A photovoltaic power station digital twin modeling method and system

Abstract

The application relates to the technical field of photovoltaic power station digital twin modeling, and discloses a photovoltaic power station digital twin modeling method and system, wherein the photovoltaic power station digital twin modeling method comprises the following steps: obtaining monitoring data of each node of a distributed photovoltaic power station, and generating a temperature-power characteristic vector; analyzing the influence of a temperature gradient by using a graph convolution network, and generating a node-specific temperature compensation matrix; calculating a deviation coefficient based on a standard temperature benchmark, and generating a temperature normalization mapping function; correcting model parameters by using an environment normalization method, and generating an environment-corrected model update vector; performing federal aggregation on the environment-corrected model update vector, and outputting a global digital twin model. The application realizes high-precision distributed photovoltaic power station digital twin modeling under environment heterogeneity conditions, and improves the accuracy and robustness of data integration across climate regions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photovoltaic power station digital twin modeling, more specifically, it relates to a photovoltaic power station digital twin modeling method and system. BACKGROUND

[0002] In the digital twin modeling scenario of distributed photovoltaic power stations, it is necessary to integrate local models of multiple geographically dispersed nodes through federated learning technology to form a unified global digital twin model. However, due to the distribution of photovoltaic power station nodes in different climate regions, there are significant differences in environmental temperature. The output power of photovoltaic modules is negatively correlated with temperature, and for every 1℃ increase in temperature, the output power of crystalline silicon photovoltaic modules decreases by about 0.4%-0.5%. The performance difference caused by temperature is misinterpreted as model bias in the traditional federated learning framework, resulting in systematic errors when aggregating model parameters.

[0003] In addition, some nodes in the distributed photovoltaic power station network may provide low-quality data and model updates due to sensor failures, equipment aging, or improper maintenance. Traditional federated learning aggregates updates from all nodes with equal weights, so that the incorrect updates from these abnormal nodes directly affect the accuracy of the global model, and cannot guarantee the credibility of distributed modeling. SUMMARY

[0004] The present application provides a photovoltaic power station digital twin modeling method and system, which realizes high-precision distributed photovoltaic power station digital twin modeling under environmental heterogeneity, and improves the accuracy and robustness of data integration across climate regions.

[0005] The present application provides a photovoltaic power station digital twin modeling method, comprising: obtaining monitoring data of each node of a distributed photovoltaic power station, and generating a temperature-power feature vector; using a graph convolution network to analyze the influence of temperature gradient and generate a node-specific temperature compensation matrix; calculating a bias coefficient based on a standard temperature reference and generating a temperature normalization mapping function; using an environmental normalization method to correct model parameters and generating an environment-corrected model update vector; federated aggregation of the environment-corrected model update vector is performed to output a global digital twin model; wherein the environmental normalization method transforms the original parameters through the temperature normalization mapping function and fine-tunes them in combination with the node-specific temperature compensation coefficient.

[0006] Further, generating a temperature-power feature vector comprises:

[0007] calculating a temperature coefficient correction factor for the photovoltaic module, wherein the temperature coefficient correction factor is equal to 1 plus the product of the power temperature coefficient and the difference between the module temperature and the standard test condition temperature;

[0008] Based on the temperature coefficient correction factor and the actual power output, a normalized temperature-power characteristic is calculated, which is equal to the actual power divided by the product of the standard condition power, the temperature coefficient correction factor and the irradiance ratio;

[0009] The temperature-power characteristics in each time window are combined into a feature vector.

[0010] Further, the influence of the temperature gradient is analyzed using a graph convolution network, which includes:

[0011] The electrical topology of the photovoltaic power station is represented as a graph, where the nodes represent photovoltaic strings or inverters, and the edges represent electrical connection relationships;

[0012] The node feature data is standardized for preprocessing to eliminate differences in the dimensions of different physical quantities;

[0013] The preprocessed temperature distribution data and electrical topology are input into the graph convolution network, which captures the propagation and influence patterns of temperature in the electrical network through multiple layers of graph convolution operations;

[0014] The graph convolution network outputs the temperature influence weights of each node, forming a node-specific temperature compensation matrix.

[0015] Further, generating a temperature normalization mapping function includes:

[0016] A standard temperature reference is set, and the deviation of the ambient temperature of each node relative to the standard temperature is calculated;

[0017] Based on the temperature deviation and the temperature characteristic curve of the photovoltaic module, a temperature normalization mapping function is generated;

[0018] The temperature normalization mapping function standardizes the influence of temperature deviation on model parameters through a nonlinear transformation, which includes applying a hyperbolic tangent function to the temperature deviation to limit the temperature influence range.

[0019] Further, the environmental normalization method includes:

[0020] The temperature deviation vector of each node is calculated;

[0021] Based on the temperature deviation vector and the temperature coefficient of the photovoltaic module, a parameter correction matrix is constructed;

[0022] The local model update vector is linearly transformed by the parameter correction matrix to obtain the temperature-normalized parameters;

[0023] The corresponding row of the node-specific temperature compensation matrix is extracted as a node compensation vector, which is multiplied element by element with the normalized parameters to generate the final environmental correction vector.

[0024] Further, before federated aggregation, it also includes:

[0025] Obtain the local verification indicators and historical update records of each node, identify abnormal updates using a statistical anomaly detection method, and generate an abnormal label matrix;

[0026] Based on the historical performance of the block chain record node, calculate the time-decaying reputation points, and generate a reputation weight vector;

[0027] The model update vector after environmental correction and the reputation weight vector are input into a weighted federated aggregation module.

[0028] Further, the statistical anomaly detection method includes:

[0029] Calculate the statistical characteristics of each node model update, including mean, variance, skewness, kurtosis and time series autocorrelation coefficient;

[0030] Use the Isolation Forest method to identify abnormal updates deviating from the group distribution;

[0031] For the detected abnormal nodes, mark 1 in the corresponding position of the abnormal label matrix, and mark 0 for normal nodes.

[0032] Further, the weighted federated aggregation includes:

[0033] Calculate the effective node set and filter out the nodes that are not marked as abnormal;

[0034] The reputation weight of the effective node is normalized;

[0035] Based on the adjusted weight and the environmental correction vector, calculate the weighted update amount;

[0036] Add the weighted update amount to the current global model parameters to generate new global model parameters.

[0037] Further, it also includes:

[0038] Evaluate the performance of the global model using validation datasets from different climate zones;

[0039] Update the temperature compensation coefficient and reputation score parameters according to the prediction error feedback to realize adaptive optimization.

[0040] The present application provides a kind of photovoltaic power station digital twin modeling system, including:

[0041] Data acquisition module, for obtaining the monitoring data of each node of distributed photovoltaic power station and generating temperature-power feature vector;

[0042] Temperature compensation analysis module, for analyzing temperature gradient influence using graph convolution network and generating node-specific temperature compensation matrix;

[0043] an environment normalization module configured to generate a temperature normalization mapping function and correct model parameters based on a standard temperature reference;

[0044] a quality screening module configured to identify abnormal nodes and calculate reputation weights;

[0045] a federated aggregation module configured to perform weighted federated aggregation to generate a global digital twin model.

[0046] The present application has the following beneficial effects:

[0047] The present application uses a graph convolution network to analyze temperature gradient influence and generate a node-specific temperature compensation matrix through an environment normalization federated protocol, and realizes standardization of model parameters under different temperature conditions by combining a temperature normalization mapping function, thereby overcoming systematic bias factors caused by temperature differences and solving the problem of errors introduced by traditional federated learning ignoring environmental factors.

[0048] Further, by introducing a quality screening method based on statistical anomaly detection and a reputation scoring system based on a blockchain, effective suppression of low-quality node contributions is realized, and model performance degradation factors caused by equal aggregation are overcome, solving the problem of malicious or faulty nodes affecting the credibility of the global model.

[0049] Therefore, the present application realizes high-precision distributed photovoltaic power station digital twin modeling under environmental heterogeneous conditions, and improves the accuracy and robustness of data integration across climate regions. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 is a photovoltaic power station digital twin modeling method flowchart of the present application;

[0051] Figure 2 is a node temperature compensation matrix thermodynamic diagram of the present application;

[0052] Figure 3 is a performance comparison diagram of different aggregation methods of the present application;

[0053] Figure 4 is a model update vector comparison diagram before and after environmental correction of the present application; DETAILED DESCRIPTION

[0054] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that discussions of these implementations are merely provided to enable those skilled in the art to better understand and thus implement the subject matter described herein, and are not intended to limit the scope of the disclosure. Various modifications can be made to the function and arrangement of elements discussed without departing from the scope of the subject matter described herein. Each example can omit, substitute, or add various processes or components also in accordance. In addition, features described in relation to one example can also be combined in other examples.

[0055] A photovoltaic power station digital twin modeling method is disclosed in at least one embodiment of the present application, as shown, comprising the following steps: Figures 1-4

[0056] Step 100: Obtain the monitoring data of each node of the distributed photovoltaic power station, and generate a temperature-power feature vector.

[0057] Specifically, the temperature distribution data of the photovoltaic array, the electrical connection topology information and the update vector of the local digital twin model are collected from each photovoltaic power station node. Among them, the temperature distribution data includes component surface temperature, ambient temperature and irradiance data; the electrical connection topology describes the connection relationship of photovoltaic strings and inverters; the local model update vector is the model weight increment obtained by each node based on local data training. Based on the collected data, the temperature-power correlation coefficient at each time is calculated, and the temperature-power feature vector is extracted.

[0058] It should be noted that the extraction of the temperature-power feature vector includes the following sub-steps:

[0059] Step 101: Calculate the temperature coefficient correction factor of the photovoltaic module , wherein is the power temperature coefficient, is the component temperature, is the standard test condition temperature (25℃).

[0060] Further, the value range of the power temperature coefficient is [-0.005, -0.004] / ℃, and the value range of the power temperature coefficient is based on the industry standard of crystalline silicon photovoltaic modules. The effective range of the component temperature is [-40℃, 85℃], which corresponds to the working temperature range of the photovoltaic module. The constraint range of the temperature coefficient correction factor is [0.6, 1.4], and when the constraint range of the temperature coefficient correction factor is exceeded, it indicates that the component is working abnormally and needs to be handled abnormally.

[0061] Step 102: Calculate the normalized temperature-power feature based on the temperature coefficient correction factor and the actual power output , wherein is the actual power, is the standard condition power, is the actual irradiance, is the standard irradiance (1000W / m²).

[0062] Further, the time dimension of the normalized temperature-power feature formula is embodied as:

[0063] ​

[0064] where is the normalized temperature-power feature (normalized temperature-power feature value), normal range [0, 1.2], time index denotes the sampling time, sampling interval The constraint range of is [1 minute, 15 minutes]. When minute, the data noise is large and affects the stability of the feature, and when minute, it is impossible to capture the fast response of light changes. All parameters , , are functions of time , reflecting real-time change characteristics.

[0065] Further, the effective range of the actual irradiance is [0, 1200] W / m², corresponding to from night to strong light conditions. The normal range of the normalized temperature-power feature is [0, 1.2], when indicates that there is a measurement anomaly or that the component is overheated, and when indicates that the component performance is severely attenuated or blocked.

[0066] Step 103: Combine the temperature-power features in each time window into a feature vector where denotes the feature vector dimension (time series length / feature number).

[0067] Further, the time series of the feature vector is expressed as:

[0068]

[0069]

[0070] where denotes the transpose operator, is the current time, is the time step, and the superscript denotes the feature value of the time window corresponding to the time . The constraint range of the time window length is [1 hour, 24 hours], ensuring that the complete daily periodic change pattern is captured.

[0071] Further, the value range of the feature vector dimension is [12, 288], corresponding to time resolutions from the hour level (12 hour windows) to the 5 minute level (288 5 minute windows). When Time data is insufficient to capture diurnal variation patterns, when The computational complexity is too high and the gain is limited.

[0072] Step 200: Analyze the effect of temperature gradient using graph convolutional networks and generate node-specific temperature compensation matrices.

[0073] The electrical topology of a photovoltaic power station is represented as shown in the figure. , where nodes Represents a photovoltaic string or inverter, side This represents electrical connectivity. Before inputting into the graph convolutional network, the node feature data undergoes standardized preprocessing: temperature data, electrical parameters, and geographic location information are respectively Z-score standardized to eliminate the impact of differences in the dimensions of different physical quantities on network training. The preprocessed temperature distribution data and electrical topology are then input into the graph convolutional network, which captures the propagation and influence patterns of temperature in the electrical network through multi-layer graph convolution operations. The graph convolutional network outputs the temperature influence weights for each node, forming a node-specific temperature compensation matrix. .

[0074] Furthermore, the node feature matrix The construction method is as follows: for nodes Its eigenvectors for:

[0075]

[0076] in This represents the transpose operator. To standardize the temperature, To standardize power, and These are standardized voltage and current, respectively. and To standardize geographic coordinates, and These represent the mean and standard deviation of the corresponding parameters, respectively. Adjacency matrix. The construction method is as follows: if node and If directly connected in an electrical network, then ,otherwise diagonal elements .

[0077] It should be noted that graph convolutional networks are a common application of graph neural networks (GNNs), and their input layer receives the feature matrix of the nodes. and adjacency matrix ,in This indicates the number of nodes participating in the computation. This represents the node feature dimension. The input node features include temperature data, electrical parameters, and geographical location information of the photovoltaic string. The output layer generates a temperature influence weight matrix. This indicates the temperature compensation relationship between the nodes.

[0078] The aforementioned graph convolutional networks include For each graph convolutional layer, the forward propagation formula is: ,in For the first Hidden states of layer (the first layer) (layer output features) To add self-connected adjacency matrices, This is the corresponding degree matrix. For the first The learnable weight matrix of the layer, This is the ReLU activation function.

[0079] Furthermore, the degree matrix The calculation method is as follows: That is, the first The degree of a node is equal to the number of nodes connected to it (including self-joins), and the degree of off-diagonal elements is equal to the number of nodes connected to it. ( ).in Indicates from arrive Sum of the nodes. Weight matrix. The Xavier initialization method is used, with initial values ​​starting from a uniform distribution. Random sampling.

[0080] Furthermore, the number of layers in a graph convolutional network The value range of is [2, 6]. When The network's expressive power is insufficient to capture complex temperature propagation patterns. Overfitting and gradient vanishing problems can easily occur. Feature dimensions of each layer. The constraint range is [16, 128], which ensures that the network has sufficient expressive power and controllable computational complexity.

[0081] The aforementioned graph convolutional network was trained using a supervised learning model, employing historical temperature-compensated data as labels. Loss function... for:

[0082]

[0083] in The predicted temperature compensation coefficient, This is the actual compensation coefficient. For regularization parameters, It is the Frobenius norm. and They represent from arrive and from arrive Sum of the nodes, Indicates from arrive The summation of network layers was performed. The optimization strategy used the Adam optimizer with an initial learning rate of 0.001, and was adjusted using a cosine annealing strategy.

[0084] Furthermore, the time constraint form of the loss function for:

[0085]

[0086]

[0087]

[0088] in The length of the training time window, The time decay weighting function ( (representing the natural exponential function). sky The attenuation coefficient is... This is an index for historical moments. and They represent from arrive and from arrive Sum of the nodes, Indicates from arrive The time window is summed. Constraints include: sky, (Normalization conditions) (Predicted value physical constraints).

[0089] Furthermore, the true temperature compensation coefficient The calculation method is as follows: based on historical operating data, calculate the nodes. and Power output ratio under the same irradiance conditions ,in and They are nodes and The standard conditional power is calculated statistically through a sliding window (30 days). When historical data is insufficient, a similarity method based on geographical distance is used for estimation. ,in As the baseline value, The geographical distance between nodes. km is the characteristic distance.

[0090] Furthermore, the regularization parameter The range of values ​​is [1×10]. -5 1×10 -2 ],when Time regularization is ineffective and prone to overfitting. Excessive time constraints lead to underfitting. Temperature compensation coefficient. The effective range is [0.5, 2.0], representing the multiple relationship of temperature influence between nodes. The learning rate decay range is [0.001, 1×10]. -6 The training termination condition is that the change in the loss function is less than 1 × 10⁻¹⁰ over 10 consecutive epochs. -4 Or reach the maximum number of training rounds of 500.

[0091] Step 300: Calculate the deviation coefficient based on the standard temperature reference and generate the temperature normalization mapping function.

[0092] Setting a standard temperature reference (e.g., 25℃), calculate the deviation of the ambient temperature at each node from the standard temperature. Based on the temperature deviation and the temperature characteristic curve of the photovoltaic module, a temperature normalization mapping function is generated. It is used to map model parameters under different temperature conditions to a unified standard temperature reference.

[0093] Furthermore, the method for obtaining the temperature characteristic curve of the photovoltaic module is as follows: power output data of each node under different temperature conditions are collected, statistically analyzed according to temperature ranges [each 5℃ is a separate range], and the power-temperature relationship curve is obtained by fitting using the least squares method. ,in It is a first-order temperature coefficient. This is a second-order temperature coefficient. The required data set for fitting is at least 100 sample points, and the fitting correlation coefficient is... Only then is it considered effective.

[0094] Furthermore, standard temperature reference The setting range is [20℃, 30℃], with 25℃ typically chosen as the international standard test condition. Temperature deviation. The effective range is [-65℃, 60℃], corresponding to the environmental temperature differences from polar regions to desert regions. When An exception handling mechanism needs to be activated at this time.

[0095] It should be noted that the temperature normalization mapping function The specific form is:

[0096]

[0097] in The input model parameter vector, This is a dimensionless temperature-sensitive adjustment parameter. The temperature response coefficient is in °C. Ensure It is a dimensionless quantity. The node temperature deviation (in °C) is the unit of measurement. The hyperbolic tangent function is used to limit the range of temperature influence. The temperature normalization mapping function standardizes the effect of temperature deviation on model parameters through nonlinear transformation.

[0098] Furthermore, the time-dependent form of the mapping function for:

[0099]

[0100] All parameters are time. The function. Time constraints include: parameter update frequency. The time correlation of temperature deviation is satisfied (times / day). ℃ (smoothness constraint), parameter adjustment cycle This ensures the timeliness of adaptive responses.

[0101] Furthermore, temperature sensitivity adjustment parameters The value range is [0.1, 0.8]. The temperature compensation effect is not obvious when Overcompensation may introduce new biases. Temperature response coefficient. The value range is [0.01, 0.05]℃ The range of values ​​for the temperature response coefficient ensures that the nonlinear characteristics of the temperature effect function are reasonable. The output multiple range of the mapping function is constrained to [0.2, 1.8] to prevent overcompensation under extreme temperature conditions.

[0102] Furthermore, parameters and The adaptive adjustment method is as follows: based on the prediction error on the validation set, the optimal value is determined by a grid search method. The search step size is 0.005℃. , The search step size is 0.1, and the evaluation metric is the root mean square error (RMSE). The search terminates when the RMSE changes by less than 0.001 for three consecutive iterations. The initial value is set to... ℃ , .

[0103] Step 400: Correct the model parameters using the environment normalization method and generate the environment-corrected model update vector.

[0104] Update the local model vectors of each node. and node-specific temperature compensation matrix The data is input into the environment normalization module. This module first applies a temperature normalization mapping function to transform the original parameters, then performs fine-tuning using node-specific temperature compensation coefficients, and outputs the environment-corrected model update vector. ,in This represents element-wise product.

[0105] Furthermore, the algorithm flow of the environment normalization processing module is as follows: Input checking: Validation The dimensions are consistent with the model parameter dimensions. (Check) The numerical range; temperature normalization: calculation ,in Update the local model vector after temperature normalization; spatial compensation: perform Numerical stability check: If Then truncation is used. .

[0106] The aforementioned environment normalization method includes the following processing steps:

[0107] Step 4001: Calculate the temperature deviation vector for each node. ,in For nodes The temperature vector, For standard reference temperature vector;

[0108] Step 4002: Construct a parameter correction matrix based on the temperature deviation vector and the temperature coefficient of the photovoltaic module. ,in For temperature coefficient, This is a diagonalization operation;

[0109] Furthermore, temperature coefficient The constraint range is [-0.006, -0.003]℃. The temperature coefficient constraint range covers the temperature characteristics of different types of photovoltaic modules. Parameter correction matrix. The diagonal element range is constrained to [0.4, 1.6] to ensure that the correction factor is within a reasonable range.

[0110] Step 4003: Perform a linear transformation on the local model update vector using the parameter correction matrix to obtain the temperature-normalized parameters. ;

[0111] Step 4004: Extract the node-specific temperature compensation matrix. The row, used as the node compensation vector, is multiplied element-wise with the normalization parameter to generate the final environment correction vector. .

[0112] Step 500: Federate the updated model vectors after environmental correction to output a global digital twin model.

[0113] Collect all node-corrected model update vectors and aggregate them using a federated aggregation method: ,in This indicates the number of nodes participating in the federated aggregation. Indicates from arrive The nodes are summed. After aggregation, a robust global digital twin model adaptable to multiple climate regions is obtained.

[0114] Furthermore, the synchronization control method for federated aggregation is as follows: setting the aggregation round interval. seconds, when the number of participating nodes When aggregation is started, This is the set of active nodes. For delayed updates from nodes, an expiration and discard policy is used: if the update delay exceeds [a certain value], the node is discarded. Then discard the node update. Perform a numerical stability check before aggregation: calculate the L2 norm of the update vector. If the threshold is exceeded Then truncation will be performed.

[0115] The input to the aforementioned federated aggregation method includes: the set of environment-corrected model update vectors for each node. and current global model parameters The output is the updated global model parameters. .

[0116] It should be noted that the output decoding process of the global digital twin model includes the following steps: [decoding the global model parameters] The data is fed into a photovoltaic power generation prediction neural network. The network receives current environmental conditions (temperature, irradiance, humidity, etc.) as input features and calculates the predicted power output, predicted power generation efficiency, and equipment health status of the photovoltaic array through forward propagation. These prediction results can be directly used for real-time monitoring, power dispatching, and maintenance decisions of photovoltaic power plants.

[0117] Furthermore, the specific structure of the photovoltaic power generation prediction neural network is as follows: the input layer has a dimension of 8 (including temperature, irradiance, humidity, wind speed, altitude, azimuth, tilt angle, and time encoding), the hidden layer adopts a 3-layer fully connected structure with the number of neurons [64, 32, 16], the activation function is ReLU, and the output layer has a dimension of 3 (power prediction, efficiency prediction, and health status).

[0118] The total number of network parameters is approximately:

[0119] One. Forward propagation formula for: .

[0120] In this embodiment of the application, in order to further improve the model quality, the following steps are included before step 500:

[0121] Step 401: Obtain the local verification metrics and historical update records of each node, identify abnormal updates using statistical anomaly detection methods, and generate an anomaly marker matrix.

[0122] Specifically, the input data is first preprocessed by normalization: the statistical characteristics of local validation metrics (usually percentage data) and historical update records are standardized to eliminate the influence of differences in the units of measurement between different data types. Then, the statistical characteristics (mean, variance) of the model updates for each node are calculated, and anomaly updates deviating from the population distribution are identified using the isolated forest method or density-based anomaly detection techniques. For detected anomalous nodes, anomaly markers are added to the anomaly labeling matrix. The corresponding position is marked as 1, and normal nodes are marked as 0.

[0123] Furthermore, the statistical feature matrix The construction method is as follows: for nodes Its eigenvectors ,in This represents the transpose operator. To update the average magnitude of the vector, For variance, and These are skewness and kurtosis, respectively. This represents the time-series autocorrelation coefficient. Indicates from arrive Summing within a time window. Time window Set to the last 30 updates.

[0124] The input to the aforementioned isolated forest method is the statistical feature matrix of the node model update vectors. ,in Represents the feature dimension (number of statistical features); the output is an anomaly score vector. A binary anomaly marker is generated by using a threshold.

[0125] Furthermore, the specific implementation steps of the Isolation Forest algorithm are as follows: Random sampling: from the feature matrix Random sampling A training subset is constructed from individual samples; random partitioning: for each isolated tree, a feature dimension is randomly selected. and segmentation threshold Recursively segment until the samples are isolated or the maximum depth is reached. Path length calculation: Calculate the isolated path length in each tree for the test sample. Abnormal scores :calculate ,in This represents the average path length. The normalization constant is This is the harmonic number. The number of trees is set to 100. Indicates the feature dimension (the number of statistical features).

[0126] Furthermore, feature dimensions The value range is [5, 15], including mean, variance, skewness, kurtosis, and time series characteristics. Anomaly score. The range is [0,1], and the anomaly detection threshold is set to 0.6. The time marker is designated as an abnormal node. The abnormal detection threshold is set based on experience and can be adjusted within the range of [0.5, 0.8] according to the system's operating conditions.

[0127] Step 402: Based on the historical performance of the blockchain-recorded nodes, calculate the time-decayed reputation score and generate a reputation weight vector.

[0128] The historical model contribution quality, data integrity, and update consistency of each node are recorded on the blockchain to ensure the immutability of these records. A time decay function is used when calculating the reputation score. ,in For the first Quality score for each contribution (dimensionless). It is the attenuation coefficient and its dimensions are consistent with the reciprocal of time (e.g., day). To ensure the exponential function The exponential part is dimensionless. To contribute time, Indicates from arrive The contribution records are summed. A normalized reputation weight vector is generated based on the reputation score.

[0129] Furthermore, the complete constraint condition for the reputation decay function is: time window constraint. Days (historical records exceeding the time window constraint are not included in the calculation), contribution record count constraint (To ensure statistical validity), time interval constraints Days (to avoid noise from frequent updates), decay weight normalization condition and numerical stability conditions (Items less than this value can be ignored). Among them... Indicates the first The timestamp of each contribution record This represents the index of contribution records.

[0130] Furthermore, quality rating The value range is [0,1], where 0 represents completely unreliable and 1 represents completely reliable. It is typically evaluated based on a comprehensive assessment of model validation accuracy, data integrity, and update consistency. (Attenuation coefficient) The value range is [0.001, 0.1] days. ,when The slow decay of historical information affects real-time performance. The rapid decay of time leads to the loss of useful historical information. The element constraints of the reputation weight vector are... and .

[0131] Step 403: Input the environment-corrected model update vector and reputation weight vector into the weighted federated aggregation module to output an environment-robust global digital twin model. Modify the federated aggregation formula as follows: ,in Used to mask the contributions of abnormal nodes.

[0132] Furthermore, the time series optimization objective function of federated aggregation for:

[0133]

[0134] in For the target model parameters, The day serves as the evaluation time window. is the regularization coefficient. Indicates from arrive Sum of time windows, Indicates from arrive The nodes are summed. Constraints include: convergence constraints. Stability constraints and time consistency constraints .

[0135] The aforementioned weighted federated aggregation module includes the following processing steps:

[0136] Step 4031: Calculate the set of valid nodes Filter out nodes that are not marked as abnormal;

[0137] Step 4032: Normalize the reputation weights of valid nodes and calculate the adjusted weights. ;

[0138] Furthermore, the adjusted weights satisfy and When the number of valid nodes At this time, the system enters protection mode, suspends federated learning, and waits for more trusted nodes to join.

[0139] Step 4033: Calculate the weighted update amount based on the adjusted weights and environmental correction vector. ,in Represents the set of valid nodes All of them Summation;

[0140] Step 4034: Add the weighted update amount to the current global model parameters to generate new global model parameters. .

[0141] In this embodiment of the application, in order to achieve continuous optimization, the following steps are also included:

[0142] Step 600: Evaluate the global model performance using validation datasets from different climate zones, and update the temperature compensation coefficient and reputation score parameters based on prediction error feedback.

[0143] Independent validation datasets were collected from various climate zones, and the prediction error of the global model under different temperature conditions was calculated. Based on the error analysis results, the weight parameters of the graph convolutional network and the decay coefficient of the reputation score were updated using gradient descent to achieve adaptive optimization.

[0144] Furthermore, the validation dataset was constructed as follows: 10 representative nodes were selected from each climate zone (tropical, temperate, polar, arid, and humid). Each node provided its operational data for the most recent 60 days as the validation set, including environmental parameters (temperature, irradiance, humidity) and actual power generation. The validation set was divided into a parameter tuning set and a test set in a 7:3 ratio. Data preprocessing employed the same standardization method as the training set to ensure data distribution consistency.

[0145] The input to the aforementioned gradient descent method is the loss function. Regarding parameters gradient The output is the updated parameters. ,in This is the learning rate.

[0146] Furthermore, the time series update rule for gradient descent is as follows: The time index In days For a moment Model parameters, For a moment The model parameters, learning rate, and time decay strategy are as follows: attenuation coefficient sky The time constraints include: maximum optimization time. Heaven, parameter rate of change constraint (To prevent drastic changes), and performance monotonicity constraints within the time window. ,in To allow for a certain growth rate.

[0147] Furthermore, the specific method for gradient calculation is as follows: the loss function is defined as... ,in To verify the sample size, This represents the actual amount of electricity generated. To predict power generation, Indicates from arrive The validation samples are summed. The gradient is calculated using the backpropagation algorithm: for the ... Layer parameters Its gradient is ,in This represents the transpose operator. For the first Layer output. Adaptive learning rate adjustment uses the Adam algorithm:

[0148]

[0149]

[0150]

[0151] in This is the first moment estimate for the Adam algorithm (the first moment estimate of the gradient in the Adam optimization algorithm). This is an estimate of the first moment from the previous moment. This is the second moment estimate of the Adam algorithm (the second moment estimate of the gradient in the Adam optimization algorithm). This is an estimate of the second moment from the previous moment. These are the updated parameters (model parameters for the next time step). This is the first-order moment estimate after bias correction (the first-order moment estimate after bias correction). This is the second-order moment estimate after bias correction (the second-order moment estimate after bias correction). , , .

[0152] Furthermore, the learning rate The range of values ​​is [1×10]. -5 [,0.01], when Slow convergence affects real-time performance. This may lead to parameter oscillations and non-convergence. The prediction error is assessed using the Mean Absolute Percentage Error (MAPE), and a parameter update mechanism is triggered when MAPE > 15%. The parameter convergence criterion is that the parameter change amplitude in five consecutive updates is less than 1 × 10⁻⁶. -4 Or it may reach the maximum number of updates, 100 times.

[0153] The following are real-world application scenarios of this invention:

[0154] A distributed digital twin modeling system was constructed using five photovoltaic power station nodes distributed across different climate zones: City A (temperate continental climate), City B (subtropical monsoon climate), City C (temperate continental arid climate), City D (temperate monsoon climate), and City E (tropical maritime monsoon climate). Each node has an installed capacity of 50MW, employing monocrystalline silicon photovoltaic modules with a power temperature coefficient of [missing information]. / ℃.

[0155] Taking the time period from 12:00 to 14:00 on a certain day in July 2024 as an example, the raw monitoring data collected by each node is shown in Table 1 below:

[0156] Table 1: Raw monitoring data for each node

[0157]

[0158] Based on the above data, the temperature-power feature vector of each node is calculated according to the method in step 100:

[0159] For Node_1 (City A node), calculate the temperature coefficient correction factor:

[0160]

[0161] Calculate the normalized temperature-power characteristic:

[0162]

[0163] Similarly, calculate the feature values ​​of other nodes, and combine the features within each time window into a vector according to step 103. (Using time windows...) Taking hourly sampling as an example, the temperature-power feature vectors of each node are obtained, as shown in Table 2 below:

[0164] Table 2: Temperature-Power Eigenvector (12:00-14:00 Time Window)

[0165]

[0166] Following step 200, the influence of temperature gradients is analyzed using a graph convolutional network. A node feature matrix is ​​constructed. The standardized node features are shown in Table 3 below:

[0167] Table 3: Standardized Node Feature Matrix

[0168]

[0169] After processing by a graph convolutional network, a node-specific temperature compensation matrix is ​​generated. Some of the elements are shown in Table 4 below:

[0170] Table 4: Temperature Compensation Matrix (Partial Elements)

[0171]

[0172] Following steps 300-400, apply the environmental normalization method. First, calculate the temperature deviation vector for each node according to sub-step 4001:

[0173] For Node_2 (City B node), the ambient temperature is 38.5℃, which is the standard temperature reference. :

[0174]

[0175] Construct the parameter correction matrix according to sub-step 4002, using the temperature coefficient. ℃ :

[0176]

[0177] According to sub-step 4003, the parameters after temperature normalization are:

[0178]

[0179] Following sub-step 4004, extract the temperature compensation matrix. The elements are multiplied one by one to generate the final environment correction vector:

[0180]

[0181]

[0182] As shown in Table 5 below:

[0183] Table 5: Comparison of model update vectors before and after environmental correction (processed according to sub-steps 4001-4004)

[0184]

[0185] Perform weighted federated aggregation according to steps 500 and 403. The reputation weight of each node is... Anomaly marker matrix (No abnormal nodes), calculate according to the aggregation formula:

[0186]

[0187] Following the processing flow of steps 4031-4034:

[0188] Step 4031: Set of Valid Nodes (No abnormal nodes)

[0189] Step 4032: Weights have been normalized.

[0190] Step 4033: Calculate the weighted update amount

[0191]

[0192]

[0193] Step 4034:

[0194] As shown in Table 6 below:

[0195] Table 6: Comparison of Federated Aggregation Results (Based on the Weighted Aggregation Formula in Step 403)

[0196]

[0197] The results show that after environmental normalization federal protocol processing, the model prediction accuracy improved from 12.8% to 7.1% of the traditional method, and the convergence speed improved by 37.8%. It effectively solved the systematic bias problem caused by temperature differences and realized high-precision digital twin modeling of distributed photovoltaic power stations across climate regions.

[0198] It is understood that data preprocessing methods known to those skilled in the art include data cleaning, data transformation, and data reduction. Data transformation includes type conversion and normalization and standardization. Although the dimensions and types of data were omitted in the description of the preceding embodiments, data preprocessing is a technical knowledge known to those skilled in the art and a prerequisite step in data processing. Therefore, the previously described well-known data preprocessing steps were not described independently.

[0199] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. A method for digital twin modeling of a photovoltaic power station, characterized in that, Includes the following steps: Acquire monitoring data from each node of a distributed photovoltaic power station and generate a temperature-power feature vector; The influence of temperature gradient is analyzed using graph convolutional networks, generating node-specific temperature compensation matrices. The analysis of the effect of temperature gradient using graph convolutional networks includes: The electrical topology of a photovoltaic power station is represented as a graph, where nodes represent photovoltaic strings or inverters, and edges represent electrical connections. Standardize and preprocess the node feature data to eliminate differences in the dimensions of different physical quantities; The preprocessed temperature distribution data and electrical topology are input into a graph convolutional network, and the propagation and influence patterns of temperature in the electrical network are captured through multi-layer graph convolution operations. The graph convolutional network outputs the temperature influence weights of each node, forming a node-specific temperature compensation matrix. The deviation coefficient is calculated based on the standard temperature reference, and a temperature normalization mapping function is generated. The generated temperature normalization mapping function includes: Set a standard temperature reference and calculate the deviation of the ambient temperature at each node from the standard temperature; A temperature normalization mapping function is generated based on the temperature deviation and the temperature characteristic curve of the photovoltaic module. The temperature normalization mapping function standardizes the influence of temperature deviation on model parameters through nonlinear transformation. The nonlinear transformation includes applying a hyperbolic tangent function to the temperature deviation to limit the range of temperature influence. The model parameters are corrected using the environment normalization method, and an environment-corrected model update vector is generated. Federated aggregation of the environmentally corrected model update vectors is performed to output a global digital twin model; The environmental normalization method transforms the original parameters through a temperature normalization mapping function and performs fine-tuning by combining a node-specific temperature compensation coefficient.

2. The photovoltaic power plant digital twin modeling method according to claim 1, characterized in that, The generated temperature-power feature vector includes: Calculate the temperature coefficient correction factor for the photovoltaic module, which is equal to 1 plus the product of the power temperature coefficient and the difference between the module temperature and the standard test condition temperature. Based on the temperature coefficient correction factor and the actual power output, the normalized temperature-power characteristic is calculated. The normalized temperature-power characteristic is equal to the actual power divided by the product of the standard condition power, the temperature coefficient correction factor, and the irradiance ratio. The irradiance ratio is the ratio of the actual irradiance to the standard irradiance. The temperature-power features within each time window are combined into a feature vector.

3. The method for digital twin modeling of photovoltaic power plants according to claim 1, characterized in that, The environment normalization method includes: Calculate the temperature deviation vector for each node; A parameter correction matrix is ​​constructed based on the temperature deviation vector and the temperature coefficient of the photovoltaic module; The local model update vector is linearly transformed using the parameter correction matrix to obtain the temperature-normalized parameters. The corresponding row of the node-specific temperature compensation matrix is ​​extracted as the node compensation vector, and multiplied element-wise with the normalization parameter to generate the final environmental correction vector.

4. The method for digital twin modeling of photovoltaic power plants according to claim 1, characterized in that, This also includes the following before performing federated aggregation: Obtain local verification metrics and historical update records for each node, identify abnormal updates using statistical anomaly detection methods, and generate an anomaly marker matrix; Based on the historical performance of nodes recorded in the blockchain, a reputation score with time decay is calculated to generate a reputation weight vector. The environmentally corrected model update vector and reputation weight vector are input into the weighted federated aggregation module.

5. The method for digital twin modeling of photovoltaic power plants according to claim 4, characterized in that, The statistical anomaly detection method includes: Calculate the statistical characteristics of the model updates at each node, including mean, variance, skewness, kurtosis, and time-series autocorrelation coefficient; Identify anomalous updates that deviate from the population distribution using the isolated forest method; For detected abnormal nodes, the corresponding position in the abnormal node marking matrix is ​​marked as 1, and normal nodes are marked as 0.

6. The method for digital twin modeling of photovoltaic power plants according to claim 4, characterized in that, The weighted federated aggregation includes: Calculate the set of valid nodes and filter out nodes that have not been marked as abnormal; The reputation weights of valid nodes are normalized. Calculate the weighted update amount based on the adjusted weights and environmental correction vector; The weighted update amount is added to the current global model parameters to generate new global model parameters.

7. The method for digital twin modeling of photovoltaic power plants according to claim 1, characterized in that, Also includes: Evaluate the global model performance using validation datasets from different climate zones; The temperature compensation coefficient and credit score parameters are updated based on the prediction error feedback to achieve adaptive optimization.

8. A photovoltaic power plant digital twin modeling system, used to execute the photovoltaic power plant digital twin modeling method according to any one of claims 1-7, characterized in that, include: The data acquisition module is used to acquire monitoring data from each node of the distributed photovoltaic power station and generate temperature-power feature vectors; The temperature compensation analysis module is used to analyze the effects of temperature gradients using graph convolutional networks and generate node-specific temperature compensation matrices. The environmental normalization module is used to generate a temperature normalization mapping function based on a standard temperature reference and to correct model parameters. The quality screening module is used to identify abnormal nodes and calculate reputation weights; The federated aggregation module is used to perform weighted federated aggregation to generate a global digital twin model.

Images