Early warning method based on industry electricity and carbon data matching

Abstract

The invention discloses an early warning method based on industry electricity and carbon data matching, and belongs to the technical field of electricity and carbon data early warning. The method is realized through five steps of data acquisition, data preprocessing, data matching analysis, early warning threshold setting, and real-time monitoring and early warning. According to the invention, through a layered and partitioned sensor deployment strategy, the power consumption and carbon emission conditions of different regions and equipment can be captured more accurately, and the comprehensiveness and accuracy of data are ensured. And the data acquisition frequency and multi-source data fusion are dynamically adjusted, so that the sensitivity to the production process and the environmental change is improved, and the real-time performance and the correlation of the data are ensured. And an advanced time sequence analysis model and an online learning mechanism are adopted, so that the model can adapt to the change of data distribution in time, and the prediction accuracy is improved. Dynamic adjustment and risk assessment of the early warning threshold ensure the sensitivity and stability of the early warning system and reduce the possibility of false alarm and missing alarm.

Description

Technical Field

[0001] The present invention relates to the technical field of electric carbon data early warning, and more specifically, to an early warning method based on industry electric carbon data matching. Background Art

[0002] With the intensification of global climate change and the proposal of carbon neutrality goals by various countries, carbon emission monitoring and management has become an important issue in the industrial field. Electricity consumption and carbon emissions are two highly correlated key indicators in industrial production. By monitoring and analyzing the matching relationship between the two, energy consumption anomalies and excessive emissions can be effectively identified, providing data support for enterprises to achieve energy conservation and emission reduction.

[0003] Incomplete data collection: Traditional energy and carbon emissions monitoring methods often rely on single-point monitoring, which cannot fully cover the data of different equipment and areas in the production process, resulting in information blind spots and insufficient monitoring.

[0004] Data processing lag: Some systems require a long processing cycle after data collection, making it difficult to detect abnormal situations in a timely manner, delaying the response time.

[0005] Single early warning mechanism: The early warning mechanisms of many existing systems are based on static rules and lack dynamic adjustment capabilities, making it difficult to adapt to the ever-changing production environment and data distribution.

[0006] Lack of visualization effect: The visualization function of traditional systems is relatively limited, and it is difficult to intuitively display the complex relationship between electricity and carbon emissions and abnormal warning information.

[0007] The application of industrial Internet of Things (IoT) and edge computing technology provides a foundation for real-time collection and processing of industrial data. The application of time series analysis and machine learning models in anomaly detection and trend prediction is becoming more and more widespread, significantly improving the accuracy and efficiency of data analysis. The advancement of visualization technology enables enterprise managers to understand data distribution and anomalies more intuitively, so as to make decisions quickly.

[0008] Based on the above background, the industry urgently needs an electric-carbon data matching analysis and early warning system that integrates real-time, accuracy and flexibility to comprehensively enhance the ability of energy consumption monitoring and carbon emission management in the industrial field. Summary of the invention

[0009] 1. Technical problems to be solved

[0010] In view of the problems existing in the prior art, the purpose of the present invention is to provide an early warning method based on industry electric carbon data matching, which can realize real-time monitoring of multi-dimensional indicators and help enterprises quickly respond and solve problems by deeply exploring the potential causes of abnormal events. The automated early warning response process and multi-channel notification function improve the efficiency and timeliness of enterprises in handling abnormal events.

[0011] 2. Technical solution

[0012] To solve the above problems, the present invention adopts the following technical solutions.

[0013] The early warning method based on industry electricity carbon data matching includes the following steps:

[0014] 1) Data collection: When collecting data, set the unique identifier of the sensor and perform regular verification. During data transmission, use checksum and hash algorithms to verify the integrity of the data to ensure that the data has not been tampered with.

[0015] 2) Data preprocessing: Clean, remove and standardize the collected power consumption data and carbon emission data, use box plot analysis to identify outliers, and combine visualization methods to help analyze data distribution;

[0016] 3) Data matching analysis: Construct a matching model between electricity consumption and carbon emissions and calculate the electricity-to-carbon ratio Where P(t) is the electricity consumption at time t, C(t) is the carbon emission at time t, and grid search and cross validation are used to optimize the model parameters;

[0017] 4) Set warning thresholds: Set warning thresholds for the electricity-to-carbon ratio based on historical data and industry standards, set warning levels, and define trigger conditions for each level; set a yellow warning when the electricity-to-carbon ratio deviates from the benchmark value by 5-10%, set an orange warning when the electricity-to-carbon ratio deviates from the benchmark value by 10-20%, and set a red warning when the electricity-to-carbon ratio deviates from the benchmark value by more than 10%. After the warning is triggered, the system automatically generates a report and sends it to relevant personnel, and records the event log for subsequent analysis;

[0018] 5) Real-time monitoring and early warning: The electricity-to-carbon ratio is monitored in real time. When the electricity-to-carbon ratio exceeds the early warning threshold, an early warning is triggered and an abnormal report is generated.

[0019] Furthermore, the data collection step includes utilizing industrial Internet of Things (IoT) devices for data transmission and using edge computing technology to perform preliminary processing on the data.

[0020] Furthermore, the data preprocessing step includes outlier detection, interpolation and standardization transformation to ensure data quality.

[0021] Furthermore, the data matching and analysis step uses a time series analysis model to predict and analyze the trend of the electricity-to-carbon ratio.

[0022] Preferably, the warning information is displayed through a visualization platform, supporting real-time dynamic dashboards and multi-platform notification push, using WebSocket for real-time data transmission, ensuring that warning information and data updates can be pushed to the user interface in real time.

[0023] An early warning system based on industry electricity carbon data matching includes the following steps:

[0024] 1) Data collection module: used to collect power consumption data and carbon emission data in real time;

[0025] 2) Data preprocessing module: used to clean and standardize data;

[0026] 3) Matching analysis module: used to calculate the electricity-to-carbon ratio and perform matching analysis;

[0027] 4) Early warning module: used to monitor the electricity-to-carbon ratio in real time and trigger abnormal early warning;

[0028] 5) Visualization module: used to display matching analysis results and warning information.

[0029] Furthermore, the warning module includes a rule engine that triggers warnings of different levels according to different threshold levels set by the user.

[0030] 3. Beneficial effects

[0031] Compared with the prior art, the advantages of the present invention are:

[0032] 1) Through the layered and partitioned sensor deployment strategy, the power consumption and carbon emissions of different areas and equipment can be captured more accurately to ensure the comprehensiveness and accuracy of the data. Dynamically adjust the data collection frequency and multi-source data fusion to enhance the sensitivity to production process and environmental changes, and ensure the real-time and relevance of the data.

[0033] 2) Using advanced time series analysis models and online learning mechanisms, the model can adapt to changes in data distribution in a timely manner and improve the accuracy of predictions. Dynamic adjustment of warning thresholds and risk assessment ensure the sensitivity and stability of the warning system and reduce the possibility of false alarms and missed reports.

[0034] 3) Real-time monitoring of multi-dimensional indicators, and in-depth mining of potential causes of abnormal events to help enterprises quickly respond to and solve problems. Automated early warning response processes and multi-channel notification functions improve the efficiency and timeliness of enterprises in handling abnormal events.

[0035] 4) The compatibility and security of the data acquisition module ensure the scalability and data security of the system. The interactivity and mobile adaptation of the visualization module enable users to conveniently and intuitively analyze and understand the electric carbon data and early warning information. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present invention are clearly and completely described below; it is obvious that the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments, and all other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making creative work are within the scope of protection of the present invention.

[0037] Embodiment 1:

[0038] Early warning method and system based on industry electric carbon data matching:

[0039] 1. Data Collection

[0040] When collecting data, in addition to setting the sensor’s unique identifier and verifying it regularly, as well as using industrial Internet of Things (IoT) devices and edge computing technology, you can further explain:

[0041] 1. Sensor deployment strategy: According to the distribution of different power consumption equipment and carbon emission sources, adopt a layered and partitioned sensor layout. For example, in large factories, for areas where high-energy-consuming equipment is concentrated, increase the density of sensors to obtain more accurate local data changes, and install high-precision sensors at concentrated carbon emission points (such as chimneys, exhaust gas treatment facility outlets, etc.) to ensure the reliability of carbon emission data.

[0042] 2. Dynamic adjustment of data collection frequency: Dynamically adjust the data collection frequency according to the different stages of industry production and equipment operation status. During peak production or unstable equipment operation, increase the frequency of data collection to capture fluctuations in power consumption and carbon emissions in a timely manner; and during the off-season or stable equipment operation stage, appropriately reduce the collection frequency to reduce data transmission and storage pressure. For example, for the steel production industry, data is collected every 5 minutes during the blast furnace ironmaking process, and can be collected every 30 minutes during routine equipment maintenance.

[0043] 3. Multi-source integration of data collection: In addition to the data collected by IoT devices, multi-source data such as the enterprise's production management system data and energy management system data can also be integrated. For example, information such as production plans and equipment operating hours can be obtained from the production management system, and correlated with power consumption and carbon emission data for a deeper understanding of the impact of production activities on power-carbon data. For example, through the product output and production process information in the production plan, the theoretical power consumption and carbon emission range can be calculated, and compared with the actual collected data for analysis, further improving data accuracy and anomaly detection capabilities.

[0044] 2. Data Preprocessing

[0045] In terms of data cleaning, denoising and standardization, we can elaborate on the following:

[0046] 1. Combination of multiple methods for outlier processing: In addition to using box plot analysis to identify outliers, you can also combine anomaly detection methods based on statistical models, such as clustering-based anomaly detection. Cluster the power consumption and carbon emission data according to different equipment types, production workshops or time periods, and those data points far away from the cluster center are considered outliers. Comprehensive analysis and verification are performed on outliers identified by different detection methods to avoid misjudgment. For example, for power consumption data points that are judged as abnormal by box plots within a certain time period, if they are also isolated in cluster analysis, there is more reason to confirm them as outliers; if they are within the normal cluster range in cluster analysis, it is necessary to further check whether there are problems in the data collection process.

[0047] 2. Intelligent algorithms for data interpolation: When performing data interpolation, intelligent algorithms based on machine learning are used, such as interpolation methods based on linear regression or decision trees. According to the historical trends and relevant characteristics of the data, a data prediction model is established to predict and interpolate the missing data. For example, for power consumption data, if the data at a certain moment is missing, the power consumption data, equipment operation status data, and production plan data of the previous period can be used as features to predict the missing value through a trained linear regression model. And after interpolation, the rationality of the interpolated data is verified by methods such as smoothness test with adjacent data points.

[0048] 3. Adaptive adjustment of standardized conversion: Adaptive standardized conversion methods are used to target the characteristics of electricity-carbon data in different industries, different enterprise sizes, and different equipment types. For example, for electricity consumption data, the standardized mean and standard deviation are dynamically determined based on the distribution range of the company's historical data and the industry average. For emerging clean energy companies, the distribution of their electricity consumption data may be very different from that of traditional high-energy-consuming companies. Through adaptive standardized conversion, data can be better comparable between different companies, which is convenient for subsequent matching analysis and early warning threshold setting.

[0049] 3. Data Matching Analysis

[0050] In terms of building a matching model between electricity consumption and carbon emissions and using a time series analysis model:

[0051] 1. Multi-factor considerations for matching models: When calculating the electricity-to-carbon ratio R(t) = (P(t)) / (C(t)), more factors that affect the electricity-to-carbon relationship are introduced as model variables. In addition to the electricity consumption P(t) and carbon emissions C(t) related to time t, factors such as equipment operating efficiency, energy structure (such as the ratio of fossil energy to clean energy), and production process parameters can also be considered. For example, for chemical production companies, process parameters such as temperature and pressure of chemical reactions will directly affect energy consumption and carbon emissions. Incorporating these parameters into the matching model can more accurately characterize the electricity-to-carbon relationship. Use multiple linear regression or more complex machine learning models (such as random forest regression) to construct a matching model that considers multiple factors. When optimizing model parameters through grid search and cross-validation, simultaneously evaluate the contribution of different factors to model performance, screen out key factors, and improve the interpretability and predictive accuracy of the model.

[0052] 2. Optimization of time series analysis models: When predicting and analyzing the trend of the electricity-to-carbon ratio, more advanced time series analysis models, such as the ARIMA-GARCH hybrid model, are used. The ARIMA model is used to capture the linear autocorrelation and trend characteristics of the electricity-to-carbon ratio data, while the GARCH model is used to describe the heteroskedasticity of the data, that is, the clustering of fluctuations. This hybrid model can better cope with the nonlinearity, non-stationarity and fluctuations that may exist in the electricity-to-carbon ratio data. During the model training process, the rolling forecast verification method is used to continuously update the model parameters to adapt to the dynamic changes of the data. For example, in the case of large fluctuations in electricity market prices leading to an unstable relationship between corporate electricity consumption and carbon emissions, the ARIMA-GARCH model can more effectively predict the future trend of the electricity-to-carbon ratio and detect potential abnormal changes in advance.

[0053] 3. Online learning and updating of models: With the improvement of enterprise production processes, equipment upgrades, and adjustments to energy structures, the electricity-carbon relationship will change. Therefore, the matching model and time series analysis model constructed should have the ability to learn and update online. When new data continues to flow in, the model can automatically detect changes in data distribution and trigger model retraining and parameter update mechanisms. For example, using incremental learning algorithms, such as incremental learning methods based on support vector machines, the model can be quickly updated to adapt to new electricity-carbon data patterns without recalculating all historical data, ensuring the long-term effectiveness of the early warning system.

[0054] 4. Warning Threshold Setting

[0055] When setting the electricity-to-carbon ratio warning threshold based on historical data and industry standards:

[0056] 1. Industry segmentation and differentiated threshold setting: Set more targeted warning thresholds for different industry segments. For example, in the manufacturing industry, the automobile manufacturing industry and the electronics manufacturing industry have different production processes and energy usage characteristics, so the reasonable range and warning threshold of their electricity-to-carbon ratio should be different. Automobile manufacturing involves a large number of metal processing and painting processes, and the energy consumption and carbon emissions are relatively high and fluctuate greatly; while electronic manufacturing is mainly precision processing and automated production lines, and the energy consumption and emissions are relatively low and stable. Through in-depth analysis and data mining of different industry segments, determine the appropriate warning threshold range for each industry, and improve the accuracy of warnings and industry adaptability.

[0057] 2. Dynamic threshold adjustment mechanism: In addition to setting the warning threshold based on a fixed deviation range, a dynamic threshold adjustment mechanism is established. Taking into account the impact of factors such as the company's production plan adjustment, seasonal changes, and market fluctuations on the electricity-to-carbon data, the warning threshold is adjusted dynamically. For example, during the high temperatures in summer, due to the extensive use of refrigeration equipment such as air conditioners, the overall electricity consumption of the company will increase, and the warning threshold of the electricity-to-carbon ratio can be appropriately relaxed accordingly; and during the period when the company carries out large-scale energy-saving and emission reduction technology transformation, as energy utilization efficiency improves, the warning threshold should be gradually tightened to encourage companies to continue to improve. By establishing a threshold adjustment model based on multi-factor influences, such as a threshold adjustment model based on a neural network, the appropriate warning threshold is dynamically output based on real-time production, environmental, market and other data inputs, making the warning system more flexible and intelligent.

[0058] 3. Risk assessment and verification of threshold setting: After setting the warning threshold, conduct risk assessment and verification. Use simulation analysis methods to substitute historical data into the warning system, observe the triggering of warnings under different threshold settings, and compare and verify with actual abnormal carbon emissions. At the same time, combined with the opinions of industry experts and corporate risk management requirements, evaluate the false alarm rate and missed alarm rate that may be caused by different warning thresholds, and optimize the thresholds. For example, if a certain threshold setting results in a false alarm rate that is too high, it may cause the company to distrust the warning system; and a missed alarm rate that is too high will cause the company to miss the opportunity to promptly discover and deal with abnormal carbon emissions. Through risk assessment and verification, find an optimal warning threshold setting that balances the false alarm rate and missed alarm rate.

[0059] V. Real-time monitoring and early warning

[0060] In terms of real-time monitoring of the electricity-to-carbon ratio and triggering early warnings to generate abnormal reports:

[0061] 1. Real-time monitoring of multi-dimensional indicators: In addition to the core indicator of electricity-to-carbon ratio, multi-dimensional indicators related to electricity-to-carbon data are also monitored, such as the power factor of electricity consumption and the component analysis of carbon emissions. The power factor can reflect the operating efficiency of the enterprise's power system. A low power factor may mean that there are problems such as excessive reactive power loss, which indirectly affects electricity consumption and carbon emissions; the component analysis of carbon emissions can help enterprises understand the source and composition of carbon emissions, which is of great significance for taking targeted emission reduction measures. When these multi-dimensional indicators show abnormal changes, they can also serve as auxiliary basis for early warning triggering, further enriching the connotation of early warning information and improving the comprehensiveness of early warning.

[0062] 2. In-depth mining and analysis of early warning information: After the early warning is triggered, the early warning information is deeply mined and analyzed. In addition to generating basic abnormality reports, including warning time, abnormal values ​​of the electricity-to-carbon ratio, related equipment information, etc., data mining technology can also be used to analyze the potential causes of abnormal events. For example, through correlation analysis, other production parameters or environmental factors that change at the same time as the abnormal electricity-to-carbon ratio, such as equipment fault codes, changes in raw material quality, meteorological conditions, etc., are found to provide support for enterprises to quickly locate the root cause of the problem. At the same time, abnormal events are classified and statistically analyzed to summarize the frequency, scope of impact and severity of different types of abnormal events, providing data basis for enterprises to formulate risk management strategies and improvement measures.

[0063] 3. Automated process of early warning response: Establish an automated process for early warning response. When an early warning is triggered, in addition to automatically generating a report and sending it to relevant personnel, a series of response measures can also be automatically initiated. For example, according to the early warning level, automatically adjust the equipment operating parameters, such as reducing the power output of high-energy-consuming equipment, optimizing production process parameters to reduce carbon emissions; or automatically dispatch backup energy equipment, such as switching to clean energy power generation equipment to improve the electricity-to-carbon ratio. At the same time, integrate the early warning information with the company's supply chain management system, financial system, etc. When an early warning occurs, automatically evaluate the impact on supply chain stability and corporate costs, prepare for response in advance, and realize a fully automated process from early warning to response, thereby improving the efficiency and timeliness of the company's response to abnormal electricity-to-carbon events.

[0064] 6. Early warning system module

[0065] 1. Data acquisition module:

[0066] 1) Compatibility of data acquisition modules: Enhance the compatibility of data acquisition modules with different types of sensors and data acquisition devices. With the development of IoT technology, new sensors and data acquisition devices are constantly emerging. Data acquisition modules should be able to easily access these new devices without large-scale system modifications. For example, by adopting common data interface standards (such as OPC-UA) and plug-in architecture, when new power sensors or carbon emission monitoring devices are connected, only the corresponding plug-ins need to be developed to achieve rapid expansion of data acquisition functions.

[0067] 2) Security protection of data acquisition module: Strengthen the security protection measures of data acquisition module to prevent data from being stolen or tampered with during the acquisition process. Use encryption transmission technology, such as SSL / TLS protocol, to encrypt the collected data before transmission; at the same time, perform identity authentication and access control on data acquisition equipment, and only authorized devices can access the data acquisition network to ensure the legitimacy and security of the data source. For example, establish an encrypted tunnel between the sensor and the data acquisition gateway to prevent data from being intercepted and cracked by hackers during wireless transmission.

[0068] 2. Data preprocessing module:

[0069] 1) Configurability of data preprocessing module: Improve the configurability of data preprocessing module, so that enterprises can flexibly adjust the parameters and rules of data cleaning, denoising and standardization according to their own data characteristics and business needs. For example, provide a visual configuration interface, enterprise users can set the threshold of outlier detection, data interpolation method and standardization conversion parameters according to different data sources and data quality conditions, without relying on professional technicians to make complex code modifications, thus improving the flexibility and adaptability of data preprocessing.

[0070] 2) Performance optimization of data preprocessing module: Optimize the performance of data preprocessing module to improve the speed and efficiency of data processing. Use distributed computing technology, such as Hadoop or Spark framework, to distribute large-scale electric carbon data and reduce data processing time. At the same time, optimize the data preprocessing algorithm, such as using data structures and algorithms based on memory computing, reduce the number of data reads and writes, and improve the efficiency of algorithm execution. For example, when performing data cleaning, use Spark's memory computing characteristics to load data into memory for processing, which greatly improves the speed of data cleaning and meets the requirements of real-time data processing.

[0071] 3. Matching analysis module:

[0072] 1) Model evaluation and monitoring of the matching analysis module: Add model evaluation and monitoring functions to the matching analysis module to track the performance and accuracy of the model in real time. Use model evaluation indicators such as mean square error (MSE) and mean absolute error (MAE) to regularly evaluate the constructed power consumption and carbon emission matching model and time series analysis model. When the model performance drops to a certain level, the model retraining and optimization mechanism is automatically triggered. At the same time, monitor the input data and output results of the model to promptly detect data anomalies or model anomalies, such as data drift, model overfitting, etc., and take corresponding measures to deal with them to ensure the stability and reliability of the matching analysis module.

[0073] 2) Enhanced model interpretability of the matching analysis module: Improve the interpretability of the model in the matching analysis module so that enterprises can better understand the decision-making process and results of the model. Use interpretable machine learning algorithms, such as decision trees, linear regression, etc., or use model interpretation tools (such as LIME, SHAP, etc.) to interpret complex machine learning models (such as random forests, neural networks, etc.). For example, by visualizing the main factors and decision rules considered by the model when calculating the electricity-to-carbon ratio through a decision tree, it helps enterprise technicians and managers understand the operating logic of the model, so as to better formulate electricity-to-carbon management strategies and decisions based on the model results.

[0074] 4. Early warning module:

[0075] 1) Dynamic adjustment of the warning level of the early warning module: Allow the early warning module to dynamically adjust the division of warning levels and trigger conditions according to the risk preferences and actual operating conditions of the enterprise. For example, when the enterprise faces strict environmental protection supervision or energy conservation and emission reduction pressure, the warning level will be divided more finely, the warning trigger threshold will be lowered, and the sensitivity of the warning will be improved; and when the production and operation of the enterprise is relatively stable and the fluctuation of electric carbon data is small, the division of warning levels and trigger conditions will be appropriately relaxed to reduce unnecessary warning interference. By providing the function of dynamic adjustment of warning levels, the early warning system can better adapt to the needs of enterprises at different stages.

[0076] 2) Multi-channel notification expansion of the early warning module: In addition to the existing notification methods, the multi-channel notification function of the early warning module is expanded. For example, in addition to sending emails and text messages to notify relevant personnel, it can also be integrated with the company's instant messaging tools (such as DingTalk, WeChat for Enterprise, etc.) to push the early warning information to the instant messaging client of relevant personnel; or display the early warning information on the company's internal electronic display screens, smart bracelets and other devices to ensure that relevant personnel can receive early warning notifications in a timely and comprehensive manner, thereby improving the efficiency of conveying early warning information.

[0077] 5. Visualization module:

[0078] 1) Enhanced interactivity of visualization module: The interactivity of visualization module is improved, so that users can more easily conduct in-depth exploration and analysis of matching analysis results and warning information. For example, data drilling function is provided, and users can click from the macro electricity-to-carbon ratio trend chart to the data details page of specific equipment or time period to view more detailed electricity consumption and carbon emission data; data filtering and screening functions are added, and users can filter and screen data according to different equipment types, production workshops, time ranges and other conditions to analyze data more accurately; at the same time, users are supported to customize the type, color, layout, etc. of visualization charts to meet the personalized needs of different users.

[0079] 2) Mobile adaptation of visualization module: Develop mobile applications for visualization modules so that enterprise managers and related personnel can view matching analysis results and warning information anytime and anywhere through mobile devices such as mobile phones and tablets. The mobile application should have a simple and clear interface design and a good user experience, and be able to quickly load data and display key information. For example, a simple dashboard design is used in the mobile application to highlight the real-time data of the electricity-to-carbon ratio, the warning level, and the latest abnormal event information, so that users can keep abreast of the company's electricity-to-carbon situation when they are out or working on the move, and make decisions and responses in a timely manner.

[0080] Through the above expansion of each link of the early warning method and system based on industry electric carbon data matching, the accuracy, flexibility, intelligence and practicality of the method and system in practical applications can be further improved, providing stronger support for the enterprise's electric carbon management and energy conservation and emission reduction.

[0081] The above is only a preferred specific implementation of the present invention; however, the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solution and its improved conception within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. The early warning method based on industry electricity carbon data matching is characterized by: The following steps are involved: 1) Data collection: When collecting data, set the unique identifier of the sensor and perform regular verification. During data transmission, use checksum and hash algorithms to verify the integrity of the data to ensure that the data has not been tampered with. 2) Data preprocessing: Clean, remove and standardize the collected power consumption data and carbon emission data, use box plot analysis to identify outliers, and combine visualization methods to help analyze data distribution; 3) Data matching analysis: Build a matching model between electricity consumption and carbon emissions and calculate the electricity-to-carbon ratio Where P(t) is the electricity consumption at time t, C(t) is the carbon emission at time t, and grid search and cross validation are used to optimize the model parameters; 4) Set warning thresholds: Set warning thresholds for the electricity-to-carbon ratio based on historical data and industry standards, set warning levels, and define trigger conditions for each level; The deviation of the carbon ratio from the benchmark value is set to 5-10% as a yellow warning, the deviation of the carbon ratio from the benchmark value is set to 10-20% as an orange warning, and the deviation of the carbon ratio from the benchmark value is set to more than the red warning. After the warning is triggered, the system automatically generates a report and sends it to the relevant personnel, and records the event log for subsequent analysis; 5) Real-time monitoring and early warning: The electricity-to-carbon ratio is monitored in real time. When the electricity-to-carbon ratio exceeds the early warning threshold, an early warning is triggered and an abnormal report is generated.

2. The early warning method based on industry electric carbon data matching according to claim 1 is characterized by:. The data collection step includes using industrial Internet of Things (IoT) devices to transmit data and using edge computing technology to perform preliminary processing on the data.

3. The early warning method based on industry electric carbon data matching according to claim 1 is characterized by: The data preprocessing steps include outlier detection, interpolation and standardization transformation to ensure data quality.

4. The early warning method based on industry electric carbon data matching according to claim 1 is characterized by: The data matching analysis step adopts a time series analysis model to predict and analyze the trend of the electricity-to-carbon ratio.

5. The early warning method based on industry electric carbon data matching according to claim 1 is characterized by: The warning information is displayed through a visualization platform, which supports real-time dynamic dashboards and multi-platform notification push, and uses WebSocket for real-time data transmission to ensure that warning information and data updates can be pushed to the user interface in real time.

6. An early warning system based on industry electricity carbon data matching, characterized by: 1) Data collection module: used to collect power consumption data and carbon emission data in real time; 2) Data preprocessing module: used to clean and standardize data; 3) Matching analysis module: used to calculate the electricity-to-carbon ratio and perform matching analysis; 4) Early warning module: used to monitor the electricity-to-carbon ratio in real time and trigger abnormal early warning; 5) Visualization module: used to display matching analysis results and warning information.

7. The early warning method based on industry electric carbon data matching according to claim 6 is characterized by: The warning module includes a rule engine that triggers warnings of different levels according to different threshold levels set by users.

Images