Carbon emission tracking and accounting method in optical storage and charging integrated system

Abstract

The invention discloses a carbon emission tracking and accounting method in an optical storage and charging integrated system, and relates to the technical field of carbon emission management, and the method comprises the steps: collecting and preprocessing the operation data of each module in the optical storage and charging integrated system, the operation data comprising photovoltaic power generation data, energy storage system charging and discharging data and charging station power utilization data; inputting the preprocessed operation data into a pre-trained carbon emission estimation model to obtain carbon emission estimation data; inputting the carbon emission estimation data into a pre-trained carbon emission tracking model to obtain refined carbon emission data; combining the refined carbon emission data with the preprocessed operation data, and carrying out carbon emission accounting; according to the accounting result, carbon emission optimization analysis is carried out, and a carbon emission report is compiled. The method not only has important theoretical value, but also shows remarkable economic and environmental benefits in practical application, and a new solution is provided for carbon emission management of an optical storage and charging integrated system.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission management, and in particular to a method for tracking and calculating carbon emissions in a photovoltaic storage and charging integrated system. Background Art

[0002] With the global emphasis on sustainable development and environmental protection, the monitoring and management of carbon emissions has become an important task for governments and enterprises around the world. As an emerging energy management model, the integrated photovoltaic storage and charging system combines photovoltaic power generation, energy storage systems and electric vehicle charging stations, which can effectively utilize renewable energy and reduce dependence on fossil fuels. However, the carbon emissions generated by the integrated photovoltaic storage and charging system during operation still need to be solved urgently.

[0003] Traditional carbon emission accounting methods often rely on static models and empirical data, lacking real-time performance and accuracy. These methods are usually unable to effectively capture dynamic changes in system operation, resulting in deviations in carbon emission estimation results. In addition, existing carbon emission monitoring technologies mostly focus on carbon emission analysis of a single module, lacking a comprehensive assessment of the entire system. This limitation makes it challenging for companies to optimize carbon emissions, and they are unable to accurately identify the contribution of each module to overall carbon emissions, thus affecting the effectiveness of optimization decisions.

[0004] In recent years, the rapid development of deep learning technology has provided new solutions for carbon emission monitoring and accounting. By building a carbon emission estimation model based on deep learning, efficient processing and analysis of large-scale data can be achieved, thereby improving the accuracy and real-time performance of carbon emission monitoring. In particular, the advantages of recursive neural networks such as long short-term memory networks (LSTM) in time series data processing make it possible to dynamically track carbon emissions in integrated photovoltaic storage and charging systems.

[0005] However, although some studies have attempted to apply deep learning to carbon emission estimation, comprehensive carbon emission tracking and accounting methods for integrated photovoltaic storage and charging systems are still relatively scarce. Therefore, developing a systematic carbon emission tracking and accounting method that can monitor and analyze carbon emission data in integrated photovoltaic storage and charging systems in real time has important theoretical significance and practical application value. Summary of the invention

[0006] The present invention provides a method for tracking and calculating carbon emissions in an integrated photovoltaic storage and charging system. By collecting and preprocessing the operating data of each module in the system, and using pre-trained carbon emission estimation models and carbon emission tracking models, a refined accounting of carbon emissions is ultimately achieved. This not only improves the accuracy of carbon emission monitoring, but also provides a scientific basis for the company's carbon emission optimization analysis.

[0007] In order to solve the above problems, the technical solution adopted by the present invention is as follows:

[0008] The present invention provides a method for tracking and calculating carbon emissions in a photovoltaic storage and charging integrated system, comprising the following steps:

[0009] S1. Collect and pre-process the operating data of each module in the photovoltaic storage and charging integrated system. The operating data includes photovoltaic power generation data, energy storage system charging and discharging data, and charging station power consumption data;

[0010] S2. Input the preprocessed operation data into the pre-trained carbon emission estimation model to obtain carbon emission estimation data;

[0011] S3. Inputting the carbon emission estimation data into a pre-trained carbon emission tracking model to obtain refined carbon emission data, wherein the carbon emission tracking model is constructed based on a deep learning convolutional neural network;

[0012] S4. Combine the refined carbon emission data with the pre-processed operation data to perform carbon emission accounting;

[0013] S5. Conduct carbon emission optimization analysis based on the accounting results and prepare a carbon emission report.

[0014] As a further description of the above technical solution: in step S1, the preprocessing method includes data cleaning, abnormal data marking and processing, feature extraction and dimensionality reduction, data denoising and data standardization.

[0015] As a further description of the above technical solution: In step S2, the carbon emission estimation model includes an input layer, a feature extraction layer, a hidden layer and an output layer, wherein the input layer is used to receive preprocessed operating data, the feature extraction layer uses a convolutional neural network to extract features from the input data to capture local features and spatial correlations in the data, and it further learns the nonlinear relationships and deep-level features in the data through multiple hidden layers, and the output layer is used to convert the output of the hidden layer into carbon emission estimation data.

[0016] As a further description of the above technical solution: During the training process of the carbon emission estimation model, the mean square error is used as the loss function, the Adam optimizer is used to update the model parameters, the parameters are adjusted by calculating the gradient of the loss function with respect to the parameters to minimize the loss function, and L2 regularization is used to prevent overfitting.

[0017] As a further description of the above technical solution: In step S3, the carbon emission tracking model is built based on the long short-term memory network, which uses the long short-term memory network to process time series data, capture the changing trends and patterns of carbon emission data over time, and track the changes in carbon emissions at each time point through the recursive structure of the LSTM network, and associate these changes with corresponding operations or modules.

[0018] As a further description of the above technical solution: The carbon emission tracking model also includes an attention mechanism layer for identifying and focusing on the modules and time periods that have the greatest impact on carbon emissions.

[0019] As a further description of the above technical solution: Step S4 specifically includes: integrating the refined carbon emission data with the pre-processed operation data one by one; using the weighted average method to calculate the carbon emissions of each module in the integrated photovoltaic storage and charging system, where the weight is set based on the operating time and efficiency of each module.

[0020] As a further description of the above technical solution: Step S5 specifically includes: analyzing the accounting results to determine the carbon emission contribution ratio of each module; preparing a carbon emission report including the carbon emissions of each module in the integrated photovoltaic storage and charging system, the total carbon emissions and the carbon emission contribution ratio.

[0021] As a further description of the above technical solution: the carbon emission report is also matched with an interactive visualization interface, allowing users to filter and view carbon emission data according to different needs and conditions, as well as perform simulation forecasts.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] 1) By combining the operating data of photovoltaic power generation, energy storage systems and charging stations, and using deep learning models to estimate and track carbon emissions, the errors in traditional methods can be effectively reduced and the accuracy of carbon emission monitoring can be improved.

[0024] 2) This method uses deep learning technologies such as LSTM to process and analyze time series data in real time, capture the dynamic changing trends of carbon emissions, and help enterprises adjust their operating strategies in a timely manner to reduce carbon emissions.

[0025] 3) By carrying out detailed accounting of the carbon emissions of each module in the integrated photovoltaic storage and charging system, the contribution ratio of each module to the overall carbon emissions can be clarified, providing a scientific basis for the company's carbon emission optimization.

[0026] 4) Carbon emission reports based on accounting results can provide enterprises with detailed carbon emission data and trend analysis, supporting enterprises to make more scientific decisions in carbon emission management and optimization.

[0027] 5) By improving the accuracy and real-time nature of carbon emissions monitoring, it helps companies better understand and manage their carbon emissions, promotes the use of renewable energy and the realization of sustainable development goals.

[0028] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the embodiments of the present invention are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 is a flow chart of a method for tracking and calculating carbon emissions in a photovoltaic storage and charging integrated system according to an embodiment;

[0031] Figure 2 It is a flow chart of carbon emission accounting described in the embodiment. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0033] Please refer to Figure 1 The embodiment of the present invention discloses a method for tracking and calculating carbon emissions in a photovoltaic storage and charging integrated system, which is as follows:

[0034] S1. Collect and pre-process the operating data of each module in the integrated photovoltaic storage and charging system.

[0035] Among them, the operating data includes photovoltaic power generation data, energy storage system charging and discharging data, and charging station electricity consumption data.

[0036] Preprocessing methods include data cleaning, abnormal data marking and processing, feature extraction and dimensionality reduction, data denoising and data standardization.

[0037] The collected data is cleaned, including noise removal, missing value filling and data smoothing operations. A low-pass filter is used to remove noise in the data to construct a park power data set. The dimensionless processing technology is used to scale the pre-processed related data information so that its range falls within the predetermined range.

[0038] Abnormal data marking and processing is to analyze and mark the common abnormal types of photovoltaic power station data according to the working characteristics of the string inverter, and divide the data in combination with the sliding window method to construct a data set for deep learning training; for abnormal data, the GRU neural network model is used for processing, and the processing accuracy of this model on actual photovoltaic power station data reaches 99.84%.

[0039] Feature extraction and dimensionality reduction are achieved by calculating the eigenvalues ​​of the matrix, constructing the covariance matrix, analyzing the correlation between the vectors in the covariance matrix, calculating the information contribution of each feature by sorting the eigenvalues, retaining the first few eigenvectors with the largest information contribution, and converting them to a new matrix. This method can reduce the computational burden and has little impact on the accuracy of the algorithm.

[0040] Denoising the data can simulate and reduce the random noise in the data through statistical methods such as the Monte Carlo algorithm, thereby improving the accuracy of the data.

[0041] We standardize the preprocessed data to make it meet the requirements of model input and improve the effect of model training and prediction accuracy.

[0042] S2. Input the preprocessed operation data into the pre-trained carbon emission estimation model to obtain carbon emission estimation data.

[0043] Among them, the carbon emission estimation model includes an input layer, a feature extraction layer, a hidden layer and an output layer; its input layer is used to receive preprocessed operating data; its feature extraction layer uses a convolutional neural network (CNN) to extract features from the input data to capture local features and spatial correlations in the data; it further learns the nonlinear relationships and deep-level features in the data through multiple hidden layers (fully connected layers); its output layer is used to convert the output of the hidden layer into carbon emission estimation data.

[0044] In this embodiment, the convolutional layer output of the carbon emission estimation model is calculated by the following formula:

[0045] (f*x)(i,j)=∑ m ∑ n f(m,n)·x(i+m,j+n)

[0046] Among them, f is the convolution kernel, x is the input data, * represents the convolution operation, and i, j represent the position on the output feature map.

[0047] Using the ReLU function as the activation function, that is, ReLU(x)=max(0,x), helps to introduce nonlinearity and enhance the expressiveness of the model.

[0048] The output calculation formula of the fully connected layer is: y=Wx+b, where W is the weight matrix, b is the bias term, and x is the input feature vector.

[0049] The training method of the carbon emission estimation model is as follows:

[0050] 1. Loss Function

[0051] The loss function used is the mean square error (MSE), and the formula is:

[0052]

[0053] Where m is the number of data points, C (i) is the actual carbon emission of the ith data point, is the predicted value of the model.

[0054] 2. Model parameter updating method: Use the Adam optimizer to update the model parameters. The parameters are adjusted by calculating the gradient of the loss function with respect to the parameters to minimize the loss function. The Adam optimizer combines the advantages of momentum and RMSprop and can adaptively adjust the learning rate. It is currently a widely used optimizer in deep learning.

[0055] 3. Regularization

[0056] L2 regularization is used to prevent overfitting. L2 regularization is implemented by adding a regularization term to the loss function. This term is the sum of the squares of the model parameters. The formula for the regularization term is:

[0057]

[0058] Among them, λ is the regularization coefficient, K is the number of layers in the model, and N k is the number of weights in the kth layer, w ki is the weight of the kth layer.

[0059] 4. Training and validation: Divide the data into training set and validation set, use the training set to train the model, and use the validation set to evaluate the performance of the model.

[0060] S3. Input the carbon emission estimation data into a pre-trained carbon emission tracking model to obtain refined carbon emission data. The carbon emission tracking model is constructed based on a convolutional neural network of deep learning.

[0061] The carbon emission tracking model is built based on long short-term memory network.

[0062] After the carbon emission estimation data obtained in step S2 is input into the carbon emission tracking model, the carbon emission tracking model uses the gating mechanism of the LSTM network to process the time series data, including the input gate, the forget gate and the output gate, to capture the changing trend and pattern of the carbon emission data over time, and through the recursive structure of the LSTM network, to track the changes in carbon emissions at each time point, and associate these changes with the corresponding operations or modules; an attention mechanism layer is added to the LSTM network, which dynamically adjusts the model focus by calculating the weights of key features in the time series data to improve the tracking accuracy; in specific implementation, the attention mechanism layer determines the key features by calculating the weighted sum of each time step, and the weight is determined by the correlation between the features of the time step and the overall data distribution.

[0063] The carbon emission tracking model ultimately outputs refined carbon emission tracking results, which can be directly used for accounting and reporting.

[0064] S4. Combine the refined carbon emission data with the pre-processed operation data to calculate carbon emissions. Figure 2 As shown, the specific steps include:

[0065] Step S4.1: integrating the refined carbon emission data obtained in step S3 with the operation data preprocessed in step S1 in a one-to-one correspondence, specifically, accurately matching the carbon emission data with the corresponding operation data through timestamps and data identifiers based on a data matching algorithm;

[0066] Step S4.2: The carbon emissions of each module in the photovoltaic storage and charging integrated system are calculated using the weighted average method, where the weight is set based on the operating time and efficiency of each module. The specific operation is to calculate the operating efficiency and operating time of each module within a certain period of time, and then calculate the weighted average value according to the preset formula: Carbon emissions = ∑ (operating time × efficiency × carbon emission factor);

[0067] Step S4.3: Record the accounting results, including the carbon emissions of each module and the total carbon emissions, and store these data in a database for subsequent analysis and report preparation.

[0068] S5. Based on the calculation results, conduct carbon emission optimization analysis and prepare a carbon emission report, which specifically includes the following steps:

[0069] Step S5.1: Analyze the calculation results in step S4 to determine the carbon emission contribution ratio of each module. The specific operation is to calculate the ratio of carbon emissions of each module to total carbon emissions. The formula is:

[0070] Carbon emission contribution ratio = module carbon emission ÷ total carbon emission × 100%;

[0071] Step S5.2: Prepare a carbon emissions report, including the carbon emissions of each module in the integrated solar-storage-charging system, total carbon emissions, carbon emissions contribution ratio and trend analysis of carbon emissions data. The specific operation is to use the report generation tool to automatically fill in the template. The template contains necessary tables and charts to intuitively display the accounting results, apply statistical analysis methods, such as moving average or exponential smoothing, to predict future carbon emissions trends, and provide these analysis results in the report.

[0072] In this embodiment, the carbon emission report is also matched with an interactive visualization interface, allowing users to filter and view carbon emission data according to different needs and conditions, as well as perform simulation forecasts.

[0073] Among them, the visualization interface is a web-based visualization interface, through which users can access the carbon emission report;

[0074] The data filter is integrated into the visualization interface, allowing users to filter carbon emission data according to time range, module type or other relevant parameters. The specific operation is to provide drop-down menus, date selectors and other input controls; the visualization interface provides a chart display area to dynamically display the filtering results. The specific operation is to update the chart in real time according to the user's selection, such as bar charts, line charts or pie charts; the visualization interface integrates simulation prediction tools, allowing users to simulate and predict future carbon emissions based on historical data. The specific operation is to provide a simulation interface where users can enter different parameters. The system calculates the prediction results based on these parameters and historical data and displays them in the chart.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for tracking and calculating carbon emissions in a photovoltaic storage and charging integrated system, characterized in that: The following steps are involved: S1. Collect and pre-process the operating data of each module in the photovoltaic storage and charging integrated system. The operating data includes photovoltaic power generation data, energy storage system charging and discharging data, and charging station power consumption data; S2. Input the preprocessed operation data into the pre-trained carbon emission estimation model to obtain carbon emission estimation data; S3. Inputting the carbon emission estimation data into a pre-trained carbon emission tracking model to obtain refined carbon emission data, wherein the carbon emission tracking model is constructed based on a deep learning convolutional neural network; S4. Combine the refined carbon emission data with the pre-processed operation data to perform carbon emission accounting; S5. Conduct carbon emission optimization analysis based on the accounting results and prepare a carbon emission report.

2. The method according to claim 1, characterized in that In step S1, the preprocessing method includes data cleaning, abnormal data marking and processing, feature extraction and dimensionality reduction, data denoising and data standardization.

3. The method according to claim 1, characterized in that In step S2, the carbon emission estimation model includes an input layer, a feature extraction layer, a hidden layer and an output layer, wherein the input layer is used to receive the preprocessed operation data. Its feature extraction layer uses a convolutional neural network to extract features from the input data, capturing local features and spatial correlations in the data. It further learns the nonlinear relationships and deep features in the data through multiple hidden layers. Its output layer is used to convert the output of the hidden layer into carbon emission estimation data.

4. The method according to claim 3, characterized in that During the training process of the carbon emission estimation model, mean square error is used as the loss function, and the Adam optimizer is used to update the model parameters. The parameters are adjusted by calculating the gradient of the loss function with respect to the parameters to minimize the loss function, and L2 regularization is used to prevent overfitting.

5. The method according to claim 1, characterized in that In step S3, the carbon emission tracking model is built based on the long short-term memory network, which uses the long short-term memory network to process time series data, capture the changing trends and patterns of carbon emission data over time, and track the changes in carbon emissions at each time point through the recursive structure of the LSTM network, and associate these changes with corresponding operations or modules.

6. The method according to claim 5, characterized in that The carbon emissions tracking model also includes an attention mechanism layer to identify and focus on the modules and time periods that have the greatest impact on carbon emissions.

7. The method according to claim 1, characterized in that Step S4 specifically includes: integrating the refined carbon emission data with the pre-processed operation data in a one-to-one correspondence; using the weighted average method to calculate the carbon emissions of each module in the integrated photovoltaic storage and charging system, where the weight is set based on the operation time and efficiency of each module.

8. The method according to claim 1, characterized in that Step S5 specifically includes: analyzing the calculation results to determine the carbon emission contribution ratio of each module; and preparing a carbon emission report including the carbon emissions of each module in the integrated photovoltaic storage and charging system, the total carbon emissions and the carbon emission contribution ratio.

9. The method according to claim 8, characterized in that The carbon emissions report is also equipped with an interactive visualization interface that allows users to filter and view carbon emissions data according to different needs and conditions, as well as perform simulation forecasts.

Images