion, which has the advantages of accurate photovoltaic panel fault diagnosis and solves the above-mentioned technical problems.

[0008] (II) Technical solution

[0009] To achieve the above object, the present invention provides the following technical solution: a photovoltaic panel fault diagnosis method based on multi-source data fusion, comprising the following steps:

[0010] S1. Data collection and annotation. The data is collected from the operation monitoring system of the photovoltaic panel and the on-site detection instrument. The collected data is stored in a structured data standard. The data includes voltage value a1, current value a2, surface temperature a3, light intensity a4, ambient humidity a5, wind speed a6, dust deposition a7, service life a8, maintenance history a9, error code a1, etc. 10 , and manually annotate the collected data;

[0011] S2, data expansion, sample generation based on the SMOTE algorithm of quantum information non-locality;

[0012] S3, inputting the expanded data into a feature extraction model to train the feature extraction model, wherein the feature extraction model extracts features using a neural network feature extraction algorithm based on adaptive fluctuation optimization;

[0013] S4, inputting the feature-extracted data into a feature dimension reduction model to train the feature dimension reduction model, wherein the feature dimension reduction model uses an autoencoder neural network based on manifold mapping learning to perform feature dimension reduction;

[0014] S5, inputting the dimension-reduced data into a classifier model to train the classifier model, wherein the classifier model adopts a high-order neural network based on gradient accumulation;

[0015] S6. Photovoltaic panel data fault diagnosis: The photovoltaic panel data to be tested is input into the feature extraction model for feature extraction, the data dimension reduction is performed through the data dimension reduction model, and the fault type is judged through the classifier model.

[0016] As a preferred technical solution of the present invention, the sample generation based on the SMOTE algorithm of quantum information non-locality in step S2 includes the following steps:

[0017] S2.1. Initialize the algorithm parameters, including the initial superposition angle θ of the quantum state b , entanglement strength ∈ b and adaptively adjusted learning rate λ b The initialization of is as follows:

[0018]

[0019] ∈ b =0.1

[0020] λ b =0.01

[0021] Among them, θ b represents the superposition state angle of the quantum bit, ∈ b represents the degree of entanglement between quantum bits, λ b represents the learning rate;

[0022] S2.2. Use nonlinear transformation to disentangle the features in the original data set. The specific expression is as follows:

[0023] zb=f b (x; Θ b )=tanh(Θ b x)

[0024] Where x is the original input eigenvector, zb is the eigenvector after disentanglement, Θ b is the weight matrix of the neural network, tanh(*) is the hyperbolic tangent function, f b (x; Θ b ) represents a nonlinear transformation function;

[0025] S2.3. By simulating the behavior of quantum superposition and entanglement, a new sample point y in the data space is generated. The generation method of the new sample is expressed as:

[0026] y=x+γ b ·(σ b ·(zb-x))

[0027] In the formula, y represents the generated new sample, γ b represents the parameter that controls the degree to which the new sample deviates from the original sample, σb is a random variable adjusted based on the entanglement degree of the quantum state, x is the original eigenvector of the input, and zb represents the eigenvector after disentanglement;

[0028] S2.4. Automatically adjust algorithm parameters θ after each iteration b ,∈ b and λ b , to optimize the subsequent sample generation process, expressed as:

[0029]

[0030] In the formula, Δθ b is the adjustment amount of the initial superposition state angle of the quantum bit, Δ∈ b is the adjustment amount of entanglement strength, is the initial superposition state angle of the updated quantum bit, is the updated entanglement strength, λ b To adjust the learning rate, and Δθ b and Δ∈ b The expression is as follows:

[0031]

[0032] Among them, η cv is the loss learning rate parameter, and They are the loss function L with respect to θ b and ∈ b The partial derivative of b is the adjustment amount of the initial superposition state angle of the quantum bit, Δ∈ b is the adjustment amount of entanglement strength;

[0033] S2.5. Repeat the above steps until the preset maximum number of iterations is reached.

[0034] As a preferred technical solution of the present invention, the random variable σ in step S2.3 based on the adjustment of the quantum state entanglement degree b According to the entanglement degree of the quantum state ∈ b and the superposition angle θ b Dynamic adjustment, the specific expression is as follows:

[0035] σ b =cos(θ b )·∈ b 2

[0036] Among them, cos(θ b ) represents the influence of the superposition state on the sample generation position, ∈ b 2represents the weight of entanglement in generating samples, σ b It is a random variable adjusted based on the degree of entanglement of the quantum state.

[0037] As a preferred technical solution of the present invention, the process of training the neural network feature extraction algorithm based on adaptive fluctuation optimization in step S3 is as follows:

[0038] S3.1. Define the neural network structure. The number of hidden layers of the neural network is 2. The first hidden layer includes 100 neurons, and the second hidden layer includes 50 neurons. The activation function is the ReLU activation function. At the same time, set the parameters of the neural network, including the initial parameters of the neural network θ0, the wave source S i Location Strength i and the initial phase φ i , the initial parameters of the neural network θ0 include the weights w of the neural network gr and bias b gr ;

[0039] S3.2. Calculate the loss L through forward propagation GR , using the neural network parameters θ on the input X GR Perform forward propagation on the , calculate the output Y and compare it with the target value T, and the expression for calculating the loss is:

[0040] Y=frs(X GR ;θ)

[0041] L GR =L GRoss (Y, T)

[0042] Among them, frs(*) represents the network structure, and the output layer of the network structure uses the preset Softmax_gr classifier to obtain the predicted label, L Gross(Y，T) represents the cross entropy loss;

[0043] S3.3. Calculate the gradient of each parameter θ based on the loss function It can be expressed as:

[0044]

[0045] in, Denotes the loss function L G The gradient of R with respect to the parameter θ, and Represent the loss function L GR The weight w of the neural network gr and the bias b of the neural network gr The partial derivative of

[0046] S3.4. Generate a wave source at the corresponding parameter position according to the direction and size of the gradient. At the t+1th iteration, each wave source S i The strength is updated to The calculation method can be expressed as:

[0047]

[0048] In the formula, Denotes the loss function L GR The gradient of the i-th parameter, α ck is the learning rate, represents the gradient of the i-th parameter, is the attenuation coefficient, Indicates that each wave source S at the tth iteration i The strength of is the amplification factor, L GR represents the loss function;

[0049] S3.5. The fluctuations of all wave sources are superimposed in the parameter space. The update amount of each parameter is determined by the superposition result of the fluctuations of all wave sources at this point. The specific expression is as follows:

[0050]

[0051] Where k is the wave number, is the neural network parameter of the t+1th iteration, represents the neural network parameters of the tth iteration, ω is the angular frequency, which represents the propagation speed and direction of the wave, Re() is the ReLU activation function, and L gr represents the cross entropy loss function, Denote the loss function as L gr The minimum fluctuation threshold value in the historical iteration number of the i-th parameter at time , || is the absolute value symbol, ng represents the total number of wave sources, They represent the intensity of the jth wave source at the t+1th iteration, The loss function is L gr The fluctuation threshold of the i-th parameter at time , its expression is as follows:

[0052]

[0053] Among them, σ0 is the initial fluctuation threshold, For L gr Layer sensitivity index of the layer, κ pq is the adjustment coefficient, e represents the natural constant, where The expression is as follows:

[0054]

[0055] Where, represents the L2 norm, Denotes the loss function L GR The gradient of the kth parameter, L GRc Indicates the total number of layers in the network; Denotes the loss function L G About L G Gradient of the Rc layer parameters; Indicates L GRc Gradients of layer parameters; Denotes the loss function L GR Gradients about the parameters of the kth layer; Represents the gradient of the k-th layer parameter;

[0056] S3.6. Dynamically adjust the attenuation coefficient β and amplification coefficient γ of each wave source. The specific expressions are as follows:

[0057]

[0058] In the formula, are the attenuation coefficient and amplification coefficient of the previous iteration respectively, They are the attenuation coefficient adjustment factor and the amplification coefficient adjustment factor, Respectively represent the attenuation coefficient and the amplification coefficient, ΔL GR is the change in loss, e represents a natural constant;

[0059] S3.7. Determine whether the preset maximum number of iterations has been reached. If not, jump to step S3.2. If reached, stop iteration and complete training.

[0060] As a preferred technical solution of the present invention, the training of the autoencoder neural network algorithm based on manifold mapping learning in step S4 includes the following steps:

[0061] S4.1. Initialize the weight w of the autoencoder using Gaussian distribution cr and bias b cr Parameters, the specific expressions are as follows:

[0062]

[0063] in, represents the weight of the i-th layer, σ ch represents the standard deviation, n” in ” is the number of nodes in the input layer;

[0064] S4.2. During the encoding process, the data x cr After the encoder, it is mapped to a low-dimensional feature space; during the decoding process, the original data is reconstructed by the decoder. The specific expression is as follows:

[0065] Data x cr is mapped to the low-dimensional feature space zr:

[0066]

[0067] The decoder reconstructs the original input x cr ^:

[0068]

[0069] Among them, Sig() is the activation function; and are the weights and biases of the encoder respectively; and are the weights and biases of the decoder;

[0070] S4.3, Riemann factors are used to implement manifold learning. The expression of the features of the Riemann factor R constraint encoding is as follows:

[0071]

[0072] Among them, μ cy represents the preset manifold center, σ R It is a parameter that controls the compactness of the manifold, exp represents an exponential function with a natural constant as the base, and zr represents a low-dimensional feature space;

[0073] S4.4. Construct the loss function Lu, the specific expression is as follows:

[0074] Lu=||x cr -x cr ^|| 2 +λ cm R

[0075] Among them, λ cm is the regularization coefficient, x cr Represents the data before encoding, x cr represents the decoded data, and R represents the Riemann factor;

[0076] S4.5, adjust network parameters through back propagation algorithm;

[0077] S4.6. Determine whether the preset maximum number of iterations has been reached. If not, jump to step S4.2. If reached, stop iteration and complete training.

[0078] As a preferred technical solution of the present invention, the training of the high-order neural network algorithm based on gradient accumulation in step S5 specifically includes the following steps:

[0079] S5.1. Initialize the parameters of the high-order neural network: Use random initialization to initialize the high-order neural network parameters W q and bias b q ;

[0080] S5.2. The data after feature dimension reduction is used as the input feature vector and processed through the network layer. The specific expression is as follows:

[0081] z q =Sig(W q ·x q +b q )

[0082] Among them, x q is the input feature vector, z q is the output of this layer, W q Represents the parameters of high-order neural network, Sig() is the Sigmoid activation function;

[0083] S5.3. Use the loss function Lg to evaluate the error between the classifier model output and the actual label. The specific expression is as follows:

[0084]

[0085] Among them, y true，i is the one-hot encoding of the true label, y pred，i is the probability output predicted by the classifier model, log(*) represents the logarithmic function, Indicates the sum of N data;

[0086] S5.4. Calculate the gradient of high-order neural network parameters and biases based on the loss function L and The specific expression is as follows:

[0087]

[0088] in, is the gradient of the loss function Lg with respect to the output; and is the gradient of the output with respect to the weights and biases;

[0089] S5.5. Accumulate the calculated gradients until the preset training cycle is reached, and the accumulated gradients of the high-order neural network parameters and biases and The expression is as follows:

[0090]

[0091] in, Indicates that the gradient is accumulated at each iteration Indicates that the gradient is accumulated at each iteration

[0092] S5.6, after the iteration of step S5.5 is completed, the accumulated gradient is clipped to limit the maximum value of the gradient. The specific expression is as follows:

[0093]

[0094] Among them, clip(*,-τ q ,τ q ) means clipping the gradient to [-τ q ,τ q ] range, τ q Represents the gradient clipping threshold;

[0095] S5.7. Determine whether the preset maximum number of iterations has been reached. If not, jump to step S5.3. If reached, stop iteration and complete training.

[0096] Compared with the prior art, the present invention provides a photovoltaic panel fault diagnosis method based on multi-source data fusion, which has the following beneficial effects:

[0097] 1. The present invention uses quantum information technology to de-entangle data features and generates high-quality training samples, which significantly improves the generalization ability of the model and its ability to process complex data patterns. The adaptive fluctuation optimization algorithm ensures the smoothness and adaptability of network parameter updates, significantly improving the stability of the model and training efficiency.

[0098] 2. The present invention ensures the retention of important features through effective encoding and decoding processes, enhances the accuracy and reliability of the model in practical applications, and the gradient accumulation strategy of high-order neural networks significantly reduces the oscillation during training and improves the convergence speed and classification accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] Figure 1 It is a schematic diagram of the process of the present invention;

[0100] Figure 2 Schematic diagram of the training process of the feature extraction model of the present invention;

[0101] Figure 3 This is a schematic diagram of the feature dimension reduction model training process of the present invention;

[0102] Figure 4 It is a schematic diagram of the training and application process of the present invention. DETAILED DESCRIPTION

[0103] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0104] See also Figure 1 - Figure 4 , a photovoltaic panel fault diagnosis method based on multi-source data fusion, comprising the following steps:

[0105] S1. Data collection and annotation constitute the original data set, which is collected from the operation monitoring system of the photovoltaic panel and the on-site detection instrument. The collected data is stored in a structured data standard. The data includes voltage value a1, current value a2, surface temperature a3, light intensity a4, ambient humidity a5, wind speed a6, dust deposition a7, service life a8, maintenance history a9, error code a1, etc. 10 , and manually annotate the collected data, with the annotated categories including "normal operation", "efficiency reduction", "local failure", and "comprehensive failure";

[0106] S2, data expansion, sample generation based on the SMOTE algorithm of quantum information non-locality;

[0107] In step S2, the sample generation based on the SMOTE algorithm of quantum information non-locality includes the following steps:

[0108] S2.1. Initialize the algorithm parameters, including the initial superposition angle θ of the quantum state b , entanglement strength ∈ b and adaptively adjusted learning rate λ b The initialization of is as follows:

[0109]

[0110] ∈ b =0.1

[0111] λ b =0.01

[0112] Among them, θ b represents the superposition state angle of the quantum bit, ∈ b represents the degree of entanglement between quantum bits, λ b represents the learning rate;

[0113] S2.2. Use nonlinear transformation to disentangle the features in the original data set. The specific expression is as follows:

[0114] zb=fb (x; Θ b )=tanh(Θ b x)

[0115] Where x is the original input eigenvector, zb is the eigenvector after disentanglement, Θ b is the weight matrix of the neural network, tanh(*) is the hyperbolic tangent function, f b (x; Θ b ) represents a nonlinear transformation function;

[0116] S2.3. By simulating the behavior of quantum superposition and entanglement, a new sample point y in the data space is generated. The generation method of the new sample is expressed as:

[0117] y=x+γ b ·(σ b ·(zx))

[0118] In the formula, y represents the generated new sample, γ b represents the parameter that controls the degree to which the new sample deviates from the original sample, σ b is a random variable based on the adjustment of quantum state entanglement degree, x is the original input eigenvector, zb represents the eigenvector after disentanglement, γ b Set to 3;

[0119] The random variable σ based on the adjustment of quantum state entanglement degree b According to the entanglement degree of the quantum state ∈ b and the superposition angle θ b Dynamic adjustment, the specific expression is as follows:

[0120] σ b =cos(θ b )·∈ b 2

[0121] Among them, cos(θ b ) represents the influence of the superposition state on the sample generation position, ∈ b 2 represents the weight of entanglement in generating samples, σ b It is a random variable adjusted based on the degree of entanglement of the quantum state;

[0122] S2.4. Automatically adjust algorithm parameters θ after each iteration b ,∈ b and λ b , to optimize the subsequent sample generation process, expressed as:

[0123]

[0124] In the formula, Δθ bis the adjustment amount of the initial superposition state angle of the quantum bit, Δ∈ b is the adjustment amount of entanglement strength, is the initial superposition state angle of the updated quantum bit, is the updated entanglement strength, λ b To adjust the learning rate, λ b is set to 0.01, and Δθ b and Δ∈ b The expression is as follows:

[0125]

[0126] Among them, η cv is the loss learning rate parameter, and They are the loss function L with respect to θ b and ∈ b The partial derivative of b is the adjustment amount of the initial superposition state angle of the quantum bit, η cv Set to 0.03, Δ∈ b is the adjustment amount of entanglement strength;

[0127] S2.5, repeat the above steps until the preset maximum number of iterations is reached, and the maximum number of iterations is set to 1000 times;

[0128] S3. Input the expanded data into the feature extraction model to train the feature extraction model. The feature extraction model uses a neural network feature extraction algorithm based on adaptive fluctuation optimization to extract features. Input the expanded data into the feature extraction model to train the feature extraction model. The present invention uses a neural network feature extraction algorithm based on adaptive fluctuation optimization, which is inspired by the fluctuation phenomenon. The fluctuation presents the characteristics of periodic change and energy transfer, and can transmit and transform energy forms in different media. The adaptive fluctuation optimization algorithm iteratively updates the weights and biases in the network by simulating the propagation and energy distribution characteristics of the fluctuation. Each parameter update is not only affected by the current gradient, but also by the cumulative influence of the fluctuation of the previous parameter update, which is similar to the fluctuation influence caused by the superposition of wave sources, making the parameter update process smoother and more adaptive, effectively avoiding the oscillation and local minimum problems in the traditional gradient descent method;

[0129] S3.1. Define the neural network structure. The number of hidden layers of the neural network is 2. The first hidden layer includes 100 neurons, and the second hidden layer includes 50 neurons. The activation function is the ReLU activation function. At the same time, set the parameters of the neural network, including the initial parameters of the neural network θ0, the wave source S i Location Strength i and the initial phase φi , the initial parameters of the neural network θ0 include the weights w of the neural network gr and bias b gr ;

[0130] Initialization can be expressed as:

[0131]

[0132] In the formula, Usually chosen as the initial position of the parameter, a i and φ i Randomly set its initial value; S i is the i-th wave source, with position Strength i and the initial phase φ i composition;

[0133] S3.2. Calculate the loss L through forward propagation GR , using the neural network parameters θ on the input X GR Perform forward propagation on the , calculate the output Y and compare it with the target value T, and the expression for calculating the loss is:

[0134] Y=frs(X GR ;θ)

[0135] L GR =L GRoss (Y, T)

[0136] Among them, frs(*) represents the network structure, and the output layer of the network structure uses the preset Softmax_gr classifier to obtain the predicted label, L Gross(Y，T) represents the cross entropy loss;

[0137] S3.3. Calculate the gradient of each parameter θ based on the loss function It can be expressed as:

[0138]

[0139] in, Denotes the loss function L GR The gradient with respect to the parameter θ is and Represent the loss function L GR The weight w of the neural network gr and the bias b of the neural network gr The partial derivative of

[0140] The chain method is used to calculate the gradient. Assume that the network output Y passes through the Softmax_gr function, then the weight w gr The gradient calculation method can be expressed as:

[0141]

[0142] In the formula, y k is the output of Softmax, z k It is the linear output before the Softmax layer.

[0143] The calculation method can be expressed as:

[0144]

[0145] The calculation method can be expressed as:

[0146]

[0147] The calculation method can be expressed as:

[0148]

[0149] S3.4. Generate a wave source at the corresponding parameter position according to the direction and size of the gradient. At the t+1th iteration, each wave source S i The strength is updated to The calculation method can be expressed as:

[0150]

[0151] In the formula, Denotes the loss function L GR The gradient of the i-th parameter, α ck is the learning rate, represents the gradient of the i-th parameter, is the attenuation coefficient, Indicates that each wave source S at the tth iteration i The strength of is the amplification factor, L GR represents the loss function, α ck Set to 0.01;

[0152] S3.5. The fluctuations of all wave sources are superimposed in the parameter space. The update amount of each parameter is determined by the superposition result of the fluctuations of all wave sources at this point. The specific expression is as follows:

[0153]

[0154] Where k is the wave number, is the neural network parameter of the t+1th iteration, represents the neural network parameters of the t-th iteration, ω is the angular frequency, which represents the propagation speed and direction of the wave, Re() is the ReLU activation function, Denote the loss function as L gr The minimum fluctuation threshold in the historical iteration number of the i-th parameter at time , L gr is the cross entropy loss function, || is the absolute value symbol, ng represents the total number of wave sources, They represent the intensity of the jth wave source at the t+1th iteration, The loss function is L gr The fluctuation threshold of the i-th parameter at time , its expression is as follows:

[0155]

[0156] Among them, τ0 is the initial fluctuation threshold, For L gr Layer sensitivity index of the layer, k pq is the adjustment coefficient, e represents the natural constant, τ0 is set to 0.1, κ pq Set to 3, where The expression is as follows:

[0157]

[0158] Among them, ||*|| represents the L2 norm, Denotes the loss function L GR The gradient of the kth parameter, L GRc Indicates the total number of layers in the network; Denotes the loss function L GR About L GRc Gradients of layer parameters; Indicates L GRc Gradients of layer parameters; Denotes the loss function L GR Gradients about the parameters of the kth layer; Represents the gradient of the k-th layer parameter;

[0159] S3.6. Dynamically adjust the attenuation coefficient β and amplification coefficient γ of each wave source. The specific expressions are as follows:

[0160]

[0161] In the formula, are the attenuation coefficient and amplification coefficient of the previous iteration respectively, They are the attenuation coefficient adjustment factor and the amplification coefficient adjustment factor, Respectively represent the attenuation coefficient and the amplification coefficient, ΔL G R is the change in loss, e is a natural constant, and the parameter and Set to 0.1 and 0.05 respectively;

[0162] S3.7, determine whether the preset maximum number of iterations has been reached, if not, jump to step S3.2, if reached, stop iteration and complete training;

[0163] S4, inputting the feature-extracted data into a feature dimension reduction model to train the feature dimension reduction model, wherein the feature dimension reduction model adopts an autoencoder neural network based on manifold mapping learning to perform feature dimension reduction, inputting the feature-extracted data into a feature dimension reduction model to train the feature dimension reduction model, the present invention adopts an autoencoder neural network based on manifold mapping learning to perform feature dimension reduction, the autoencoder neural network not only learns the intrinsic structure of the data in the encoding stage, but also attempts to reconstruct the input through the decoding process to ensure that the features after dimension reduction can still retain key information, the autoencoder neural network model structure consists of an encoder and a decoder, the encoder is responsible for mapping high-dimensional input data to a low-dimensional feature space, and the decoder reconstructs the original input data from this low-dimensional space, the present invention utilizes Riemann factors based on manifold learning to optimize the representation of the feature space, and by utilizing the geometric properties on the Riemann manifold, the local structure of the data can be better understood, and more accurate data reconstruction can be achieved;

[0164] The training of the autoencoder neural network algorithm based on manifold mapping learning in step S4 includes the following steps:

[0165] S4.1. Initialize the weight w of the autoencoder using Gaussian distribution cr and bias b cr Parameters, the specific expressions are as follows:

[0166]

[0167] in, represents the weight of the i-th layer, σ ch represents the standard deviation. Preferably, n” in ” is the number of nodes in the input layer;

[0168] S4.2. During the encoding process, the data x cr After the encoder, it is mapped to a low-dimensional feature space; during the decoding process, the original data is reconstructed by the decoder. The specific expression is as follows:

[0169] Data x cr is mapped to the low-dimensional feature space zr:

[0170]

[0171] The decoder reconstructs the original input xcr ^:

[0172]

[0173] Among them, Sig() is the activation function; and are the weights and biases of the encoder respectively; and are the weights and biases of the decoder;

[0174] Weight By presetting it in the way of principal component analysis, the calculation method can be expressed as:

[0175]

[0176] Where V k is from the covariance matrix The matrix composed of the first k eigenvectors obtained, Σ k is the corresponding diagonal matrix of eigenvalues;

[0177] S4.3, Riemann factors are used to implement manifold learning. The expression of the features of the Riemann factor R constraint encoding is as follows:

[0178]

[0179] Among them, μ cy represents the preset manifold center, σ R It is a parameter that controls the compactness of the manifold, exp represents an exponential function with a natural constant as the base, and zr represents a low-dimensional feature space;

[0180] S4.4. Construct the loss function Lu, the specific expression is as follows:

[0181] Lu=||x cr -x cr ^|| 2 +λ cm R

[0182] Among them, λ cm is the regularization coefficient, R represents the Riemann factor, and the reconstruction error ||x cr -x cr ^|| 2 The calculation method can be expressed as:

[0183]

[0184] In the formula, z i is the value of z in the i-th dimension, μ i is the value of the preset manifold center in the i-th dimension;

[0185] S4.5, adjust network parameters through back propagation algorithm;

[0186] By weight For example, the calculation method of its gradient can be expressed as:

[0187]

[0188] Furthermore, the gradient It is calculated by the error back propagation method and can be expressed as:

[0189]

[0190] and, Through the decoder weights The transposed representation of can be expressed as:

[0191]

[0192] Furthermore, since z is and x cr The result obtained by the activation function after the linear combination of The calculation method can be expressed as:

[0193]

[0194] Where Sig′() is the derivative of the Sigmoid activation function, diag(vs) represents a diagonal matrix with vector vs as the diagonal element, is x cr The transpose of .

[0195] Furthermore, the gradient descent method is used to update the parameters, and the update method can be expressed as:

[0196]

[0197] In the formula, α rv is the learning rate, t represents the number of iterations, is the updated weight parameter, is the weight parameter before updating, preferably, α rv Set to 0.05;

[0198] S4.6, determine whether the preset maximum number of iterations has been reached, if not, jump to step S4.2, if reached, stop iteration and complete training, the maximum number of iterations is set to 1000;

[0199] S5, inputting the dimension-reduced data into a classifier model to train the classifier model, wherein the classifier model adopts a high-order neural network based on gradient accumulation;

[0200] The training of the high-order neural network algorithm based on gradient accumulation in step S5 specifically includes the following steps:

[0201] S5.1. Initialize the parameters of the high-order neural network: Use random initialization to initialize the high-order neural network parameters W q and bias b q ;

[0202] S5.2. The data after feature dimension reduction is used as the input feature vector and processed through the network layer. The specific expression is as follows:

[0203] z q =Sig(W q ·x q +b q )

[0204] Among them, x q is the input feature vector, z q is the output of this layer, W q Represents the parameters of high-order neural network, Sig() is the Sigmoid activation function;

[0205] S5.3. Use the loss function Lg to evaluate the error between the classifier model output and the actual label. The specific expression is as follows:

[0206]

[0207] Among them, y true，i is the one-hot encoding of the true label, y pred，i is the probability output predicted by the classifier model, log(*) represents the logarithmic function, Indicates the sum of N data;

[0208] y predi The calculation method is expressed as:

[0209]

[0210] In the formula, z q，i is the value of the i-th node in the output layer of the neural network;

[0211] S5.4. Calculate the gradient of high-order neural network parameters and biases based on the loss function Lg and The specific expression is as follows:

[0212]

[0213] in, is the gradient of the loss function Lg with respect to the output; and is the gradient of the output with respect to the weights and biases;

[0214] Further, and The calculation method is expressed as:

[0215]

[0216] Furthermore, the derivative Sig′() of the Sigmoid function is calculated as follows:

[0217] Sig'(zq)=Sig(zq)·(1-Sig(zq))

[0218] S5.5. Accumulate the calculated gradients until the preset training cycle is reached. The preset training cycle is 20 iterations. The accumulated gradients of the high-order neural network parameters and biases are accumulated. and The expression is as follows:

[0219]

[0220] in, Indicates that the gradient is accumulated at each iteration Indicates that the gradient is accumulated at each iteration

[0221] S5.6, after the iteration of step S5.5 is completed, the accumulated gradient is clipped to limit the maximum value of the gradient. The specific expression is as follows:

[0222]

[0223] Among them, clip(*, -τ q , τ q ) means clipping the gradient to [-τ q , τ q ] range, τ q Represents the gradient clipping threshold;

[0224] The accumulated gradient is used to update the parameters, which is expressed as:

[0225]

[0226] In the formula, ← represents the update symbol, τ q is the gradient clipping threshold; λ q is the learning rate. Preferably, τ q Set to 0.1, λ q Set to 0.01.

[0227] In one embodiment, the implementation of gradient clipping is expressed as:

[0228]

[0229] In the formula, g represents the gradient to be clipped, τ q is the gradient clipping threshold

[0230] S5.7, determine whether the preset maximum number of iterations has been reached, if not, jump to step S5.3, if reached, stop iteration and complete training;

[0231] S6. Photovoltaic panel data fault diagnosis: The photovoltaic panel data to be tested is input into the feature extraction model for feature extraction, the data dimension reduction model is used for data dimension reduction, and the fault type is judged by the classifier model. The fault categories include "normal operation", "efficiency reduction", "local fault" and "comprehensive fault".

[0232] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A photovoltaic panel fault diagnosis method based on multi-source data fusion, characterized in that: The following steps are involved: S1. Data collection and annotation constitute the original data set, which is collected from the operation monitoring system of the photovoltaic panel and the on-site detection instrument. The collected data is stored in a structured data standard. The data includes voltage value a1, current value a2, surface temperature a3, light intensity a4, ambient humidity a5, wind speed a6, dust deposition a7, service life a8, maintenance history a9, error code a1, etc. 10 , and manually annotate the collected data; S2, data expansion, sample generation based on the SMOTE algorithm of quantum information non-locality; S3, inputting the expanded data into a feature extraction model to train the feature extraction model, wherein the feature extraction model extracts features using a neural network feature extraction algorithm based on adaptive fluctuation optimization; S4, inputting the feature-extracted data into a feature dimension reduction model to train the feature dimension reduction model, wherein the feature dimension reduction model uses an autoencoder neural network based on manifold mapping learning to perform feature dimension reduction; S5, inputting the dimension-reduced data into a classifier model to train the classifier model, wherein the classifier model adopts a high-order neural network based on gradient accumulation; S6. Photovoltaic panel data fault diagnosis: The photovoltaic panel data to be tested is input into the feature extraction model for feature extraction, the data dimension reduction is performed through the data dimension reduction model, and the fault type is judged through the classifier model.

2. A photovoltaic panel fault diagnosis method based on multi-source data fusion according to claim 1, characterized in that: The sample generation based on the SMOTE algorithm of quantum information non-locality in step S2 includes the following steps: S2.

1. Initialize algorithm parameters, including the initial superposition angle θ of the quantum state b , entanglement strength ∈ b and adaptively adjusted learning rate λ b The initialization of is as follows: ∈ b ＝0.1 l b =0.01 Among them, θ b represents the superposition state angle of the quantum bit, ∈ b represents the degree of entanglement between quantum bits, λ b represents the learning rate; S2.2, use nonlinear transformation to disentangle the features in the original data set collected in step S1, the specific expression is as follows: zb=f b (x;Θ b )=tanh(Θ b ·x) Where x is the original input eigenvector, zb is the eigenvector after disentanglement, Θ b is the weight matrix of the neural network, tanh() is the hyperbolic tangent function, f b (x; Θ b ) represents a nonlinear transformation function; S2.

3. By simulating the behavior of quantum superposition and entanglement, a new sample point y in the data space is generated. The generation method of the new sample is expressed as: y=x+γ b (s b ·(zb-x)) In the formula, y represents the generated new sample, γ b represents the parameter that controls the degree to which the new sample deviates from the original sample, σ b is a random variable adjusted based on the entanglement degree of the quantum state, x is the original eigenvector of the input, and zb represents the eigenvector after disentanglement; S2.

4. Automatically adjust algorithm parameters θ after each iteration b ,∈ b and λ b , to optimize the subsequent sample generation process, expressed as: In the formula, Δθ b is the adjustment amount of the initial superposition state angle of the quantum bit, Δ∈ b is the adjustment amount of entanglement strength, is the initial superposition state angle of the updated quantum bit, is the updated entanglement strength, λ b To adjust the learning rate, and Δθ b and Δ∈ b The expression is as follows: Among them, η cv is the loss learning rate parameter, and They are the loss function L with respect to θ b and ∈ b The partial derivative of b is the adjustment amount of the initial superposition state angle of the quantum bit, Δ∈ b is the adjustment amount of entanglement strength; S2.

5. Repeat the above steps until the preset maximum number of iterations is reached.

3. A photovoltaic panel fault diagnosis method based on multi-source data fusion according to claim 2, characterized in that: The random variable σ in step S2.3 is adjusted based on the quantum state entanglement degree. b According to the entanglement degree of the quantum state ∈ b and the superposition angle θ b Dynamic adjustment, the specific expression is as follows: s b =cos(θ b )·∈ b 2 Among them, cos(θ b ) represents the influence of the superposition state on the sample generation position, ∈ b 2 represents the weight of entanglement in generating samples, σ b It is a random variable adjusted based on the degree of entanglement of the quantum state.

4. A photovoltaic panel fault diagnosis method based on multi-source data fusion according to claim 2, characterized in that: The process of training the neural network feature extraction algorithm based on adaptive fluctuation optimization in step S3 is as follows: S3.

1. Define the neural network structure. The number of hidden layers of the neural network is 2. The first hidden layer includes 100 neurons, and the second hidden layer includes 50 neurons. The activation function is the ReLU activation function. At the same time, set the parameters of the neural network, including the initial parameters of the neural network θ0, the wave source S i Location Strength i and the initial phase φ i , the initial parameters of the neural network θ0 include the weights w of the neural network gr and bias b gr ; S3.

2. Calculate the loss L through forward propagation GR , using the neural network parameters θ on the input X GE Perform forward propagation on the , calculate the output Y and compare it with the target value T, and the expression for calculating the loss is: Y=frs(X GR ;θ) L GR =L GRoss (Y , T) Among them, frs(*) represents the network structure, and the output layer of the network structure uses the preset Softmax_gr classifier to obtain the predicted label, L Gross(Y,T) represents the cross entropy loss; S3.

3. Based on the loss function L GR Calculate the gradient of each parameter θ The expression is: in, Denotes the loss function L GR The gradient with respect to the parameter θ is and Respectively represent the loss function L GR The weight w of the neural network gr and the bias b of the neural network gr The partial derivative of S3.

4. Generate a wave source at the corresponding parameter position according to the direction and size of the gradient. At the t+1th iteration, each wave source S i The strength is updated to The calculation method can be expressed as: In the formula, Denotes the loss function L GR The gradient of the i-th parameter, α ck is the learning rate, represents the gradient of the i-th parameter, is the attenuation coefficient, Indicates that each wave source S at the tth iteration i The strength of is the amplification factor, L GR represents the loss function; S3.

5. The fluctuations of all wave sources are superimposed in the parameter space. The update amount of each parameter is determined by the superposition result of the fluctuations of all wave sources at this point. The specific expression is as follows: Where k is the wave number, is the neural network parameter of the t+1th iteration, represents the neural network parameters of the tth iteration, ω is the angular frequency, which represents the propagation speed and direction of the wave, Re() is the ReLU activation function, and L gr represents the cross entropy loss function, Denote the loss function as L gr The minimum fluctuation threshold value in the historical iteration number of the i-th parameter at time , | | is the absolute value symbol, ng represents the total number of wave sources, They represent the intensity of the jth wave source at the t+1th iteration, The loss function is L gr The fluctuation threshold of the i-th parameter at time , its expression is as follows: Among them, τ0 is the initial fluctuation threshold, For L gr Layer sensitivity index of the layer, κ pq is the adjustment coefficient, e represents the natural constant, where The expression is as follows: Among them, ||*|| represents the L2 norm, Denotes the loss function L GR The gradient of the kth parameter, L GRc Indicates the total number of layers in the network; Denotes the loss function L GR About L GRc Gradients of layer parameters; Indicates L GRc Gradients of layer parameters; Denotes the loss function L GR Gradients about the parameters of the kth layer; Represents the gradient of the k-th layer parameter; S3.

6. Dynamically adjust the attenuation coefficient β and amplification coefficient γ of each wave source. The specific expressions are as follows: In the formula, are the attenuation coefficient and amplification coefficient of the previous iteration respectively, They are the attenuation coefficient adjustment factor and the amplification coefficient adjustment factor, Respectively represent the attenuation coefficient and the amplification coefficient, ΔL GR is the change in loss, e represents a natural constant; S3.

7. Determine whether the preset maximum number of iterations has been reached. If not, jump to step S3.

2. If reached, stop iteration and complete training.

5. The photovoltaic panel fault diagnosis method based on multi-source data fusion according to claim 2 is characterized in that: The training of the autoencoder neural network algorithm based on manifold mapping learning in step S4 comprises the following steps: S4.

1. Initialize the weight w of the autoencoder using Gaussian distribution cr and bias b cr Parameters, the specific expressions are as follows: in, represents the weight of the i-th layer, σ ch represents the standard deviation, n” in ” is the number of nodes in the input layer; S4.

2. During the encoding process, the data x cr After the encoder, it is mapped to a low-dimensional feature space; during the decoding process, the original data is reconstructed by the decoder. The specific expression is as follows: Data x cr is mapped to the low-dimensional feature space zr: The decoder reconstructs the original input x cr ^: Among them, Sig() is the activation function; and are the weights and biases of the encoder respectively; and are the weights and biases of the decoder; S4.3, Riemann factors are used to implement manifold learning. The expression of the features of the Riemann factor R constraint encoding is as follows: Among them, μ cy represents the preset manifold center, σ R It is a parameter that controls the compactness of the manifold, exp represents an exponential function with a natural constant as the base, and zr represents a low-dimensional feature space; S4.

4. Construct the loss function Lu, the specific expression is as follows: Lu=||x cr -x cr ^|| 2 +λ cm R Among them, λ cm is the regularization coefficient, x cr Represents the data before encoding, x cr ^ represents the decoded data, R represents the Riemann factor, ||x cr -x cr ^|| 2 represents the reconstruction error; S4.5, adjust network parameters through back propagation algorithm; S4.

6. Determine whether the preset maximum number of iterations has been reached. If not, jump to step S4.

2. If reached, stop iteration and complete training.

6. A photovoltaic panel fault diagnosis method based on multi-source data fusion according to claim 2, characterized in that: The training of the high-order neural network algorithm based on gradient accumulation in step S5 specifically includes the following steps: S5.

1. Initialize the parameters of the high-order neural network: Use random initialization to initialize the high-order neural network parameters W q and bias b q ; S5.

2. The data after feature dimension reduction is used as the input feature vector and processed through the network layer. The specific expression is as follows: z q =Sig(W q ·x q +b q ) Among them, x q is the input feature vector, z q is the output of this layer, W q Represents the parameters of high-order neural network, Sig() is the Sigmoid activation function; S5.

3. Use the loss function Lg to evaluate the error between the classifier model output and the actual label. The specific expression is as follows: Among them, y true,i is the one-hot encoding of the true label, y pred,i is the probability output predicted by the classifier model, log(*) represents the logarithmic function, Indicates the sum of N data; S5.

4. Calculate the gradient of high-order neural network parameters and biases based on the loss function Lg and The specific expression is as follows: in, is the gradient of the loss function Lg with respect to the output; and is the gradient of the output with respect to the weights and biases; S5.

5. Accumulate the calculated gradients until the preset training cycle is reached, and the accumulated gradients of the high-order neural network parameters and biases and The expression is as follows: in, Indicates that the gradient is accumulated at each iteration Indicates that the gradient is accumulated at each iteration S5.6, after the iteration of step S5.5 is completed, the accumulated gradient is clipped to limit the maximum value of the gradient. The specific expression is as follows: Among them, clip(*,-τ q ,τ q ) means clipping the gradient to [-τ q ,τ q ] range, τ q Represents the gradient clipping threshold; S5.

7. Determine whether the preset maximum number of iterations has been reached. If not, jump to step S5.

3. If reached, stop iteration and complete training.

Images