Internet of things system rigid-flex board full life cycle carbon footprint accounting method and device

Abstract

The invention provides an Internet of Things system rigid-flex board full life cycle carbon footprint accounting method and device, and the method comprises the steps: carrying out the feature extraction according to carbon emission activity data, and generating carbon emission activity features; according to different life cycle stages, in combination with a carbon emission factor method, through a pre-trained carbon emission prediction model, prediction is carried out according to the carbon emission activity characteristics, and carbon emission prediction results corresponding to the life cycle stages are generated; and according to the carbon emission prediction result and the global warming potential value corresponding to the life cycle stage, generating a carbon footprint accounting result corresponding to the life cycle stage, thereby generating a full-life-cycle carbon footprint accounting report, and realizing accurate accounting of the full-life-cycle carbon footprint of the rigid-flex board. The carbon footprint is scientifically managed, the accuracy of carbon emission data is improved, meanwhile, the efficiency of carbon footprint accounting is improved, a real-time monitoring and early warning function is provided, and a transportation route is optimized to reduce carbon emission in the transportation process.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, particularly to the field of artificial intelligence technology, and more particularly to a method and device for calculating the full life cycle carbon footprint of a rigid-flex board in an Internet of Things system. Background Art

[0002] With the increasing severity of climate change, reducing carbon emissions has become an important task for enterprise operation. In the electronics manufacturing industry, the rigid-flex board, as a key component of complex electronic systems, is widely used in industries such as industry and medical care, and its usage has been continuously increasing in recent years. The carbon emissions generated during its production and use cannot be ignored. Due to the complex production process of the rigid-flex board, there are significant differences in the energy demand and usage efficiency of different production processes, the aging of production equipment will affect its energy efficiency and emission performance, and the supply chain involves multiple links and multiple suppliers and is difficult to coordinate, etc. In related technologies, data collection can only be carried out through the record sheets of each process link or hardware devices. Traditional carbon footprint calculation methods often have problems such as incomplete data collection, complex calculation processes, and large errors, and it is difficult to meet the needs of modern industry for accurate carbon footprint calculation. Summary of the Invention

[0003] An object of the present invention is to provide a method for calculating the full life cycle carbon footprint of a rigid-flex board in an Internet of Things system, which can achieve accurate calculation of the full life cycle carbon footprint of the rigid-flex board, scientifically manage the carbon footprint, improve the accuracy of carbon emission data, and at the same time improve the efficiency of carbon footprint calculation, provide a real-time monitoring and warning function, and optimize the transportation route to reduce carbon emissions during transportation. Another object of the present invention is to provide a device for calculating the full life cycle carbon footprint of a rigid-flex board in an Internet of Things system. Still another object of the present invention is to provide a computer-readable medium. Yet another object of the present invention is to provide a computer device.

[0004] To achieve the above objects, on the one hand, the present invention discloses a method for calculating the full life cycle carbon footprint of a rigid-flex board in an Internet of Things system, including:

[0005] Receiving the carbon emission activity data of the rigid-flex board sent by the Internet of Things system;

[0006] Performing feature extraction according to the carbon emission activity data to generate carbon emission activity features;

[0007] According to different life cycle stages, combining the carbon emission factor method, and through a pre-trained carbon emission prediction model, predicting according to the carbon emission activity features to generate a carbon emission prediction result corresponding to the life cycle stage;

[0008] Generate the carbon footprint accounting results corresponding to the life cycle stage based on the predicted carbon emissions results and global warming potential values corresponding to the life cycle stage;

[0009] Generate a carbon footprint accounting report for the entire life cycle based on the carbon footprint accounting results corresponding to each life cycle stage.

[0010] Preferably, before extracting features from the carbon emission activity data to generate carbon emission activity features, it further includes:

[0011] Clean the carbon emission activity data to generate the cleaned carbon emission activity data;

[0012] Standardize the cleaned carbon emission activity data to generate the standardized carbon emission activity data.

[0013] Preferably, extracting features from the carbon emission activity data to generate carbon emission activity features includes:

[0014] Extract features from the carbon emission activity data through the mean clustering algorithm to generate initial activity features;

[0015] Reduce the dimension of the initial activity features through the principal component analysis method to generate carbon emission activity features.

[0016] Preferably, the life cycle stage is the production stage, and the carbon emission activity features include production carbon emission factors, production direct energy consumption characteristics, and production indirect energy consumption characteristics;

[0017] According to different life cycle stages, combined with the carbon emission factor method, through a pre-trained carbon emission prediction model, predict based on the carbon emission activity features to generate the carbon emission prediction results corresponding to the life cycle stage, including:

[0018] Through a pre-constructed production direct carbon emission prediction model, predict the carbon emissions of production carbon emission factors and production direct energy consumption characteristics to generate production direct carbon emission prediction results;

[0019] Through a pre-constructed production indirect carbon emission prediction model, predict the carbon emissions of production carbon emission factors and production indirect energy consumption characteristics to generate production indirect carbon emission prediction results;

[0020] Generate the carbon emission prediction results corresponding to the production stage based on the production direct carbon emission prediction results and the production indirect carbon emission prediction results.

[0021] Preferably, the method further includes:

[0022] Obtain historical production direct energy consumption data and historical production indirect energy consumption data;

[0023] Using the association rule mining algorithm, data mining is respectively carried out on the historical production direct energy consumption data and the historical production indirect energy consumption data to generate a historical production direct multi-source data set and a historical production indirect multi-source data set;

[0024] Through the historical production direct multi-source data set, the random forest model and the convolutional neural network are trained to construct a production direct carbon emission prediction model;

[0025] Through the historical production indirect multi-source data set, the random forest model and the convolutional neural network are trained to construct a production indirect carbon emission prediction model.

[0026] Preferably, the life cycle stage is the transportation stage, and the carbon emission activity characteristics include the transportation vehicle energy consumption characteristics, the transportation distance characteristics, and the transportation tool carbon emission factor;

[0027] According to different life cycle stages, combined with the carbon emission factor method, through the pre-trained carbon emission prediction model, prediction is carried out according to the carbon emission activity characteristics to generate the carbon emission prediction results corresponding to the life cycle stage, including:

[0028] Through the pre-constructed transportation carbon emission prediction model, carbon emission prediction is carried out on the transportation vehicle energy consumption characteristics, the transportation distance characteristics, and the transportation tool carbon emission factor to generate the carbon emission prediction results corresponding to the transportation stage.

[0029] Preferably, the method further includes:

[0030] Obtain the historical transportation activity data set;

[0031] Perform data preprocessing on the historical transportation activity data set to generate the preprocessed historical transportation activity characteristics;

[0032] Through the historical transportation activity characteristics, the random forest model is trained to construct an initial transportation prediction model;

[0033] Through the Bayesian optimization algorithm, the initial transportation prediction model is optimized to construct a transportation carbon emission prediction model.

[0034] Preferably, the life cycle stage is the use stage, and the carbon emission activity characteristics include the energy consumption characteristics of the rigid-flexible printed circuit board and the use carbon emission factor;

[0035] According to different life cycle stages, combined with the carbon emission factor method, through the pre-trained carbon emission prediction model, prediction is carried out according to the carbon emission activity characteristics to generate the carbon emission prediction results corresponding to the life cycle stage, including:

[0036] Through a pre-built carbon emission prediction model, predict the carbon emissions based on the energy consumption characteristics of rigid-flex printed circuit boards and the carbon emission factors used, and generate the carbon emission prediction results corresponding to the usage stage.

[0037] Preferably, the method further includes:

[0038] Obtain the historical usage activity dataset;

[0039] Perform data preprocessing on the historical usage activity dataset to generate the preprocessed historical usage activity characteristics;

[0040] Train the support vector machine through the historical usage activity characteristics to construct an initial usage prediction model;

[0041] Optimize the initial usage prediction model through the Bayesian optimization algorithm to construct a carbon emission prediction model for usage.

[0042] Preferably, the life cycle stage is the waste treatment stage, and the carbon emission activity characteristics include the direct energy consumption characteristics of waste treatment, the indirect energy consumption characteristics of waste treatment, the energy consumption characteristics of waste transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools;

[0043] According to different life cycle stages, combined with the carbon emission factor method, through a pre-trained carbon emission prediction model, predict according to the carbon emission activity characteristics, and generate the carbon emission prediction results corresponding to the life cycle stage, including:

[0044] Through a pre-built carbon emission prediction model for waste transportation, predict the carbon emissions based on the energy consumption characteristics of waste transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools, and generate the carbon emission prediction results for waste transportation;

[0045] Through a pre-built direct carbon emission prediction model for waste disassembly, predict the carbon emissions based on the direct energy consumption characteristics of waste treatment, and generate the direct carbon emission prediction results for waste disassembly;

[0046] Through a pre-built indirect carbon emission prediction model for waste disassembly, predict the carbon emissions based on the indirect energy consumption characteristics of waste treatment, and generate the indirect carbon emission prediction results for waste disassembly;

[0047] Generate the carbon emission prediction results corresponding to the waste treatment stage based on the carbon emission prediction results for waste transportation, the direct carbon emission prediction results for waste disassembly, and the indirect carbon emission prediction results for waste disassembly.

[0048] Preferably, the method further includes:

[0049] Obtain the historical waste transportation activity dataset, the historical direct energy consumption data for waste treatment, and the historical energy consumption data for production waste treatment;

[0050] Preprocess the historical waste transportation activity dataset to generate the preprocessed historical waste transportation activity features;

[0051] Train the random forest model using the historical waste transportation activity features to construct an initial waste transportation prediction model;

[0052] Optimize the initial waste transportation prediction model using the Bayesian optimization algorithm to construct a waste transportation carbon emission prediction model;

[0053] Use the association rule mining algorithm to perform data mining on the historical direct energy consumption data for waste treatment and the historical energy consumption data for production waste treatment respectively to generate a historical direct multi-source dataset for waste treatment and a historical indirect multi-source dataset for waste treatment;

[0054] Train the random forest model and the convolutional neural network using the historical direct multi-source dataset for waste treatment to construct a waste disassembly direct carbon emission prediction model;

[0055] Train the random forest model and the convolutional neural network using the historical indirect multi-source dataset for waste treatment to construct a waste disassembly indirect carbon emission prediction model.

[0056] Preferably, generate the carbon footprint accounting results corresponding to the life cycle stage according to the carbon emission prediction results corresponding to the life cycle stage and the global warming potential value, including:

[0057] Multiply the global warming potential value by the carbon emission prediction results corresponding to the life cycle stage to generate the carbon footprint accounting results corresponding to the life cycle stage.

[0058] Preferably, generate a carbon footprint accounting report for the entire life cycle according to the carbon footprint accounting results corresponding to each life cycle stage, including:

[0059] Summarize the carbon footprint accounting results corresponding to each life cycle stage to generate the total carbon emissions of the rigid-flex board for the entire life cycle;

[0060] Generate a carbon footprint accounting report for the entire life cycle according to the total carbon emissions of the rigid-flex board for the entire life cycle and the carbon footprint accounting results corresponding to each life cycle stage, and visualize the carbon footprint accounting report for the entire life cycle.

[0061] The present invention also discloses a carbon footprint accounting device for the entire life cycle of a rigid-flex board in an Internet of Things system, including:

[0062] A receiving unit, configured to receive the carbon emission activity data of the hardware-software integrated board sent by the Internet of Things system;

[0063] A feature extraction unit, configured to extract features based on the carbon emission activity data to generate carbon emission activity features;

[0064] A carbon emission prediction unit, configured to, according to different life cycle stages, in combination with the carbon emission factor method, through a pre-trained carbon emission prediction model, predict according to the carbon emission activity features to generate a carbon emission prediction result corresponding to the life cycle stage;

[0065] A carbon footprint accounting unit, configured to generate a carbon footprint accounting result corresponding to the life cycle stage according to the carbon emission prediction result corresponding to the life cycle stage and the global warming potential value;

[0066] A full life cycle carbon footprint reporting unit, configured to generate a full life cycle carbon footprint accounting report according to the carbon footprint accounting results corresponding to each life cycle stage.

[0067] Preferably, the device further includes:

[0068] A data cleaning unit, configured to clean the carbon emission activity data to generate cleaned carbon emission activity data;

[0069] A data standardization unit, configured to perform standardization processing on the cleaned carbon emission activity data to generate standardized carbon emission activity data.

[0070] Preferably, the feature extraction unit is specifically configured to extract features from the carbon emission activity data through a mean clustering algorithm to generate initial activity features; and perform dimensionality reduction processing on the initial activity features through a principal component analysis method to generate carbon emission activity features.

[0071] Preferably, the life cycle stage is the production stage, and the carbon emission activity features include production carbon emission factors, production direct energy consumption features, and production indirect energy consumption features;

[0072] The carbon emission prediction unit is specifically configured to predict the carbon emissions of the production carbon emission factors and the production direct energy consumption features through a pre-constructed production direct carbon emission prediction model to generate a production direct carbon emission prediction result; predict the carbon emissions of the production carbon emission factors and the production indirect energy consumption features through a pre-constructed production indirect carbon emission prediction model to generate a production indirect carbon emission prediction result; and generate a carbon emission prediction result corresponding to the production stage according to the production direct carbon emission prediction result and the production indirect carbon emission prediction result.

[0073] Preferably, the device further includes:

[0074] A historical production data acquisition unit for acquiring historical direct energy consumption data and historical indirect energy consumption data of production;

[0075] A production data correlation mining unit for respectively performing data mining on the historical direct energy consumption data and historical indirect energy consumption data of production through an association rule mining algorithm to generate a historical direct multi-source data set and a historical indirect multi-source data set of production;

[0076] A training unit for a production direct carbon emission prediction model for training a random forest model and a convolutional neural network through the historical direct multi-source data set of production to construct a production direct carbon emission prediction model;

[0077] A training unit for a production indirect carbon emission prediction model for training a random forest model and a convolutional neural network through the historical indirect multi-source data set of production to construct a production indirect carbon emission prediction model.

[0078] Preferably, the life cycle stage is the transportation stage, and the carbon emission activity characteristics include transportation vehicle energy consumption characteristics, transportation distance characteristics, and transportation tool carbon emission factors;

[0079] A carbon emission prediction unit specifically for predicting carbon emissions of transportation vehicle energy consumption characteristics, transportation distance characteristics, and transportation tool carbon emission factors through a pre-constructed transportation carbon emission prediction model to generate a carbon emission prediction result corresponding to the transportation stage.

[0080] Preferably, the device further includes:

[0081] A historical transportation data acquisition unit for acquiring a historical transportation activity data set;

[0082] A historical transportation data preprocessing unit for preprocessing the historical transportation activity data set to generate preprocessed historical transportation activity characteristics;

[0083] An initial transportation prediction model training unit for training a random forest model through the historical transportation activity characteristics to construct an initial transportation prediction model;

[0084] A transportation carbon emission prediction model optimization unit for optimizing the initial transportation prediction model through a Bayesian optimization algorithm to construct a transportation carbon emission prediction model.

[0085] Preferably, the life cycle stage is the usage stage, and the carbon emission activity characteristics include the energy consumption characteristics of rigid-flex printed circuit boards and usage carbon emission factors;

[0086] The carbon emission prediction unit is specifically used to predict the carbon emissions by using the pre - constructed carbon emission prediction model for the energy consumption characteristics of the rigid - flexible printed circuit board and the carbon emission factors in use, and generate the carbon emission prediction results corresponding to the use stage.

[0087] Preferably, the device further includes:

[0088] The historical usage data acquisition unit is used to acquire the historical usage activity dataset;

[0089] The historical usage data pre - processing unit is used to pre - process the historical usage activity dataset and generate the pre - processed historical usage activity characteristics;

[0090] The initial usage prediction model training unit is used to train the support vector machine through the historical usage activity characteristics to construct the initial usage prediction model;

[0091] The carbon emission prediction model optimization unit in use is used to optimize the initial usage prediction model through the Bayesian optimization algorithm to construct the carbon emission prediction model in use.

[0092] Preferably, the life cycle stage is the waste treatment stage, and the carbon emission activity characteristics include the direct energy consumption characteristics of waste treatment, the indirect energy consumption characteristics of waste treatment, the energy consumption characteristics of waste transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools;

[0093] The carbon emission prediction unit is specifically used to predict the carbon emissions of waste transportation vehicles through the pre - constructed waste transportation carbon emission prediction model for the energy consumption characteristics of waste transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools, and generate the waste transportation carbon emission prediction results; predict the carbon emissions of direct waste dismantling through the pre - constructed direct carbon emission prediction model of waste dismantling for the direct energy consumption characteristics of waste treatment, and generate the direct carbon emission prediction results of waste dismantling; predict the carbon emissions of indirect waste dismantling through the pre - constructed indirect carbon emission prediction model of waste dismantling for the indirect energy consumption characteristics of waste treatment, and generate the indirect carbon emission prediction results of waste dismantling; generate the carbon emission prediction results corresponding to the waste treatment stage according to the waste transportation carbon emission prediction results, the direct carbon emission prediction results of waste dismantling, and the indirect carbon emission prediction results of waste dismantling.

[0094] Preferably, the device further includes:

[0095] The historical waste data acquisition unit is used to acquire the historical waste transportation activity dataset, the historical direct energy consumption data of waste treatment, and the historical energy consumption data of production waste treatment;

[0096] A historical waste data preprocessing unit for preprocessing a historical waste transportation activity dataset to generate preprocessed historical waste transportation activity features;

[0097] An initial waste transportation prediction model training unit for training a random forest model through historical waste transportation activity features to construct an initial waste transportation prediction model;

[0098] A waste transportation carbon emission prediction model optimization unit for optimizing the initial waste transportation prediction model through a Bayesian optimization algorithm to construct a waste transportation carbon emission prediction model;

[0099] A waste data association mining unit for respectively performing data mining on historical direct energy consumption data of waste treatment and historical energy consumption data of production waste treatment through an association rule mining algorithm to generate a historical direct multi-source dataset of waste treatment and a historical indirect multi-source dataset of waste treatment;

[0100] A waste dismantling direct carbon emission prediction model training unit for training a random forest model and a convolutional neural network through the historical direct multi-source dataset of waste treatment to construct a waste dismantling direct carbon emission prediction model;

[0101] A waste dismantling indirect carbon emission prediction model training unit for training a random forest model and a convolutional neural network through the historical indirect multi-source dataset of waste treatment to construct a waste dismantling indirect carbon emission prediction model.

[0102] Preferably, a carbon footprint accounting unit is specifically configured to multiply the global warming potential value by the carbon emission prediction result corresponding to the life cycle stage to generate a carbon footprint accounting result corresponding to the life cycle stage.

[0103] Preferably, a full life cycle carbon footprint reporting unit is specifically configured to summarize the carbon footprint accounting results corresponding to each life cycle stage to generate the total carbon emissions of the rigid-flex printed circuit board in the full life cycle; generate a full life cycle carbon footprint accounting report according to the total carbon emissions of the rigid-flex printed circuit board in the full life cycle and the carbon footprint accounting results corresponding to each life cycle stage, and visualize the full life cycle carbon footprint accounting report.

[0104] The present invention also discloses a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the above-mentioned method is implemented.

[0105] The present invention also discloses a computer device including a memory and a processor, the memory is used for storing information including program instructions, the processor is used for controlling the execution of the program instructions, and when the processor executes the program, the above-mentioned method is implemented.

[0106] The present invention also discloses a computer program product, including computer programs / instructions, which, when executed by a processor, implement the method as described above.

[0107] The present invention extracts features based on carbon emission activity data to generate carbon emission activity features; according to different life cycle stages, in combination with the carbon emission factor method, through a pre-trained carbon emission prediction model, predicts based on the carbon emission activity features to generate a carbon emission prediction result corresponding to the life cycle stage; generates a carbon footprint accounting result corresponding to the life cycle stage according to the carbon emission prediction result corresponding to the life cycle stage and the global warming potential value, thereby generating a carbon footprint accounting report for the entire life cycle, which can achieve accurate accounting of the carbon footprint of the rigid-flex printed circuit board throughout the life cycle, scientifically manage the carbon footprint, improve the accuracy of carbon emission data, and at the same time improve the efficiency of carbon footprint accounting, provide a real-time monitoring and early warning function, and optimize the transportation route to reduce carbon emissions during transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0108] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0109] Figure 1 It is a flowchart of a method for calculating the carbon footprint of a rigid-flex printed circuit board throughout the life cycle in an Internet of Things system provided by an embodiment of the present invention;

[0110] Figure 2 It is a flowchart of another method for calculating the carbon footprint of a rigid-flex printed circuit board throughout the life cycle in an Internet of Things system provided by an embodiment of the present invention;

[0111] Figure 3 It is a flowchart of predicting the carbon emissions in the production stage provided by an embodiment of the present invention;

[0112] Figure 4 It is a flowchart of predicting the carbon emissions in the transportation stage provided by an embodiment of the present invention;

[0113] Figure 5 It is a flowchart of predicting the carbon emissions in the usage stage provided by an embodiment of the present invention;

[0114] Figure 6 It is a flowchart of predicting the carbon emissions in the waste treatment stage provided by an embodiment of the present invention;

[0115] Figure 7Schematic diagram of a device for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system provided by an embodiment of the present invention;

[0116] Figure 8 Schematic diagram of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0117] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0118] It should be noted that a method and a device for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system disclosed in the present application can be used in the field of artificial intelligence technology, and can also be used in any field other than the field of artificial intelligence technology. The application field of the method and the device for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system disclosed in the present application is not limited.

[0119] To facilitate the understanding of the technical solutions provided in the present application, the relevant content of the technical solutions of the present application will be described first. The present invention proposes a method for calculating the full - life - cycle carbon footprint of a software - hardware integrated board based on an Internet of Things system, aiming to achieve accurate calculation and scientific management of the carbon footprint of the software - hardware integrated board through a software - hardware integrated system that is connected to production and logistics. The present invention builds a cloud data center, configures data collection and processing software, and develops a front - end user interface or application program; installs Internet of Things sensors and monitoring devices (such as smart meters, gas flow meters, GHG emission monitors, etc.) to collect carbon emission data in real - time during production, transportation, use, and waste treatment. The collected data is processed for calculating the carbon emissions in four stages. The carbon emissions in each stage of production, transportation, use, and waste treatment are summarized to obtain the total carbon emissions of the full - life - cycle of the software - hardware integrated board, and key points of carbon emissions and emission reduction suggestions are analyzed and given.

[0120] Next, taking the device for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system as the execution subject as an example, the implementation process of the method for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system provided by an embodiment of the present invention will be described. It can be understood that the execution subject of the method for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system provided by an embodiment of the present invention includes but is not limited to the device for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system.

[0121] Figure 1 Flowchart of a method for calculating the full - life - cycle carbon footprint of a software - hardware integrated board in an Internet of Things system provided by an embodiment of the present invention, asFigure 1 As shown, the method includes:

[0122] Step 101: Receive the carbon emission activity data of the rigid-flex board sent by the Internet of Things system.

[0123] In the embodiments of the present invention, the Internet of Things system includes, but is not limited to, Internet of Things sensors and detection devices. The Internet of Things sensors and detection devices are used to collect real-time carbon emission data at key nodes such as rigid-flex board production factories, transportation vehicles, usage sites, and waste treatment stations, including but not limited to data such as energy consumption, waste gas emissions, and waste treatment.

[0124] In the embodiments of the present invention, the Internet of Things system transmits the collected data to the cloud data center through a wireless network or a wired network to achieve centralized storage and management of the data. In the cloud data center, the collected data is cleaned, integrated, and analyzed using big data analysis and machine learning algorithms to calculate the carbon emissions at each stage.

[0125] Step 102: Extract features based on the carbon emission activity data to generate carbon emission activity features.

[0126] In the embodiments of the present invention, the quartile statistical method is used to detect some missing values and outliers in the carbon emission activity data. After removing the data points that do not conform to the normal production process caused by sensor failures, data entry errors, or other accidental factors, the missing values are filled by the interpolation method to achieve data cleaning; the Z-score standardization method is used to standardize the carbon emission activity data for subsequent analysis.

[0127] Specifically, features closely related to the carbon footprint are extracted from the standardized data using the k-means clustering method to generate carbon emission activity features.

[0128] Step 103: According to different life cycle stages, combined with the carbon emission factor method, through a pre-trained carbon emission prediction model, predict according to the carbon emission activity features to generate a carbon emission prediction result corresponding to the life cycle stage.

[0129] In the embodiments of the present invention, the dimension of the carbon emission activity features is reduced by the principal component analysis method (PCA), and the most important features are retained; the retained important features are divided into a training set and a test set. The training set is used to train the random forest model, and the test set is used to evaluate the performance of the model. The training set data is used to train the model by constructing multiple decision trees and voting or averaging to predict the result. The test set data is used to verify the accuracy of the model. The evaluation metrics include the mean square error (MSE) and the coefficient of determination (R 2 ) etc.

[0130] In the embodiments of the present invention, the inputs of the trained carbon emission prediction model are various activity data involved in the product production process, parameters related to carbon emissions, important features (principal component scores) retained after preprocessing and dimensionality reduction, original features, and other environmental factors. Among them, the various activity data involved in the product production process include, but are not limited to, energy consumption, material use, transportation distance, and processing process; the parameters related to carbon emissions include, but are not limited to, the carbon emission factors corresponding to each activity link, and the carbon emission factors corresponding to each activity link include, but are not limited to, production carbon emission factors, transportation tool carbon emission factors, and usage carbon emission factors. For example, the CO2 emissions generated per kilogram of energy consumption; the original features include, but are not limited to, product weight, production time, and energy type; the environmental factors include, but are not limited to, climate conditions and geographical regions. The carbon emission factors corresponding to each life cycle stage and each activity link can be obtained by querying relevant professional databases. For example, the carbon emission factors of raw materials in the production stage and the carbon emission factors of different transportation tools in the transportation stage. In the production stage, the carbon emission prediction results are mainly the direct emissions (process emissions and fugitive emissions) and indirect emissions (electricity and heat consumption) during the production process.

[0131] In the embodiments of the present invention, the outputs of the trained carbon emission prediction model include, but are not limited to, the predicted carbon emissions and the components of carbon emissions. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e); the components of carbon emissions include, but are not limited to, direct emissions (Scope 1), indirect emissions (Scope 2), and other related emissions (Scope 3).

[0132] In the embodiments of the present invention, the carbon emission prediction model can adopt the random forest algorithm combined with the carbon emission factor method, use historical data to train the model, continuously adjust the model parameters, and improve the calculation accuracy. The data set is divided into a training set and a validation set, and may also include a test set under certain circumstances. Through the k-fold cross-validation method, the accuracy and generalization ability of the model are verified. By using the random forest model to examine the relationships between different variables, understand which variables have positive or negative effects on the carbon footprint. At the same time, a prediction model is constructed to predict the carbon footprint according to various characteristics in the product production process, and is used to evaluate the impact of different production processes or product designs on the carbon footprint.

[0133] In the embodiments of the present invention, random forests are used for feature selection, and the feature input of a convolutional neural network (CNN) is optimized by recursively removing the features with the lowest importance, thereby improving the model performance. When performing model fusion, it is ensured that the datasets used by all models are consistent, and the same data preprocessing steps are carried out, including but not limited to feature selection, data cleaning, normalization, etc. One or more trained random forest models are used to predict the validation set or the test set, and the prediction results of each model are collected. A new model (learner) is used to combine the prediction results of other models to form a meta-model (meta-learner). An additional neural network layer is constructed to learn how to combine the outputs of different models. The prediction results of other random forest models are used as features to train the meta-model for final prediction.

[0134] Step 104: Generate the carbon footprint accounting result corresponding to the life cycle stage according to the carbon emission prediction result corresponding to the life cycle stage and the global warming potential value.

[0135] In the embodiments of the present invention, the carbon footprint activities of the whole life cycle of the rigid-flex printed circuit board are calculated according to the carbon emission prediction result corresponding to the life cycle stage output by the carbon emission prediction model. Specifically, the obtained global warming potential value (Global Warming Potential, abbreviated as: GWP) is multiplied by the carbon emission prediction result corresponding to the life cycle stage to generate the carbon footprint accounting result corresponding to the life cycle stage.

[0136] Step 105: Generate a carbon footprint accounting report for the whole life cycle according to the carbon footprint accounting results corresponding to each life cycle stage.

[0137] Specifically, the carbon footprint accounting results corresponding to each life cycle stage are summarized and analyzed to generate a carbon footprint accounting report for the whole life cycle.

[0138] Furthermore, the carbon footprint accounting results corresponding to each life cycle stage are displayed in various chart forms using an open-source chart visualization tool (Matplotlib), intuitively reflecting the sources of carbon emissions in complex production processes.

[0139] In the technical solution provided by the embodiment of the present invention, carbon emission activity data of a rigid-flex board is received from an Internet of Things (IoT) system; feature extraction is performed on the carbon emission activity data to generate carbon emission activity features; according to different life cycle stages, in combination with the carbon emission factor method, through a pre-trained carbon emission prediction model, carbon emissions are predicted based on the carbon emission activity features to generate carbon emission prediction results corresponding to the life cycle stages; according to the carbon emission prediction results corresponding to the life cycle stages and the global warming potential value, carbon footprint accounting results corresponding to the life cycle stages are generated; according to the carbon footprint accounting results corresponding to each life cycle stage, a carbon footprint accounting report for the entire life cycle is generated, which can achieve accurate accounting of the carbon footprint of the rigid-flex board throughout its life cycle, scientifically manage the carbon footprint, improve the accuracy of carbon emission data, and at the same time improve the efficiency of carbon footprint accounting, provide a real-time monitoring and early warning function, and optimize the transportation route to reduce carbon emissions during transportation.

[0140] Figure 2 FIG. is a flowchart of another method for calculating the carbon footprint of a rigid-flex board in the Internet of Things system provided by the embodiment of the present invention. As Figure 2 shown, the method includes:

[0141] Step 201: Receive carbon emission activity data of a rigid-flex board sent by an Internet of Things system.

[0142] In the embodiment of the present invention, each step is executed by a device for calculating the carbon footprint of a rigid-flex board in the Internet of Things system.

[0143] In the embodiment of the present invention, the Internet of Things system includes but is not limited to Internet of Things sensors and detection devices, including but not limited to energy consumption monitors, greenhouse gas (GHG) emission monitors, continuous emission monitoring systems (CEMS), and waste treatment tracking systems. The sensors are installed on the rigid-flex board to monitor energy consumption data such as its power consumption and heat consumption.

[0144] The energy consumption monitor is used to monitor the energy consumption of electricity, gas, steam, etc. of each production line and equipment (such as injection molding machines, ovens, cooling systems, etc.) in the factory. The energy consumption monitor includes but is not limited to smart meters, gas flow meters, and steam flow meters. A dedicated energy consumption monitor is installed on each production line as a carbon emission data collection point, and a unique identifier is generated for each energy consumption monitor using industrial Internet identification technology and the SHA-256 hashing algorithm.

[0145] The GHG emission monitor is installed at the exhaust port or chimney of the production workshop to monitor the emissions of greenhouse gases such as CO2 and NO x in the waste gas in real time. A GHG emission monitor is installed at each process point as a carbon emission data collection point, and a unique identifier is generated for each monitor using industrial Internet identification technology and the SHA-256 hashing algorithm.

[0146] The continuous emission monitoring system (CEMS) includes, but is not limited to, gas analyzers and flow monitoring devices for monitoring greenhouse gas components, concentrations, and flows.

[0147] The waste treatment tracking system is used to classify and weigh the waste generated during the production process and track the entire process of its transportation to the waste treatment station. The waste treatment tracking system includes, but is not limited to, RFID tags (for identifying product types), weighing sensors, GPS trackers (for transport vehicles), etc.

[0148] In the embodiments of the present invention, the Internet of Things system transmits the collected carbon emission activity data to the cloud data center through a wireless network or a wired network to achieve centralized storage and management of the data. Inside the factory, each data collection point can be connected to the data center through Ethernet to achieve stable and reliable data transmission. For wireless connection scenarios that require long distance and low power consumption, such as long-distance data transmission between the waste treatment station and the main plant area, technologies such as long-range radio (LoRa), narrowband Internet of Things (NB-IoT), or general packet radio service (GPRS) can be used.

[0149] The transport layer security / secure sockets layer (TLS / SSL) protocol is used to encrypt the transmitted data to ensure the security of the data during transmission. The collected data is transmitted to the cloud data center through the data transmission layer to achieve centralized storage and management of the data. A strict data management and access permission system is established to ensure that only authorized users can access and process the relevant data.

[0150] Step 202: Clean the carbon emission activity data to generate the cleaned carbon emission activity data.

[0151] Specifically, the quartile statistical method is used to detect some missing values and outliers in the carbon emission activity data. After removing the data points that do not conform to the normal production process caused by sensor failures, data entry errors, or other accidental factors, the missing values are filled by interpolation to generate the cleaned carbon emission activity data.

[0152] Step 203: Standardize the cleaned carbon emission activity data to generate the standardized carbon emission activity data.

[0153] Specifically, the Z-score standardization method is used to standardize the cleaned carbon emission activity data to generate the standardized carbon emission activity data for subsequent analysis.

[0154] Step 204: Extract features from the carbon emission activity data through the mean clustering algorithm to generate the initial activity features.

[0155] Specifically, the k-means clustering method was used to extract features closely related to the carbon footprint from the standardized data, generating initial activity features.

[0156] Step 205: Through the principal component analysis (PCA), dimensionality reduction processing was performed on the initial activity features to generate carbon emission activity features.

[0157] In the embodiment of the present invention, through PCA, dimensionality reduction processing was performed on the initial activity features, retaining the most important information to generate carbon emission activity features.

[0158] In the embodiment of the present invention, PCA effectively reduces the feature dimension of the dataset. Although the number of features is reduced, the features that contribute the most to the data variation are still retained, thereby reducing the computational complexity while maintaining the essential structure and key characteristics of the data as much as possible.

[0159] Step 206: According to different life cycle stages, combined with the carbon emission factor method, through a pre-trained carbon emission prediction model, predictions were made based on the carbon emission activity features to generate carbon emission prediction results corresponding to the life cycle stages.

[0160] In the embodiment of the present invention, the life cycle stage is the production stage, and the carbon emission activity features include production carbon emission factors, production direct energy consumption features, and production indirect energy consumption features. The production direct energy consumption features are the features collected by GHG emission monitors installed at each process flow point, including but not limited to greenhouse gas concentration and emissions; the production indirect energy consumption features are the features collected by special energy consumption metering instruments installed on each production line, including but not limited to greenhouse gas components, concentration, and flow rate.

[0161] Figure 3 It is a flowchart of carbon emission prediction in the production stage provided by the embodiment of the present invention. As Figure 3 shown, step 206 specifically includes:

[0162] Step 2061: Construct a production direct carbon emission prediction model and a production indirect carbon emission prediction model.

[0163] In the embodiment of the present invention, step 2061 specifically includes:

[0164] Step a1: Obtain historical production direct energy consumption data and historical production indirect energy consumption data.

[0165] In the embodiment of the present invention, the historical production direct energy consumption data and historical production indirect energy consumption data can be obtained from a PostgreSQL relational database. The PostgreSQL relational database is used to store the data collected by sensors.

[0166] Step a2: Using the association rule mining algorithm, perform data mining on the historical direct energy consumption data and historical indirect energy consumption data of production respectively to generate a historical direct multi-source dataset of production and a historical indirect multi-source dataset of production.

[0167] In the embodiments of the present invention, the association rule mining algorithm includes but is not limited to the Apriori algorithm or the FP-growth algorithm. Specifically, the association rule mining algorithm is used to identify frequent item sets. Candidate rules are generated based on the frequent item sets, and the support, confidence, and lift of each rule are calculated. Meaningful association rules are selected according to predetermined evaluation criteria (such as minimum support, minimum confidence). Using the selected association rules, relevant activity data are combined into a historical direct multi-source dataset / historical indirect multi-source dataset of production. The characteristics and distribution of the dataset are clarified through exploratory data analysis (EDA) techniques.

[0168] Step a3: Through the historical direct multi-source dataset of production, train the random forest model and the CNN to construct a production direct carbon emission prediction model.

[0169] Specifically, through the historical direct multi-source dataset of production, train the random forest model to generate a direct feature selection model and an initial prediction result; through the historical direct multi-source dataset, the direct feature extraction model, and the initial prediction result, train the CNN to generate a production direct carbon emission prediction model. Among them, the initial prediction result is the prediction result generated by the random forest model, and the direct feature selection model optimizes the feature input of the CNN by recursively removing the least important features, thereby improving the model performance.

[0170] Step a4: Through the historical indirect multi-source dataset of production, train the random forest model and the convolutional neural network to construct a production indirect carbon emission prediction model.

[0171] Specifically, through the historical indirect multi-source dataset of production, train the random forest model to generate an indirect feature selection model and an initial prediction result; through the historical indirect multi-source dataset, the indirect feature extraction model, and the initial prediction result, train the CNN to generate a production indirect carbon emission prediction model. Among them, the initial prediction result is the prediction result generated by the random forest model, and the indirect feature selection model optimizes the feature input of the CNN by recursively removing the least important features, thereby improving the model performance.

[0172] Step 2062: Through the pre-constructed production direct carbon emission prediction model, perform carbon emission prediction on the production carbon emission factor and the production direct energy consumption characteristics to generate a production direct carbon emission prediction result.

[0173] Specifically, the production carbon emission factor and the production direct energy consumption characteristics are input into the production direct carbon emission prediction model for carbon emission prediction, and the production direct carbon emission prediction result is generated. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e).

[0174] Step 2063: Through the pre-constructed production indirect carbon emission prediction model, carbon emission prediction is carried out on the production carbon emission factor and the production indirect energy consumption characteristics, and the production indirect carbon emission prediction result is generated.

[0175] Specifically, the production carbon emission factor and the production indirect energy consumption characteristics are input into the production indirect carbon emission prediction model for carbon emission prediction, and the production indirect carbon emission prediction result is generated. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e).

[0176] Step 2064: According to the production direct carbon emission prediction result and the production indirect carbon emission prediction result, the carbon emission prediction result corresponding to the production stage is generated.

[0177] Specifically, the production direct carbon emission prediction result and the production indirect carbon emission prediction result are added together to generate the carbon emission prediction result corresponding to the production stage.

[0178] In the embodiments of the present invention, the life cycle stage is the transportation stage, and the carbon emission activity characteristics include transportation vehicle energy consumption characteristics, transportation distance characteristics, and transportation tool carbon emission factors. The transportation vehicle energy consumption characteristics include, but are not limited to, collecting fuel consumption or power consumption data of transportation vehicles through Internet of Things devices or on-vehicle sensors; the transportation distance characteristics are the transportation distance characteristics calculated using GIS technology based on the starting point and ending point of the vehicle; the transportation tool carbon emission factors are the carbon emission factors of different transportation tools obtained from authoritative data sources.

[0179] Figure 4 For the flowchart of carbon emission prediction in the transportation stage provided by the embodiments of the present invention, as Figure 4 shown, step 206 specifically includes:

[0180] Step 3061: Construct a transportation carbon emission prediction model.

[0181] In the embodiments of the present invention, step 3061 specifically includes:

[0182] Step b1: Obtain a historical transportation activity dataset.

[0183] In the embodiments of the present invention, the historical transportation activity dataset includes transportation vehicle energy consumption characteristics, transportation distance characteristics, and transportation tool carbon emission factors, and the above data are integrated into the cloud data center of the present application. The historical transportation activity dataset can be obtained from a database.

[0184] Step b2: Perform data preprocessing on the historical transportation activity dataset to generate preprocessed historical transportation activity features.

[0185] Specifically, detect partial missing values and outliers in the historical transportation activity dataset through the quartile statistical method. After removing the data points that do not conform to the normal transportation process caused by sensor failures, data entry errors, or other accidental factors, then fill in the missing values through the interpolation method to generate a cleaned historical transportation activity dataset; use the Z-score standardization method to standardize the cleaned historical transportation activity dataset, convert the data into a unified format and dimension with the datasets of other stages to generate a standardized historical transportation activity dataset for subsequent analysis; the PostgreSQL relational database is used to store the data collected by the sensors; the k-means clustering method is used to extract the features closely related to the carbon footprint from the standardized data, namely, the historical transportation activity features.

[0186] In the embodiments of the present invention, the historical transportation activity features include but are not limited to transportation distance, fuel consumption or power consumption, and carbon emission factor.

[0187] Step b3: Train the random forest model through the historical transportation activity features to construct an initial transportation prediction model.

[0188] Specifically, input the historical transportation activity features into the random forest model for model training, establish the relationship between fuel consumption or power consumption, transportation distance, and carbon emission factor, and construct an initial transportation prediction model.

[0189] In the embodiments of the present invention, the main outputs of the initial transportation prediction model include: predicting the carbon footprint of the product based on the input features, usually in kilograms of CO2 equivalent (kg CO2e); the contribution degree of different features to the total carbon emissions, helping to analyze which factors have the greatest impact on carbon emissions; the estimated uncertainty of the prediction, including the upper and lower limits of the prediction interval; the relative importance score of each input feature, identifying which features are the most critical for carbon emission prediction; the confidence score for each prediction result, reflecting the confidence level of the model in the prediction result; the trend prediction of carbon emissions, mainly the carbon emissions changing over time.

[0190] Step b4: Optimize the initial transportation prediction model through the Bayesian optimization algorithm to construct a transportation carbon emission prediction model.

[0191] In the embodiments of the present invention, use the Bayesian optimization technique combined with the carbon emission factor method to calculate the carbon emissions during the product transportation process, and use the k-fold cross-validation method to verify the accuracy of the model calculation to construct a transportation carbon emission prediction model. Specifically, it includes:

[0192] 1. Define the optimization objective: Minimize the total carbon emissions of the product or the carbon emissions per unit of product as the optimization objective function.

[0193] 2. Construct a Bayesian optimization framework: Adopt Gaussian Process Optimization (GPO for short) and apply it to the random forest model. Establish a probability model (such as a Gaussian process model) to represent the uncertainty predicted by the random forest model.

[0194] 3. Initialize Bayesian optimization: Provide initial parameters for the Bayesian optimization algorithm, including the initial search space and parameter distribution.

[0195] 4. Execute Bayesian optimization: Conduct iterative search through the Bayesian optimization algorithm, continuously adjusting the input parameters (such as activity data) of the random forest model. In each iteration, use the random forest model to predict the adjusted carbon emissions and feedback the prediction results to the Bayesian model. The Bayesian model updates the parameter distribution according to the prediction results to guide the next search.

[0196] 5. Evaluate the optimization results: Evaluate the optimized carbon emissions to determine whether the expected optimization objective has been achieved. Analyze which parameter adjustments are most effective in reducing carbon emissions during the optimization process.

[0197] 6. Iteration and improvement: If the optimization results are not satisfactory, continue to iterate, adjust the parameters of Bayesian optimization, or adjust the random forest model until the optimization results reach the ideal conditions to construct a transportation carbon emissions prediction model.

[0198] Step 3062: Through the pre-constructed transportation carbon emissions prediction model, predict the carbon emissions of the transportation vehicle energy consumption characteristics, transportation distance characteristics, and transportation tool carbon emission factors, and generate the carbon emissions prediction results corresponding to the transportation stage.

[0199] Specifically, input the transportation vehicle energy consumption characteristics, transportation distance characteristics, and transportation tool carbon emission factors into the transportation carbon emissions prediction model for carbon emissions prediction, and generate the carbon emissions prediction results corresponding to the transportation stage. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e).

[0200] In the embodiment of the present invention, the life cycle stage is the usage stage, and the carbon emission activity characteristics include the energy consumption characteristics of the rigid-flex board and the usage carbon emission factor; the energy consumption characteristics of the rigid-flex board include, but are not limited to, the energy consumption data such as the power consumption and heat consumption of the rigid-flex board monitored by sensors installed on the rigid-flex board. The sensor data is transmitted to the cloud data center in real time through GPRS; the PostgreSQL relational database is used to store the data collected by the sensors for subsequent analysis.

[0201] Figure 5 The flowchart of predicting the carbon emission amount in the usage stage provided by the embodiment of the present invention is as Figure 5 shown, and step 206 specifically includes:

[0202] Step 4061: Construct a usage carbon emission prediction model.

[0203] In the embodiment of the present invention, step 4061 specifically includes:

[0204] Step c1: Obtain a historical usage activity dataset.

[0205] In the embodiment of the present invention, the historical usage activity dataset includes energy consumption data such as the power consumption and heat consumption of transporting the rigid-flex board, and the above data is integrated into the cloud data center of the present application. The historical transportation activity dataset can be obtained from the database.

[0206] Step c2: Perform data preprocessing on the historical usage activity dataset to generate preprocessed historical usage activity characteristics.

[0207] Specifically, through the quartile statistical method, some missing values and outliers existing in the historical usage activity dataset are detected. After removing the data points that do not conform to the normal transportation process caused by sensor failures, data entry errors or other accidental factors, the missing values are filled by the interpolation method to generate a cleaned historical usage activity dataset; the Z-score normalization method is used to perform normalization processing on the cleaned historical usage activity dataset to convert the data into a unified format and dimension with the datasets of other stages to generate a normalized historical usage activity dataset for subsequent analysis; the PostgreSQL relational database is used to store the data collected by the sensors; the k-means clustering method is used to extract the characteristics closely related to the carbon footprint from the normalized data for the historical usage activity dataset to generate historical usage activity characteristics.

[0208] In the embodiment of the present invention, the historical usage activity characteristics include, but are not limited to, the energy consumption amount and the usage time, and the energy consumption amount includes, but is not limited to, the power consumption and heat consumption.

[0209] Step c3: Train the model of the support vector machine (SVM) using historical usage activity features to construct an initial usage prediction model.

[0210] Specifically, select the support vector machine (SVM) to train the model with data such as the energy consumption data (such as power consumption and heat energy consumption), usage time, and service life of the rigid-flex printed circuit board, establish the relationship between energy consumption and carbon emissions, and construct an initial usage prediction model.

[0211] Step c4: Optimize the initial usage prediction model using the Bayesian optimization algorithm to construct a usage carbon emission prediction model.

[0212] In the embodiment of the present invention, the Bayesian optimization technique is combined with the carbon emission factor method to calculate the carbon emissions during the product usage process, and the k-fold cross-validation method is used to verify the accuracy of the model calculation, and a usage carbon emission prediction model is constructed. Specifically, it includes:

[0213] 1. Define the optimization objective: Use minimizing the total carbon emissions of the product or the carbon emissions per unit product as the objective function for optimization.

[0214] 2. Construct the Bayesian optimization framework: Adopt Gaussian Process Optimization (GPO) and apply it to SVM. Establish a probability model (such as a Gaussian process model) to represent the uncertainty predicted by the random forest model.

[0215] 3. Initialize the Bayesian optimization: Provide the initial parameters for the Bayesian optimization algorithm, including the initial search space and parameter distribution.

[0216] 4. Execute the Bayesian optimization: Perform iterative search through the Bayesian optimization algorithm, and continuously adjust the input parameters (such as activity data) of the SVM model. In each iteration, use the SVM model to predict the adjusted carbon emissions and feed the prediction results back to the Bayesian model. The Bayesian model updates the parameter distribution according to the prediction results to guide the next search.

[0217] 5. Evaluate the optimization results: Evaluate the optimized carbon emissions to determine whether the expected optimization objective has been achieved. Analyze which parameter adjustments are most effective in reducing carbon emissions during the optimization process.

[0218] 6. Iteration and improvement: If the optimization results are not satisfactory, continue to iterate, adjust the parameters of the Bayesian optimization, or adjust the SVM model until the optimization results reach the ideal conditions, and construct a usage carbon emission prediction model.

[0219] Step 4062: Predict the carbon emissions of the rigid-flex printed circuit board by using the pre-constructed carbon emissions prediction model for energy consumption characteristics and carbon emission factors during use, and generate the carbon emissions prediction results corresponding to the use stage.

[0220] Specifically, input the energy consumption characteristics of the rigid-flex printed circuit board and the carbon emission factors during use into the carbon emissions prediction model for carbon emissions prediction, and generate the carbon emissions prediction results corresponding to the use stage. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e).

[0221] In the embodiments of the present invention, the life cycle stage is the waste treatment stage, and the carbon emission activity characteristics include the direct energy consumption characteristics of waste treatment, the indirect energy consumption characteristics of waste treatment, the energy consumption characteristics of waste transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools; the energy consumption characteristics of waste transportation vehicles include, but are not limited to, collecting the fuel consumption or power consumption data of transportation vehicles through Internet of Things devices or on-vehicle sensors; the transportation distance characteristics are the transportation distance characteristics calculated by using GIS technology based on the starting point and ending point of the vehicle; the carbon emission factors of transportation tools are obtained from authoritative data sources for different transportation tools; the direct energy consumption characteristics of waste treatment are the characteristics collected by GHG emission monitors installed at each process point, including but not limited to greenhouse gas concentration and emissions; the indirect energy consumption characteristics of waste treatment are the characteristics collected by dedicated energy consumption metering instruments installed on each production line, including but not limited to greenhouse gas composition, concentration, and flow rate.

[0222] Establish a recycling mechanism for the production waste and products of rigid-flex printed circuit boards. After the production waste or the product is monitored to be no longer in use, collect the recycling records of the discarded rigid-flex printed circuit boards, including the recycling time, location, quantity, etc. Collect the data during the disassembly and treatment process, such as disassembly energy consumption, material classification after disassembly, etc. Collect the carbon emission source data of the final treatment method, such as the quantity and method of landfill, incineration, and recycling.

[0223] Figure 6 The flowchart of carbon emissions prediction in the waste treatment stage provided by the embodiments of the present invention is as Figure 6 shown. Step 206 specifically includes:

[0224] Step 5061: Construct a carbon emissions prediction model for waste transportation, a direct carbon emissions prediction model for waste disassembly, and an indirect carbon emissions prediction model for waste disassembly.

[0225] In the embodiments of the present invention, step 5061 specifically includes:

[0226] Step d1: Obtain the historical waste transportation activity dataset, the historical direct energy consumption data of waste treatment, and the historical energy consumption data of production waste treatment.

[0227] In the embodiments of the present invention, the historical waste transportation activity dataset includes the energy consumption characteristics of transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools, and the above data are integrated into the cloud data center of the present application. The historical transportation activity dataset can be obtained from the database.

[0228] Step d2: Perform data preprocessing on the historical waste transportation activity dataset to generate the preprocessed historical waste transportation activity characteristics.

[0229] Specifically, the quartile statistical method is used to detect some missing values and outliers in the historical waste transportation activity dataset. After removing the data points that do not conform to the normal transportation process caused by sensor failures, data entry errors, or other accidental factors, the missing values are filled by the interpolation method to generate the cleaned historical waste transportation activity dataset; the Z-score standardization method is used to standardize the cleaned historical waste transportation activity dataset to convert the data into a unified format and dimension with the datasets in other stages, generating the standardized historical waste transportation activity dataset for subsequent analysis; the PostgreSQL relational database is used to store the data collected by the sensors; the k-means clustering method is used to extract the characteristics closely related to the carbon footprint from the standardized data for the historical waste transportation activity dataset, generating the historical waste transportation activity characteristics.

[0230] Step d3: Train the random forest model through the historical waste transportation activity characteristics to construct an initial waste transportation prediction model.

[0231] Specifically, the historical waste transportation activity characteristics are input into the random forest model for model training to establish the relationship between fuel consumption or power consumption, transportation distance, and carbon emission factors, and an initial waste transportation prediction model is constructed.

[0232] In the embodiments of the present invention, the main outputs of the initial waste transportation prediction model include: predicting the carbon footprint of the product according to the input characteristics, usually in kilograms of CO2 equivalent (kg CO2e); the contribution degree of different characteristics to the total carbon emissions to help analyze which factors have the greatest impact on carbon emissions; the estimated uncertainty of the prediction, including the upper and lower limits of the prediction interval; the relative importance score of each input characteristic to identify which characteristics are the most critical for carbon emission prediction; the confidence score for each prediction result to reflect the confidence level of the model in the prediction result; the trend prediction of carbon emissions, mainly the carbon emissions changing over time.

[0233] Step d4: Optimize the initial waste transportation prediction model through the Bayesian optimization algorithm to construct a waste transportation carbon emission prediction model.

[0234] In the embodiments of the present invention, the Bayesian optimization technique is used in combination with the carbon emission factor method to calculate the carbon emissions during the waste transportation process. The k-fold cross-validation method is adopted to verify the accuracy of the model calculation, and a prediction model for the carbon emissions of waste transportation is constructed. Specifically, it includes:

[0235] 1. Define the optimization objective: Minimize the total carbon emissions of the product or the carbon emissions per unit product as the optimization objective function.

[0236] 2. Construct a Bayesian optimization framework: Adopt Gaussian Process Optimization (GPO for short) and apply it to the random forest model. Establish a probability model (such as a Gaussian process model) to represent the uncertainty of the random forest model prediction.

[0237] 3. Initialize Bayesian optimization: Provide the initial parameters for the Bayesian optimization algorithm, including the initial search space and parameter distribution.

[0238] 4. Execute Bayesian optimization: Conduct iterative search through the Bayesian optimization algorithm, and continuously adjust the input parameters (such as activity data) of the random forest model. In each iteration, use the random forest model to predict the adjusted carbon emissions and feedback the prediction results to the Bayesian model. The Bayesian model updates the parameter distribution according to the prediction results to guide the next search.

[0239] 5. Evaluate the optimization results: Evaluate the optimized carbon emissions to determine whether the expected optimization goal has been achieved. Analyze which parameter adjustments are most effective in reducing carbon emissions during the optimization process.

[0240] 6. Iteration and improvement: If the optimization results are not satisfactory, continue to iterate, adjust the parameters of Bayesian optimization, or adjust the random forest model until the optimization results reach the ideal conditions, and construct a prediction model for transportation carbon emissions.

[0241] Step d5: Through the association rule mining algorithm, perform data mining on the historical direct energy consumption data of waste treatment and the historical energy consumption data of production waste treatment respectively to generate a historical direct multi-source dataset for waste treatment and a historical indirect multi-source dataset for waste treatment.

[0242] In the embodiments of the present invention, the association rule mining algorithm includes, but is not limited to, the Apriori algorithm or the FP-growth algorithm. Specifically, the association rule mining algorithm is used to identify frequent item sets. Candidate rules are generated based on the frequent item sets, and the support, confidence, and lift of each rule are calculated. Meaningful association rules are screened out according to predetermined evaluation criteria (such as minimum support and minimum confidence). Using the screened association rules, relevant activity data are combined into a historical waste treatment direct multi-source dataset / historical waste treatment indirect multi-source dataset. The characteristics and distribution of the dataset are clarified through exploratory data analysis (EDA) techniques.

[0243] Step d6: Train the random forest model and the convolutional neural network through the historical waste treatment direct multi-source dataset to construct a waste dismantling direct carbon emission prediction model.

[0244] Specifically, train the random forest model through the historical waste treatment direct multi-source dataset to generate a direct feature selection model and an initial prediction result; train the CNN through the historical waste treatment direct multi-source dataset, the direct feature extraction model, and the initial prediction result to generate a waste dismantling direct carbon emission prediction model. Among them, the initial prediction result is the prediction result generated by the random forest model, and the direct feature selection model optimizes the feature input of the CNN by recursively removing the least important features, thereby improving the model performance.

[0245] Step d7: Train the random forest model and the convolutional neural network through the historical waste treatment indirect multi-source dataset to construct a waste dismantling indirect carbon emission prediction model.

[0246] Specifically, train the random forest model through the historical waste treatment indirect multi-source dataset to generate an indirect feature selection model and an initial prediction result; train the CNN through the historical waste treatment indirect multi-source dataset, the indirect feature extraction model, and the initial prediction result to generate a waste dismantling indirect carbon emission prediction model. Among them, the initial prediction result is the prediction result generated by the random forest model, and the indirect feature selection model optimizes the feature input of the CNN by recursively removing the least important features, thereby improving the model performance.

[0247] Step 5062: Predict the carbon emissions of the waste transportation vehicle energy consumption characteristics, transportation distance characteristics, and transportation tool carbon emission factors through the pre-constructed waste transportation carbon emission prediction model to generate a waste transportation carbon emission prediction result.

[0248] Specifically, the energy consumption characteristics, transportation distance characteristics of waste transportation vehicles, and carbon emission factors of transportation tools are input into the waste transportation carbon emission prediction model to predict carbon emissions, generating the waste transportation carbon emission prediction results. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e).

[0249] Step 5063: Through the pre-constructed direct carbon emission prediction model for waste disassembly, predict the carbon emissions of the direct energy consumption characteristics of waste treatment, generating the direct carbon emission prediction results for waste disassembly.

[0250] Specifically, input the direct energy consumption characteristics of waste treatment into the direct carbon emission prediction model for waste disassembly to predict carbon emissions, generating the direct carbon emission prediction results for waste disassembly. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e).

[0251] Step 5064: Through the pre-constructed indirect carbon emission prediction model for waste disassembly, predict the carbon emissions of the indirect energy consumption characteristics of waste treatment, generating the indirect carbon emission prediction results for waste disassembly.

[0252] Specifically, input the indirect energy consumption characteristics of waste treatment into the indirect carbon emission prediction model for waste disassembly to predict carbon emissions, generating the indirect carbon emission prediction results for waste disassembly. Carbon emissions are usually measured in kilograms of CO2 equivalent (kg CO2e).

[0253] Step 5065: According to the waste transportation carbon emission prediction results, the direct carbon emission prediction results for waste disassembly, and the indirect carbon emission prediction results for waste disassembly, generate the carbon emission prediction results corresponding to the waste treatment stage.

[0254] Specifically, add the waste transportation carbon emission prediction results, the direct carbon emission prediction results for waste disassembly, and the indirect carbon emission prediction results for waste disassembly to generate the carbon emission prediction results corresponding to the waste treatment stage.

[0255] Step 207: Multiply the global warming potential value by the carbon emission prediction results corresponding to the life cycle stage to generate the carbon footprint accounting results corresponding to the life cycle stage.

[0256] In the embodiments of the present invention, the GWP value can be obtained from the reports released by authoritative organizations.

[0257] Specifically, multiply the GWP value by the carbon emission prediction results corresponding to the life cycle stage to generate the carbon footprint accounting results corresponding to the life cycle stage. For example, in the production stage, the carbon emission prediction results are mainly the direct emissions (process emissions and fugitive emissions) and indirect emissions (electricity and heat consumption) during the production process.

[0258] Step 208: Summarize the carbon footprint accounting results corresponding to each life cycle stage to generate the total carbon emissions of the rigid-flex printed circuit board throughout its life cycle.

[0259] In the embodiments of the present invention, the carbon footprint accounting results corresponding to each life cycle stage are added together to generate the total carbon emissions of the rigid-flex printed circuit board throughout its life cycle.

[0260] Step 209: Generate a carbon footprint accounting report for the entire life cycle based on the total carbon emissions of the rigid-flex printed circuit board throughout its life cycle and the carbon footprint accounting results corresponding to each life cycle stage, and visualize the carbon footprint accounting report for the entire life cycle.

[0261] In the embodiments of the present invention, the carbon emissions in each stage of production, transportation, use, and waste treatment are summarized to obtain the total carbon emissions of the rigid-flex printed circuit board throughout its life cycle.

[0262] In the embodiments of the present invention, based on the total carbon emissions of the rigid-flex printed circuit board throughout its life cycle and the carbon footprint accounting results corresponding to each life cycle stage, a carbon footprint accounting report for the entire life cycle is compiled to provide enterprises with detailed carbon emission data, emission reduction potential analysis, and optimization suggestions. The carbon footprint accounting report for the entire life cycle includes, but is not limited to, product carbon footprint, emission reduction hotspots, emission reduction potential analysis, cost-benefit of emission reduction measures, and emission reduction strategies.

[0263] Among them, calculating the product carbon footprint includes: calculating the carbon footprint of the product life cycle using the constructed model and analyzing the carbon emission contributions of each stage; identifying emission reduction hotspots includes: analyzing the carbon footprint calculation results to identify the stage or activity with the largest carbon emissions, thereby identifying emission reduction hotspots; emission reduction potential analysis includes: for each emission reduction hotspot, analyzing possible emission reduction measures, such as improving processes, using low-carbon materials, optimizing transportation routes, etc., and estimating the potential emission reduction amount of each emission reduction measure; evaluating the cost-benefit of emission reduction measures includes analyzing the input costs and potential economic benefits of each emission reduction measure, including saved energy costs, reduced carbon emission fines, etc.; formulating emission reduction strategies includes: formulating specific emission reduction strategies and action plans based on emission reduction potential analysis and cost-benefit evaluation.

[0264] In the embodiments of the present invention, in the production stage, the system dynamics method can also be used for hyperparameter optimization to formulate emission reduction strategies, such as optimizing transportation routes, replacing low-emission vehicles, etc., including but not limited to:

[0265] 1. Construct a system dynamics model according to each stage of the product life cycle. The model should include links such as raw material acquisition, manufacturing, transportation, use, and waste treatment. Determine the variables and parameters in the model, mainly the carbon emission factors, energy consumption, material use, etc. of various activities.

[0266] 2. Collect data related to the product carbon footprint, including historical carbon emission data, energy consumption data, production activity data, etc. Determine the values of various parameters in the model, which will directly affect the output results of the model.

[0267] 3. Determine the hyperparameters that have a significant impact on the carbon footprint calculation results, mainly the energy efficiency, transportation distance, and raw material selection in the production process.

[0268] 4. Observe the changes in the product's carbon footprint under different hyperparameter settings. Analyze the impact of hyperparameter changes on system behavior, including carbon emissions, costs, and efficiency.

[0269] 5. Use optimization algorithms (such as genetic algorithms, particle swarm optimization, Bayesian optimization, etc.) to optimize the hyperparameters.

[0270] 6. Based on the optimization results, formulate specific emission reduction strategies, including improving production processes, optimizing the supply chain, and using low-carbon materials. Analyze the cost-effectiveness and implementation difficulties of different emission reduction strategies.

[0271] In the embodiment of the present invention, during the use stage, genetic algorithms and particle swarm optimization algorithms can also be used to search the optimization solution space and generate optimized carbon emission strategies.

[0272] The present invention realizes the accurate accounting of the carbon footprint of the rigid-flex printed circuit board throughout its life cycle, scientifically manages the carbon footprint, improves the accuracy of carbon emission data, and at the same time improves the efficiency of carbon footprint accounting, provides a real-time monitoring and early warning function, and optimizes the transportation route to reduce carbon emissions during transportation.

[0273] The technical solution of the present invention realizes the accurate monitoring and tracking of carbon emissions in each link through the integration of advanced sensor technology, the Internet of Things, and data acquisition and analysis systems, improving the accuracy and reliability of carbon emission data. The system can collect emission data in real time during production, transportation, energy use and other stages, comprehensively track the carbon footprint throughout the life cycle, help enterprises fully understand their carbon emission situations, and propose optimization measures to effectively reduce carbon emissions. Through intelligent analysis, enterprises can optimize energy utilization, reduce resource waste, and improve resource utilization efficiency. In addition, the data transparency and visualization functions provided by the system not only meet regulatory compliance requirements, but also help enterprises enhance their environmental protection images and promote green development. This technical solution also promotes the innovation of energy conservation and emission reduction technologies, provides data support for enterprises, and promotes the application of more efficient and energy-saving processes. Through precise monitoring, enterprises can reduce energy consumption and abnormal emissions, reduce operating costs, and improve economic benefits. In summary, the present invention provides an efficient and accurate carbon emission management solution for enterprises, which not only helps enterprises achieve green and sustainable development, but also enhances their market competitiveness and social responsibility.

[0274] The full-life-cycle carbon footprint accounting method is the most widely used and highly recognized carbon emission accounting method. In addition to the carbon emission monitoring solution based on Internet of Things sensors and big data analysis adopted in the present invention, a solution based on artificial intelligence and machine learning can also be adopted. Through prediction and optimization algorithms, historical and real-time data can be combined to make a relatively accurate prediction of carbon emissions, so as to achieve dynamic adjustment of the carbon footprint during the production and operation process. As another alternative solution, a data platform integrating data collection, storage, processing and analysis is constructed; historical energy consumption data, raw material consumption data, production environment data, etc. of the flexible-rigid printed circuit board production are collected through the key performance indicator method, big data text mining and semantic correlation analysis; the long short-term memory network (LSTM) of the deep learning model is used to perform time series processing on the collected data to form an analysis model; the LDA topic clustering method is used to train the model, including adjusting the network structure, learning rate and regularization parameters; the system dynamics method is used to input real-time data to predict carbon emissions of the trained model; according to the prediction results, the working state of the production equipment is dynamically adjusted, such as reducing energy consumption and optimizing the production process; the Delphi method combined with the fuzzy analytic hierarchy process is used to regularly evaluate the effect of the dynamic adjustment measures to verify the accuracy of the model and the effectiveness of the adjustment strategy.

[0275] It should be noted that in the technical solution of this application, the acquisition, storage, use, processing, etc. of data all comply with the relevant regulations of laws and regulations. The user information in the embodiments of this application is obtained through legal and compliant channels, and the acquisition, storage, use, processing, etc. of the user information have obtained the authorization and consent of the customers.

[0276] It should be noted that the information collected in this application is information and data authorized by the users or fully authorized by all parties, and the processing of the relevant data, such as collection, storage, use, processing, transmission, provision, disclosure and application, all comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0277] It should be noted that the technical solution provided in this application provides corresponding operation entrances for users to choose to agree or refuse the automated decision-making results; if the user chooses to refuse, the expert decision-making process will be entered.

[0278] In the technical solution of the method for calculating the full - life - cycle carbon footprint of the software - hardware integrated board of the Internet of Things system provided by the embodiment of the present invention, carbon emission activity data of the software - hardware integrated board sent by the Internet of Things system is received; feature extraction is performed according to the carbon emission activity data to generate carbon emission activity features; according to different life - cycle stages, combined with the carbon emission factor method, through a pre - trained carbon emission prediction model, prediction is performed according to the carbon emission activity features to generate a carbon emission prediction result corresponding to the life - cycle stage; according to the carbon emission prediction result corresponding to the life - cycle stage and the global warming potential value, a carbon footprint calculation result corresponding to the life - cycle stage is generated; according to the carbon footprint calculation results corresponding to each life - cycle stage, a full - life - cycle carbon footprint calculation report is generated, which can achieve accurate calculation of the full - life - cycle carbon footprint of the software - hardware integrated board, scientifically manage the carbon footprint, improve the accuracy of carbon emission data, and at the same time improve the efficiency of carbon footprint calculation, provide a real - time monitoring and early - warning function, and optimize the transportation route to reduce carbon emissions during transportation.

[0279] Figure 7 FIG. is a schematic structural diagram of an apparatus for calculating the full - life - cycle carbon footprint of the software - hardware integrated board of the Internet of Things system provided by the embodiment of the present invention. This apparatus is used to execute the above - mentioned method for calculating the full - life - cycle carbon footprint of the software - hardware integrated board of the Internet of Things system, as Figure 7 shown. The apparatus includes: a receiving unit 11, a feature extraction unit 12, a carbon emission prediction unit 13, a carbon footprint calculation unit 14, and a full - life - cycle carbon footprint report unit 15.

[0280] The receiving unit 11 is used to receive the carbon emission activity data of the software - hardware integrated board sent by the Internet of Things system.

[0281] The feature extraction unit 12 is used to perform feature extraction according to the carbon emission activity data to generate carbon emission activity features.

[0282] The carbon emission prediction unit 13 is used to perform prediction according to different life - cycle stages, combined with the carbon emission factor method, through a pre - trained carbon emission prediction model, according to the carbon emission activity features, to generate a carbon emission prediction result corresponding to the life - cycle stage.

[0283] The carbon footprint calculation unit 14 is used to generate a carbon footprint calculation result corresponding to the life - cycle stage according to the carbon emission prediction result corresponding to the life - cycle stage and the global warming potential value.

[0284] The full - life - cycle carbon footprint report unit 15 is used to generate a full - life - cycle carbon footprint calculation report according to the carbon footprint calculation results corresponding to each life - cycle stage.

[0285] In the embodiment of the present invention, the apparatus further includes: a data cleaning unit 16 and a data standardization unit 17.

[0286] The data cleaning unit 16 is used to clean the carbon emission activity data and generate the cleaned carbon emission activity data.

[0287] The data standardization unit 17 is used to standardize the cleaned carbon emission activity data and generate the standardized carbon emission activity data.

[0288] In the embodiment of the present invention, the feature extraction unit 12 is specifically configured to extract features from the carbon emission activity data through the mean clustering algorithm to generate initial activity features; and perform dimensionality reduction processing on the initial activity features through the principal component analysis method to generate carbon emission activity features.

[0289] In the embodiment of the present invention, the life cycle stage is the production stage, and the carbon emission activity features include production carbon emission factors, production direct energy consumption features, and production indirect energy consumption features; the carbon emission prediction unit 13 is specifically configured to predict the carbon emissions of the production carbon emission factors and the production direct energy consumption features through a pre-constructed production direct carbon emission prediction model to generate a production direct carbon emission prediction result; predict the carbon emissions of the production carbon emission factors and the production indirect energy consumption features through a pre-constructed production indirect carbon emission prediction model to generate a production indirect carbon emission prediction result; and generate a carbon emission prediction result corresponding to the production stage according to the production direct carbon emission prediction result and the production indirect carbon emission prediction result.

[0290] In the embodiment of the present invention, the device further includes: a historical production data acquisition unit 18, a production data association mining unit 19, a production direct carbon emission prediction model training unit 20, and a production indirect carbon emission prediction model training unit 21.

[0291] The historical production data acquisition unit 18 is used to acquire historical production direct energy consumption data and historical production indirect energy consumption data.

[0292] The production data association mining unit 19 is used to perform data mining on the historical production direct energy consumption data and the historical production indirect energy consumption data respectively through the association rule mining algorithm to generate a historical production direct multi-source data set and a historical production indirect multi-source data set.

[0293] The production direct carbon emission prediction model training unit 20 is used to train a random forest model and a convolutional neural network through the historical production direct multi-source data set to construct a production direct carbon emission prediction model.

[0294] The production indirect carbon emission prediction model training unit 21 is used to train a random forest model and a convolutional neural network through the historical production indirect multi-source data set to construct a production indirect carbon emission prediction model.

[0295] In an embodiment of the present invention, the life cycle stage is the transportation stage, and the carbon emission activity characteristics include the energy consumption characteristics of transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools; the carbon emission prediction unit 13 is specifically configured to predict the carbon emissions of the energy consumption characteristics of transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools through a pre-constructed transportation carbon emission prediction model, and generate a carbon emission prediction result corresponding to the transportation stage.

[0296] In an embodiment of the present invention, the device further includes: a historical transportation data acquisition unit 22, a historical transportation data preprocessing unit 23, an initial transportation prediction model training unit 24, and a transportation carbon emission prediction model optimization unit 25.

[0297] The historical transportation data acquisition unit 22 is used to acquire a historical transportation activity data set.

[0298] The historical transportation data preprocessing unit 23 is used to perform data preprocessing on the historical transportation activity data set to generate preprocessed historical transportation activity characteristics.

[0299] The initial transportation prediction model training unit 24 is used to train a random forest model through the historical transportation activity characteristics to construct an initial transportation prediction model.

[0300] The transportation carbon emission prediction model optimization unit 25 is used to optimize the initial transportation prediction model through a Bayesian optimization algorithm to construct a transportation carbon emission prediction model.

[0301] In an embodiment of the present invention, the life cycle stage is the usage stage, and the carbon emission activity characteristics include the energy consumption characteristics of rigid-flex boards and the usage carbon emission factors; the carbon emission prediction unit 13 is specifically configured to predict the carbon emissions of the energy consumption characteristics of rigid-flex boards and the usage carbon emission factors through a pre-constructed usage carbon emission prediction model, and generate a carbon emission prediction result corresponding to the usage stage.

[0302] In an embodiment of the present invention, the device further includes: a historical usage data acquisition unit 26, a historical usage data preprocessing unit 27, an initial usage prediction model training unit 28, and a usage carbon emission prediction model optimization unit 29.

[0303] The historical usage data acquisition unit 26 is used to acquire a historical usage activity data set.

[0304] The historical usage data preprocessing unit 27 is used to perform data preprocessing on the historical usage activity data set to generate preprocessed historical usage activity characteristics.

[0305] The initial usage prediction model training unit 28 is used to train a support vector machine through the historical usage activity characteristics to construct an initial usage prediction model.

[0306] The usage carbon emission prediction model optimization unit 29 is used to optimize the initial usage prediction model through the Bayesian optimization algorithm to construct a usage carbon emission prediction model.

[0307] In the embodiment of the present invention, the life cycle stage is the waste treatment stage, and the carbon emission activity characteristics include the direct energy consumption characteristics of waste treatment, the indirect energy consumption characteristics of waste treatment, the energy consumption characteristics of waste transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools; the carbon emission prediction unit 13 is specifically used to predict the carbon emissions of the energy consumption characteristics of waste transportation vehicles, the transportation distance characteristics, and the carbon emission factors of transportation tools through a pre-constructed waste transportation carbon emission prediction model to generate a waste transportation carbon emission prediction result; predict the carbon emissions of the direct energy consumption characteristics of waste treatment through a pre-constructed direct carbon emission prediction model for waste disassembly to generate a direct carbon emission prediction result for waste disassembly; predict the carbon emissions of the indirect energy consumption characteristics of waste treatment through a pre-constructed indirect carbon emission prediction model for waste disassembly to generate an indirect carbon emission prediction result for waste disassembly; generate a carbon emission prediction result corresponding to the waste treatment stage according to the waste transportation carbon emission prediction result, the direct carbon emission prediction result for waste disassembly, and the indirect carbon emission prediction result for waste disassembly.

[0308] In the embodiment of the present invention, the device further includes: a historical waste data acquisition unit 30, a historical waste data preprocessing unit 31, an initial waste transportation prediction model training unit 32, a waste transportation carbon emission prediction model optimization unit 33, a waste data correlation mining unit 34, a direct carbon emission prediction model training unit 35 for waste disassembly, and an indirect carbon emission prediction model training unit 36 for waste disassembly.

[0309] The historical waste data acquisition unit 30 is used to acquire a historical waste transportation activity data set, historical direct energy consumption data for waste treatment, and historical energy consumption data for production waste treatment.

[0310] The historical waste data preprocessing unit 31 is used to preprocess the historical waste transportation activity data set to generate preprocessed historical waste transportation activity characteristics.

[0311] The initial waste transportation prediction model training unit 32 is used to train a random forest model through historical waste transportation activity characteristics to construct an initial waste transportation prediction model.

[0312] The waste transportation carbon emission prediction model optimization unit 33 is used to optimize the initial waste transportation prediction model through the Bayesian optimization algorithm to construct a waste transportation carbon emission prediction model.

[0313] The discarded data association mining unit 34 is used to perform data mining on the historical discarded treatment direct energy consumption data and the historical production discarded treatment energy consumption data respectively through the association rule mining algorithm, and generate the historical discarded treatment direct multi-source data set and the historical discarded treatment indirect multi-source data set.

[0314] The waste dismantling direct carbon emission prediction model training unit 35 is used to train the random forest model and the convolutional neural network through the historical discarded treatment direct multi-source data set, and construct the waste dismantling direct carbon emission prediction model.

[0315] The waste dismantling indirect carbon emission prediction model training unit 36 is used to train the random forest model and the convolutional neural network through the historical discarded treatment indirect multi-source data set, and construct the waste dismantling indirect carbon emission prediction model.

[0316] In the embodiment of the present invention, the carbon footprint accounting unit 14 is specifically used to multiply the global warming potential value by the carbon emission prediction result corresponding to the life cycle stage to generate the carbon footprint accounting result corresponding to the life cycle stage.

[0317] In the embodiment of the present invention, the full life cycle carbon footprint reporting unit 15 is specifically used to summarize the carbon footprint accounting results corresponding to each life cycle stage to generate the total carbon emissions of the rigid-flex printed circuit board in the full life cycle; according to the total carbon emissions of the rigid-flex printed circuit board in the full life cycle and the carbon footprint accounting results corresponding to each life cycle stage, generate the carbon footprint accounting report of the full life cycle, and visualize the carbon footprint accounting report of the full life cycle.

[0318] In the solution of the embodiment of the present invention, receive the carbon emission activity data of the rigid-flex printed circuit board sent by the Internet of Things system; extract features according to the carbon emission activity data to generate carbon emission activity features; according to different life cycle stages, combine the carbon emission factor method, and through the pre-trained carbon emission prediction model, predict according to the carbon emission activity features to generate the carbon emission prediction result corresponding to the life cycle stage; generate the carbon footprint accounting result corresponding to the life cycle stage according to the carbon emission prediction result corresponding to the life cycle stage and the global warming potential value; generate the carbon footprint accounting report of the full life cycle according to the carbon footprint accounting results corresponding to each life cycle stage, which can realize the accurate accounting of the carbon footprint of the rigid-flex printed circuit board in the full life cycle, scientifically manage the carbon footprint, improve the accuracy of carbon emission data, improve the efficiency of carbon footprint accounting, provide a real-time monitoring and early warning function, and optimize the transportation route to reduce carbon emissions during transportation.

[0319] The systems, devices, modules or units illustrated in the above embodiments may be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer device. Specifically, the computer device may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0320] An embodiment of the present invention provides a computer device, including a memory and a processor. The memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions. When the program instructions are loaded and executed by the processor, the steps of the above embodiment of the full-life-cycle carbon footprint accounting method for the software-hardware combined board of the Internet of Things system are implemented. For specific descriptions, refer to the above embodiment of the full-life-cycle carbon footprint accounting method for the software-hardware combined board of the Internet of Things system.

[0321] Next, refer to Figure 8 , which shows a schematic structural diagram of a computer device 600 suitable for implementing the embodiments of the present application.

[0322] As Figure 8 shown, the computer device 600 includes a central processing unit (CPU) 601, which can perform various appropriate operations and processes according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage section 608 into the random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the computer device 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0323] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that the computer program read from it can be installed in the storage section 608 as needed.

[0324] In particular, according to an embodiment of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present invention includes a computer program product that includes a computer program tangibly embodied on a machine-readable medium, the computer program including program code for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 609, and / or installed from the removable medium 611.

[0325] Computer-readable media includes both permanent and non-permanent, removable and non-removable media implemented by any method or technology for storing information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0326] For convenience of description, the above devices are described by function as various units respectively. Of course, when implementing the present application, the functions of each unit can be implemented in one or more software and / or hardware.

[0327] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce means for implementing the functions specified in one Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0328] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the processes Figure 1 one or more processes and / or blocks Figure 1 specified in a block or blocks.

[0329] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the processes Figure 1 one or more processes and / or blocks Figure 1 specified in a block or blocks.

[0330] It should also be noted that the terms "including", "comprising", or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a series of elements includes not only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the statement "including a..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0331] In the technical solution of this application, the acquisition, storage, use, processing, etc. of data all comply with the relevant provisions of national laws and regulations.

[0332] It should be noted that in the embodiments of this application, some industry-existing solutions such as certain software, components, models, etc. may be mentioned. They should be considered exemplary, and their purpose is only to illustrate the feasibility in the implementation of the technical solution of this application, but it does not mean that the applicant has already or necessarily used this solution.

[0333] Those skilled in the art should understand that the embodiments of this application can be provided as a method, a system, or a computer program product. Therefore, this application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0334] This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The application can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0335] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the corresponding parts of the method embodiments for relevant details.

[0336] The above description is only for the embodiments of this application and is not intended to limit this application. For those skilled in the art, various changes and modifications can be made to this application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of this application shall be included within the scope of the claims of this application.

Claims

1. A method for calculating the carbon footprint of a hard-and-soft board in an Internet of Things system throughout its life cycle, characterized in that: The method comprises: Receive carbon emission activity data of the hard-and-soft board sent by the IoT system; Extract features based on the carbon emission activity data to generate carbon emission activity features; According to different life cycle stages, combined with the carbon emission factor method, a pre-trained carbon emission prediction model is used to predict according to the characteristics of the carbon emission activities to generate carbon emission prediction results corresponding to the life cycle stages; Generate carbon footprint calculation results corresponding to the life cycle stage based on the carbon emission prediction results and global warming potential values ​​corresponding to the life cycle stage; Based on the carbon footprint accounting results corresponding to each life cycle stage, a carbon footprint accounting report for the entire life cycle is generated.

2. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1 is characterized in that: Before extracting features according to the carbon emission activity data to generate carbon emission activity features, the method further includes: Clean the carbon emission activity data to generate cleaned carbon emission activity data; The cleaned carbon emission activity data is standardized to generate standardized carbon emission activity data.

3. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1 is characterized in that: The extracting features according to the carbon emission activity data to generate carbon emission activity features includes: The feature extraction of carbon emission activity data is performed through the mean clustering algorithm to generate initial activity features; The initial activity characteristics are reduced in dimension through principal component analysis to generate carbon emission activity characteristics.

4. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1 is characterized in that: The life cycle stage is the production stage, and the carbon emission activity characteristics include production carbon emission factors, production direct energy consumption characteristics, and production indirect energy consumption characteristics; According to different life cycle stages, combined with the carbon emission factor method, a pre-trained carbon emission prediction model is used to predict according to the carbon emission activity characteristics to generate carbon emission prediction results corresponding to the life cycle stage, including: By using a pre-built production direct carbon emission prediction model, carbon emission prediction is performed on the production carbon emission factor and the production direct energy consumption characteristics to generate a production direct carbon emission prediction result; By using a pre-built production indirect carbon emission prediction model, carbon emission prediction is performed on the production carbon emission factor and the production indirect energy consumption characteristics to generate a production indirect carbon emission prediction result; The carbon emission prediction results corresponding to the production stage are generated according to the production direct carbon emission prediction results and the production indirect carbon emission prediction results.

5. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 4 is characterized in that: The method further comprises: Obtain historical production direct energy consumption data and historical production indirect energy consumption data; By using an association rule mining algorithm, data mining is performed on the historical production direct energy consumption data and the historical production indirect energy consumption data to generate a historical production direct multi-source data set and a historical production indirect multi-source data set; The random forest model and convolutional neural network are trained by using the historical direct production multi-source data set to construct the direct production carbon emissions prediction model; The random forest model and convolutional neural network are trained through the historical production indirect multi-source data set to construct the production indirect carbon emissions prediction model.

6. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1, characterized in that: The life cycle stage is the transportation stage, and the carbon emission activity characteristics include the energy consumption characteristics of the transportation vehicle, the transportation distance characteristics and the carbon emission factor of the transportation tool; According to different life cycle stages, combined with the carbon emission factor method, a pre-trained carbon emission prediction model is used to predict according to the carbon emission activity characteristics to generate carbon emission prediction results corresponding to the life cycle stage, including: Through the pre-constructed transportation carbon emission prediction model, the carbon emission prediction is carried out on the energy consumption characteristics of the transportation vehicle, the transportation distance characteristics and the carbon emission factor of the transportation tool to generate the carbon emission prediction result corresponding to the transportation stage.

7. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 6 is characterized in that: The method further comprises: Obtain historical transportation activity datasets; Performing data preprocessing on the historical transportation activity data set to generate preprocessed historical transportation activity features; The random forest model is trained based on the historical transportation activity characteristics to construct an initial transportation prediction model; The initial transportation prediction model is optimized by using a Bayesian optimization algorithm to construct the transportation carbon emission prediction model.

8. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1 is characterized in that: The life cycle stage is the use stage, and the carbon emission activity characteristics include the energy consumption characteristics of the rigid-flex board and the carbon emission factor of use; According to different life cycle stages, combined with the carbon emission factor method, a pre-trained carbon emission prediction model is used to predict according to the carbon emission activity characteristics to generate carbon emission prediction results corresponding to the life cycle stage, including: The carbon emission prediction model constructed in advance is used to predict the carbon emission of the rigid-flexible board based on its energy consumption characteristics and the carbon emission factor, thereby generating a carbon emission prediction result corresponding to the use stage.

9. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 8 is characterized in that: The method further comprises: Obtain historical usage activity datasets; Performing data preprocessing on the historical usage activity data set to generate preprocessed historical usage activity features; Using the historical usage activity features, model training is performed on a support vector machine to construct an initial usage prediction model; The initial usage prediction model is optimized by using a Bayesian optimization algorithm to construct the usage carbon emission prediction model.

10. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1, characterized in that: The life cycle stage is the waste treatment stage, and the carbon emission activity characteristics include direct energy consumption characteristics of waste treatment, indirect energy consumption characteristics of waste treatment, energy consumption characteristics of waste transportation vehicles, transportation distance characteristics and carbon emission factors of transportation tools; According to different life cycle stages, combined with the carbon emission factor method, a pre-trained carbon emission prediction model is used to predict according to the carbon emission activity characteristics to generate carbon emission prediction results corresponding to the life cycle stage, including: The carbon emission prediction model for waste transportation is used to predict the carbon emission of the waste transportation vehicle based on its energy consumption characteristics, transportation distance characteristics and carbon emission factor, thereby generating a prediction result for the carbon emission of waste transportation; By using a pre-built direct carbon emission prediction model for waste disassembly, carbon emission prediction is performed on the direct energy consumption characteristics of the waste treatment to generate a direct carbon emission prediction result for waste disassembly; By using a pre-built waste disassembly indirect carbon emission prediction model, carbon emission prediction is performed on the waste treatment indirect energy consumption characteristics to generate a waste disassembly indirect carbon emission prediction result; The carbon emission prediction results corresponding to the waste treatment stage are generated according to the waste transportation carbon emission prediction results, the waste dismantling direct carbon emission prediction results and the waste dismantling indirect carbon emission prediction results.

11. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 10, characterized in that: The method further comprises: Obtain historical waste transportation activity datasets, historical waste treatment direct energy consumption data, and historical production waste treatment energy consumption data; Performing data preprocessing on the historical waste transportation activity data set to generate preprocessed historical waste transportation activity features; The random forest model is trained based on the historical waste transportation activity characteristics to construct an initial waste transportation prediction model; The initial waste transportation prediction model is optimized by using a Bayesian optimization algorithm to construct the waste transportation carbon emission prediction model; By using an association rule mining algorithm, data mining is performed on the historical waste treatment direct energy consumption data and the historical production waste treatment energy consumption data to generate a historical waste treatment direct multi-source data set and a historical waste treatment indirect multi-source data set; The random forest model and convolutional neural network are trained by using the historical waste treatment direct multi-source data set to construct a prediction model for direct carbon emissions from waste dismantling; The random forest model and convolutional neural network are trained through the historical waste treatment indirect multi-source data set to construct the waste disassembly indirect carbon emission prediction model.

12. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1, characterized in that: The carbon footprint calculation results corresponding to the life cycle stage are generated according to the carbon emission prediction results and global warming potential values ​​corresponding to the life cycle stage, including: The global warming potential value is multiplied by the carbon emission prediction result corresponding to the life cycle stage to generate the carbon footprint calculation result corresponding to the life cycle stage.

13. The method for calculating the carbon footprint of a rigid-soft board in an Internet of Things system throughout its life cycle according to claim 1, characterized in that: The carbon footprint accounting report for the entire life cycle is generated based on the carbon footprint accounting results corresponding to each life cycle stage, including: Summarize the carbon footprint calculation results corresponding to each life cycle stage to generate the total carbon emissions of the rigid-flex board throughout its life cycle; Based on the total carbon emissions of the rigid-flex PCB throughout its life cycle and the carbon footprint accounting results corresponding to each life cycle stage, a carbon footprint accounting report for the entire life cycle is generated and visualized.

14. A carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system, characterized in that: The device comprises: A receiving unit, used for receiving carbon emission activity data of the rigid-flex board sent by the Internet of Things system; A feature extraction unit, used to extract features based on the carbon emission activity data to generate carbon emission activity features; A carbon emission prediction unit is used to generate carbon emission prediction results corresponding to the life cycle stage by combining the carbon emission factor method with a pre-trained carbon emission prediction model according to the carbon emission activity characteristics according to different life cycle stages; A carbon footprint calculation unit is used to generate carbon footprint calculation results corresponding to the life cycle stage according to the carbon emission prediction results and global warming potential values ​​corresponding to the life cycle stage; The full life cycle carbon footprint reporting unit is used to generate a full life cycle carbon footprint accounting report based on the carbon footprint accounting results corresponding to each life cycle stage.

15. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 14, characterized in that: The device also includes: A data cleaning unit, used for cleaning the carbon emission activity data to generate cleaned carbon emission activity data; The data standardization unit is used to perform standardization processing on the cleaned carbon emission activity data to generate standardized carbon emission activity data.

16. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 14, characterized in that: The feature extraction unit is specifically used to extract features from carbon emission activity data through a mean clustering algorithm to generate initial activity features; and to perform dimensionality reduction processing on the initial activity features through a principal component analysis method to generate carbon emission activity features.

17. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 14, characterized in that: The life cycle stage is the production stage, and the carbon emission activity characteristics include production carbon emission factors, production direct energy consumption characteristics, and production indirect energy consumption characteristics; The carbon emissions prediction unit is specifically used to predict the carbon emissions of the production carbon emission factors and the direct production energy consumption characteristics through a pre-constructed direct production carbon emissions prediction model, and generate a direct production carbon emissions prediction result; to predict the carbon emissions of the production carbon emission factors and the indirect production energy consumption characteristics through a pre-constructed indirect production carbon emissions prediction model, and generate an indirect production carbon emissions prediction result; and to generate a carbon emissions prediction result corresponding to the production stage based on the direct production carbon emissions prediction result and the indirect production carbon emissions prediction result.

18. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 17, characterized in that: The device also includes: A historical production data acquisition unit, used to acquire historical production direct energy consumption data and historical production indirect energy consumption data; A production data association mining unit, used to perform data mining on the historical production direct energy consumption data and the historical production indirect energy consumption data respectively through an association rule mining algorithm to generate a historical production direct multi-source data set and a historical production indirect multi-source data set; A production direct carbon emissions prediction model training unit is used to train a random forest model and a convolutional neural network through the historical production direct multi-source data set to construct the production direct carbon emissions prediction model; The production indirect carbon emissions prediction model training unit is used to train the random forest model and the convolutional neural network through the historical production indirect multi-source data set to construct the production indirect carbon emissions prediction model.

19. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 14, characterized in that: The life cycle stage is the transportation stage, and the carbon emission activity characteristics include the energy consumption characteristics of the transportation vehicle, the transportation distance characteristics and the carbon emission factor of the transportation tool; The carbon emission prediction unit is specifically used to predict the carbon emissions of the transport vehicle energy consumption characteristics, transport distance characteristics and transport tool carbon emission factors through a pre-constructed transport carbon emission prediction model, and generate a carbon emission prediction result corresponding to the transport stage.

20. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 19, characterized in that: The device also includes: A historical transportation data acquisition unit, used to acquire historical transportation activity data sets; A historical transportation data preprocessing unit, used to perform data preprocessing on the historical transportation activity data set to generate preprocessed historical transportation activity features; An initial transportation prediction model training unit is used to train a random forest model through the historical transportation activity characteristics to construct an initial transportation prediction model; The transport carbon emission prediction model optimization unit is used to optimize the initial transport prediction model through a Bayesian optimization algorithm to construct the transport carbon emission prediction model.

21. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 14, characterized in that: The life cycle stage is the use stage, and the carbon emission activity characteristics include the energy consumption characteristics of the rigid-flex board and the carbon emission factor of use; The carbon emission prediction unit is specifically used to predict the carbon emissions of the rigid-flexible board based on its energy consumption characteristics and carbon emission factors through a pre-built carbon emission prediction model, and generate a carbon emission prediction result corresponding to the use stage.

22. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 21, characterized in that: The device also includes: A historical usage data acquisition unit, used to acquire a historical usage activity data set; A historical usage data preprocessing unit, used to perform data preprocessing on the historical usage activity data set to generate preprocessed historical usage activity features; An initial usage prediction model training unit, used to perform model training on a support vector machine through the historical usage activity features to construct an initial usage prediction model; The carbon emission prediction model optimization unit is used to optimize the initial usage prediction model through a Bayesian optimization algorithm to construct the carbon emission prediction model.

23. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 14, characterized in that: The life cycle stage is the waste treatment stage, and the carbon emission activity characteristics include direct energy consumption characteristics of waste treatment, indirect energy consumption characteristics of waste treatment, energy consumption characteristics of waste transportation vehicles, transportation distance characteristics and carbon emission factors of transportation tools; The carbon emission prediction unit is specifically used to predict the carbon emissions of the waste transportation vehicle energy consumption characteristics, transportation distance characteristics and transportation tool carbon emission factors through a pre-constructed waste transportation carbon emission prediction model, and generate a waste transportation carbon emission prediction result; to predict the direct energy consumption characteristics of the waste treatment through a pre-constructed waste dismantling direct carbon emission prediction model, and generate a waste dismantling direct carbon emission prediction result; to predict the indirect energy consumption characteristics of the waste treatment through a pre-constructed waste dismantling indirect carbon emission prediction model, and generate a waste dismantling indirect carbon emission prediction result; and to generate a carbon emission prediction result corresponding to the waste treatment stage based on the waste transportation carbon emission prediction result, the waste dismantling direct carbon emission prediction result and the waste dismantling indirect carbon emission prediction result.

24. The carbon footprint accounting device for the entire life cycle of a rigid-soft board in an Internet of Things system according to claim 23, characterized in that: The device also includes: A historical waste data acquisition unit, used to acquire historical waste transportation activity data sets, historical waste treatment direct energy consumption data, and historical production waste treatment energy consumption data; A historical waste data preprocessing unit, used to perform data preprocessing on the historical waste transportation activity data set to generate preprocessed historical waste transportation activity features; An initial waste transportation prediction model training unit is used to train a random forest model based on the historical waste transportation activity characteristics to construct an initial waste transportation prediction model; A waste transportation carbon emission prediction model optimization unit, used to optimize the initial waste transportation prediction model through a Bayesian optimization algorithm to construct the waste transportation carbon emission prediction model; A waste data association mining unit is used to perform data mining on the historical waste treatment direct energy consumption data and the historical production waste treatment energy consumption data respectively through an association rule mining algorithm to generate a historical waste treatment direct multi-source data set and a historical waste treatment indirect multi-source data set; A waste dismantling direct carbon emission prediction model training unit, used to train the random forest model and convolutional neural network through the historical waste treatment direct multi-source data set to construct the waste dismantling direct carbon emission prediction model; The waste dismantling indirect carbon emission prediction model training unit is used to train the random forest model and the convolutional neural network through the historical waste treatment indirect multi-source data set to construct the waste dismantling indirect carbon emission prediction model.

25. The device for calculating the carbon footprint of a rigid-soft board in the entire life cycle of an Internet of Things system according to claim 14, characterized in that: The carbon footprint calculation unit is specifically used to multiply the global warming potential value with the carbon emission prediction result corresponding to the life cycle stage to generate the carbon footprint calculation result corresponding to the life cycle stage.

26. The carbon footprint accounting device for the entire life cycle of a hard-and-soft board in an Internet of Things system according to claim 14, characterized in that: The full life cycle carbon footprint reporting unit is specifically used to summarize the carbon footprint accounting results corresponding to each life cycle stage to generate the total carbon emissions of the full life cycle of the rigid-flex board; based on the total carbon emissions of the full life cycle of the rigid-flex board and the carbon footprint accounting results corresponding to each life cycle stage, a full life cycle carbon footprint accounting report is generated, and the full life cycle carbon footprint accounting report is visualized.

27. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for calculating the carbon footprint of a hard-and-soft board in a IoT system throughout its life cycle as described in any one of claims 1 to 13 is implemented.

28. A computer device comprising a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, characterized in that: When the program instructions are loaded and executed by the processor, the method for calculating the carbon footprint of the entire life cycle of the soft-hard combination board of the Internet of Things system as described in any one of claims 1 to 13 is implemented.

29. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by the processor, the method for calculating the carbon footprint of a hard-and-soft board in a IoT system throughout its life cycle as described in any one of claims 1 to 13 is implemented.

Images