Intelligent smelting factory data visual management method and system

Abstract

The invention discloses an intelligent smelting factory data visual management method and system, and relates to the technical field of smelting data analysis, and the method comprises the steps: defining the grade of a smelting working condition state, dividing the grade of the smelting working condition state in detail according to the smelting condition, and providing a reliable basis for the configuration of a template of a subsequent smelting event. A template of each smelting event is configured according to historical smelting factory data and smelting working condition state grades, a current smelting event on a smelting factory data axis is matched by means of the smelting event templates, smelting data are converted into specific smelting events, and the accuracy of smelting event recognition is improved. All the current smelting events are associated, integrated and complemented based on the types and attributes of the current smelting events, the smelting events are integrated, the accuracy and adaptability of smelting data visualization are improved, surface smelting data are fused, deep information is output, the complex and changeable smelting requirements are met, and the smelting efficiency is improved. Decision assistance and direct management are better provided for managers.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smelting data analysis, and particularly relates to a smart smelting plant data visualization management method and system. BACKGROUND

[0002] In a smart smelting plant, the production process involves massive and complex data, covering equipment operating parameters, process indicators, quality detection data, etc. The traditional management method relies on manual recording and analysis, which is inefficient and prone to errors, and it is difficult to grasp the plant operation status in real time. With the development of industrial internet, big data, artificial intelligence and other technologies, data visualization management is possible. Through real-time data collection by sensors, big data technology is used for storage and processing, and then visualization technology is used to present the data in the form of intuitive charts, graphs, etc. such as line charts to show the temperature change trend of equipment, bar charts to compare the quality of different batches of products, etc. This enables management personnel to quickly understand the rules and problems behind the data, timely identify potential risks, optimize production processes, improve production efficiency and product quality, reduce operating costs, and meet the intelligent and fine management needs of the plant.

[0003] In the prior art, the visualization of smelting data only displays the directly collected data, which cannot directly provide reference and decision support for management personnel, resulting in poor accuracy and adaptability of smelting plant data visualization management, which cannot meet the complex and variable needs of smelting.

[0004] Therefore, how to fuse surface smelting data to output deep information to improve the accuracy and adaptability of smelting plant data visualization management is a technical problem to be solved at present. SUMMARY

[0005] The purpose of the present application is to solve the problem of poor accuracy and adaptability of smelting plant data visualization management in the prior art, and a smart smelting plant data visualization management method is proposed, which comprises, obtaining historical smelting plant data of all smelting events, defining smelting condition state levels, and configuring a template for each smelting event according to the historical smelting plant data and the smelting condition state levels; obtaining smelting plant data in the current period, expanding the smelting plant data in the form of a time axis to obtain a smelting plant data axis, and confirming all smelting event types; matching the current smelting events on the smelting plant data axis by means of the smelting event template, and counting the attributes of each current smelting event; associating and integrating all current smelting events based on the current smelting event types and attributes to obtain smelting deep information, and visualizing the smelting deep information.

[0006] In some embodiments of the present application, the smelting plant data includes smelting equipment data, smelting production data, smelting energy data and smelting quality data.

[0007] In some embodiments of the present application, the smelting condition state level is defined, including, According to the factors involved in the smelting equipment data, smelting production data, smelting energy data and smelting quality data, the fluctuation period of each parameter in the smelting equipment data, smelting production data, smelting energy data and smelting quality data is determined; According to the fluctuation period of each parameter in the smelting equipment data, smelting production data, smelting energy data and smelting quality data, the corresponding historical smelting plant data is divided to generate multiple sample data in each parameter, and the average value range and the change rate of each parameter are determined based on the multiple sample data; The smelting condition state value of each parameter is obtained by evaluation and analysis of the average value range of each parameter, and the smelting condition state level is defined based on the smelting condition state value and the change rate.

[0008] In some embodiments of the present application, according to the factors involved in the smelting equipment data, smelting production data, smelting energy data and smelting quality data, the fluctuation period of each parameter in the smelting equipment data, smelting production data, smelting energy data and smelting quality data is determined, including, The first fluctuation period of each parameter is obtained by statistical analysis of the historical smelting plant data of the smelting equipment data, smelting production data, smelting energy data and smelting quality data respectively; The factors involved in each parameter in the smelting equipment data, smelting production data, smelting energy data and smelting quality data are collected, and the second fluctuation period of each parameter is obtained by comprehensive analysis of the factors; The fluctuation period of each parameter is determined based on the first fluctuation period and the second fluctuation period of each parameter.

[0009] In some embodiments of the present application, the template of each smelting event is configured according to the historical smelting plant data and the smelting condition state level, including, The smelting features and smelting events are extracted from the historical smelting plant data, the smelting feature set matched by each smelting event is determined by principal component analysis, the smelting feature basic association rule of each smelting event is mined using Apriori algorithm, and all smelting condition state levels under each smelting event are counted; Based on all smelting condition state levels under a single smelting event, the smelting feature threshold, smelting feature category and weight of each smelting feature in the smelting feature basic association rule are adjusted, so as to configure the template of each smelting event under each smelting condition state level.

[0010] In some embodiments of this application, matching the current smelting event on the smelting plant data axis using a smelting event template includes, Determine the range of smelting operating condition status levels within the current period. Within the range of smelting operating condition status levels, match the smelting events on the smelting plant data axis using the matching degree according to the template of each smelting event under each smelting operating condition status level, and identify the current smelting event. The current smelting events include confirmed smelting events and suspected smelting events.

[0011] In some embodiments of this application, after statistically analyzing the attributes of each current smelting event, the method further includes, Analyze the relationships between all different smelting events, output pre-selected smelting depth information based on the relationships between different smelting events, and construct a relationship network between different smelting events and smelting depth information output relationship.

[0012] In some embodiments of this application, all current smelting events are correlated, integrated, and complemented based on the current smelting event type and attributes to obtain deeper smelting information, including: Mark the current smelting event type and attributes on the correlation network between smelting events and smelting depth information, and use different tags to mark confirmed smelting events and suspected smelting events; The state transition probability of each suspected smelting event is calculated using a Markov chain. Based on the state transition probability, the corresponding suspected smelting event is transformed into a confirmed smelting event. On the basis of the correlation network between smelting events and smelting depth information, multiple candidate smelting depth information is output. Based on the attributes of the determined smelting events, the correlation between the determined smelting events, and the state transition probability, the credibility of each candidate smelting depth information is calculated, and the final smelting depth information is output.

[0013] Correspondingly, this application also provides a data visualization management system for intelligent smelting plants, including, The first module is used to obtain historical smelting plant data for all smelting events, define smelting operating condition status levels, and configure a template for each smelting event based on historical smelting plant data and smelting operating condition status levels. The second module is used to obtain the smelting plant data within the current time period, expand the smelting plant data in the form of a time axis to obtain the smelting plant data axis, and confirm all smelting event types. The third module is used to match the current smelting event on the smelting plant data axis based on the smelting event template, and to count the attributes of each current smelting event. The fourth module is used to associate, integrate, and complement all current smelting events based on the current smelting event type and attributes to obtain in-depth smelting information, and then visualize this in-depth smelting information.

[0014] Compared with the prior art, the beneficial effects of this invention are as follows: 1. Define smelting condition status levels to comprehensively classify smelting conditions into detailed levels, providing a reliable foundation for configuring templates for subsequent smelting events. Configure templates for each smelting event based on historical smelting plant data and smelting condition status levels, adjusting the rules for each smelting event determined by historical data according to the smelting condition status levels.

[0015] 2. By using smelting event templates to match current smelting events on the smelting plant's data axis, smelting data is transformed into specific smelting events, reducing errors caused by data redundancy. Furthermore, matching smelting events using templates improves the accuracy of smelting event identification. Based on the current smelting event type and attributes, all current smelting events are correlated, integrated, and complementary to obtain deeper smelting information. Integrating smelting events outputs more in-depth smelting information, improving the accuracy and adaptability of smelting data visualization. By merging surface-level smelting data and outputting deeper information, it meets the complex and ever-changing smelting needs, better providing decision support and straightforward management for managers. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the intelligent smelting plant data visualization management method proposed in this invention. Figure 2 This is a schematic diagram of the structure of the intelligent smelting plant data visualization management system proposed in this invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0018] Reference Figure 1 The intelligent smelting plant data visualization management method includes the following steps: Step S101: Obtain historical smelting plant data for all smelting events, define smelting operating condition status levels, and configure a template for each smelting event based on the historical smelting plant data and smelting operating condition status levels.

[0019] In some embodiments of this application, the smelting plant data includes smelting equipment data, smelting production data, smelting energy data, and smelting quality data.

[0020] In this embodiment, historical smelting plant data related to all smelting events are acquired, covering smelting equipment data, smelting production data, smelting energy data, and smelting quality data, providing a comprehensive and accurate data foundation for subsequent smelting condition analysis, event template configuration, event matching, and in-depth information mining.

[0021] Smelting equipment data: Data content: Equipment operating parameters: such as equipment speed, temperature, pressure, vibration frequency, etc. Taking blast furnace ironmaking as an example, parameters such as blast furnace top pressure and hot blast stove temperature directly reflect the equipment operating status. Values ​​that are too high or too low may indicate equipment abnormalities or changes in smelting conditions.

[0022] Equipment maintenance records include information such as equipment repair time, repair details, and replaced parts. These records help understand the equipment's health status and lifespan; for example, frequently repaired parts may be high-risk areas for equipment failure.

[0023] Equipment Failure Records: Record the time of equipment failure, the symptoms of the failure, the cause of the failure, and the results of the handling. For example, if a rolling mill experiences a roll breakage failure, record the location of the breakage and possible causes (such as excessive rolling force, roll material problems, etc.) to provide a basis for subsequent analysis.

[0024] How to obtain Sensor data acquisition: Sensors are installed at key parts of the equipment to collect operating parameters in real time and transmit them to a database via a data acquisition system. For example, a temperature sensor is installed at the oxygen lance location in the converter to monitor the oxygen lance temperature in real time.

[0025] Equipment Management System: Retrieves equipment maintenance and fault records from the enterprise's Equipment Management System (EAM). These systems typically maintain detailed records of equipment maintenance and fault conditions, and the relevant data can be obtained through interfaces or data export functions.

[0026] Smelting production data: Data content: Production plan: This includes information such as production tasks, production batches, and production schedule. For example, the plan for a certain month is to produce 1,000 tons of steel of a specific specification, which will be produced in 5 batches.

[0027] Process parameters, such as smelting temperature, smelting time, and raw material ratio, directly affect product quality and production efficiency in electrolytic aluminum production, including the current intensity and electrolysis temperature of the electrolytic cell.

[0028] Production data: Records the actual output of each production batch or time period. For example, if the actual output of a batch of steel is 200 tons, comparing it with the planned output can help assess production performance.

[0029] How to obtain Manufacturing Execution System (MES): The MES system is responsible for monitoring and managing the production process, and production plans, process parameters, and output data can be obtained from this system. Through integration with the MES system, real-time or periodic data acquisition can be achieved.

[0030] Production reports: Companies typically generate production reports regularly to record key data from the production process. These reports can be obtained from relevant departments, and the data can be organized and entered.

[0031] Smelting energy data: Data content: Energy consumption: This refers to the consumption of energy sources such as electricity, coal, and natural gas. In steel enterprises, processes such as blast furnace ironmaking and converter steelmaking consume a large amount of energy. Accurately recording energy consumption helps in assessing energy utilization efficiency.

[0032] Energy prices: Fluctuations in energy prices at different times can affect production costs. For example, electricity prices differ significantly between peak and off-peak periods; recording energy prices allows for cost analysis and optimization.

[0033] Energy quality data, such as the calorific value of coal and the calorific value of natural gas, are crucial. The quality of energy directly impacts the smelting process and product quality; for example, low-calorific-value coal can lead to unstable blast furnace temperatures.

[0034] How to obtain Energy metering system: Metering equipment, such as electricity meters and coal flow meters, is installed at the energy supply and usage stages to collect energy consumption data in real time and transmit it to the energy management system.

[0035] Energy suppliers: Partner with energy suppliers to obtain energy price and quality data. Some energy suppliers provide detailed energy quality reports and pricing information.

[0036] Smelting quality data: Data content: Product quality indicators include the chemical composition, mechanical properties, and surface quality of steel. For copper smelting products, the purity of copper and the content of impurities are important quality indicators.

[0037] Quality inspection records: Record information such as the time of product inspection, inspection method, and inspection results. For example, when using a spectrometer to detect the chemical composition of steel, record the inspection data and the personnel involved.

[0038] Non-conforming product record: Record the quantity, cause, and handling method of non-conforming products. For example, if a batch of steel is judged to be non-conforming because its mechanical properties do not meet the standards, the cause needs to be analyzed and corrective measures taken.

[0039] How to obtain Quality Inspection System: This system retrieves product quality indicators and inspection records from the company's Quality Inspection System (QIS). It typically performs real-time monitoring and recording of various product quality indicators.

[0040] Quality Management Department: The quality management department is responsible for the management and control of product quality. You can obtain records of non-conforming products and related analysis reports from this department.

[0041] In some embodiments of this application, smelting operating condition levels are defined, including... The fluctuation period of each parameter in the smelting equipment data, smelting production data, smelting energy data, and smelting quality data is determined based on the factors involved in each of these data. The historical smelting plant data is divided according to the fluctuation cycle of each parameter in the smelting equipment data, smelting production data, smelting energy data, and smelting quality data. Multiple sample data are generated for each parameter, and the average range and rate of change of each parameter are determined based on the multiple sample data. The smelting condition status value of each parameter is obtained by evaluating and analyzing the average range of each parameter, and the smelting condition status level is defined based on the smelting condition status value and the rate of change.

[0042] In this embodiment, the fluctuation period is determined by combining historical data of smelting equipment, smelting production, smelting energy, and smelting quality, along with their respective influencing factors, because the fluctuation period varies for each parameter. The fluctuation period of each parameter is used to divide the corresponding historical smelting plant data, generating multiple sample data sets for each parameter. That is, the same parameter is divided according to the time dimension of the fluctuation period, resulting in multiple sample data sets, each with the same time frame. Based on the determined fluctuation period, the historical smelting plant data is divided into multiple time periods, generating multiple sample data sets for each parameter. For example, the temperature data of a certain equipment is divided into 4.5-hour periods, resulting in temperature data samples for multiple time periods. Based on the average range of each parameter, its smelting operating condition status value (describing the safety and stability of the smelting operating condition) is evaluated. The calculation formula for the smelting operating condition status level, defined based on the smelting operating condition status value and the rate of change, is as follows: ; in, The 4 represents the smelting operating condition level, indicating four types of data: smelting equipment data, smelting production data, smelting energy data, and smelting quality data. Let the operating condition weights be defined for the i-th type of data among the four types of data: smelting equipment data, smelting production data, smelting energy data, and smelting quality data. This refers to the corrected smelting operating condition value for the first category of data among four types: smelting equipment data, smelting production data, smelting energy data, and smelting quality data. for The maximum value in, The second constant of the i-th type of data, This represents the correction of the maximum value to the average value. The second constant is to balance the magnitude of the correction function. n is the number of parameters for the i-th type of data, which includes four types of data: smelting equipment data, smelting production data, smelting energy data, and smelting quality data. Let the influence weight of the j-th parameter be the data of the i-th class. For the smelting operating condition value under the i-th type of data, Let j be the rate of change of the j-th parameter under the i-th data type. The first constant corresponding to the j-th parameter. This represents the correction of the rate of change to the smelting operating condition value; the first constant is used to balance the magnitude of the correction function. The third constant is used to balance the magnitude of the smelting condition level, and [] is the rounding symbol.

[0043] In some embodiments of this application, the fluctuation period of each parameter in the smelting equipment data, smelting production data, smelting energy data, and smelting quality data is determined based on factors related to each of these factors, including: Statistical analysis was conducted on historical smelting plant data for smelting equipment, smelting production, smelting energy, and smelting quality respectively to obtain the first fluctuation period of each parameter. Collect the factors involved in each parameter from smelting equipment data, smelting production data, smelting energy data, and smelting quality data, and comprehensively analyze the factors to obtain the second fluctuation period of each parameter; The fluctuation period of each parameter is determined based on the first and second fluctuation periods of each parameter.

[0044] In this embodiment, time series analysis is performed on the historical data of each parameter, using methods such as autocorrelation analysis and Fourier transform to identify the periodic characteristics of the data. For example, equipment temperature data may show periodic fluctuations every 4 hours. Based on the time series analysis results, the initial fluctuation period of each parameter is determined and denoted as the first fluctuation period. For example, the first fluctuation period of equipment temperature is 4 hours.

[0045] Collect the influencing factors for each parameter, including: Equipment factors: equipment aging, maintenance cycle, changes in equipment load, etc.

[0046] Production factors: adjustments to production plans, changes in process parameters, changes in raw materials, etc.

[0047] Energy factors: stability of energy supply, fluctuations in energy prices, changes in energy quality, etc.

[0048] Quality factors: changes in product quality standards, adjustments to testing methods, handling of non-conforming products, etc.

[0049] Use statistical analysis methods (such as regression analysis, analysis of variance) or expert experience to assess the extent to which each factor affects the parameter fluctuation.

[0050] For example, a production plan adjustment might cause the equipment temperature fluctuation cycle to become 5 hours. Combining the analysis results of influencing factors, the first fluctuation cycle is adjusted to obtain the second fluctuation cycle. For example, the second fluctuation cycle of the equipment temperature is 5 hours. The first and second fluctuation cycles are then combined, and a weighted average or other comprehensive calculation is performed to determine the final fluctuation cycle of each parameter. For example, the final fluctuation cycle of the equipment temperature is 4.5 hours, taking into account the results of historical data analysis and adjustments to influencing factors.

[0051] In some embodiments of this application, a template for each smelting event is configured based on historical smelting plant data and smelting operating condition levels, including: Smelting features and events are extracted from historical smelting plant data. Principal component analysis is used to determine the set of smelting features matched for each smelting event. The Apriori algorithm is used to mine the basic association rules of smelting features for each smelting event. All smelting operating condition status levels under each smelting event are statistically analyzed. The smelting feature threshold, smelting feature category, and weight of each smelting feature in the basic association rules of smelting features are adjusted based on all smelting condition status levels under a single smelting event, thereby configuring the template for each smelting event under each smelting condition status level.

[0052] In this embodiment, smelting features are extracted from historical data. These features can include equipment parameters (such as temperature and pressure), production parameters (such as smelting time and raw material ratio), energy parameters (such as electricity consumption), and quality parameters (such as product purity). For example, features such as furnace top pressure, hot blast temperature, and molten iron composition are extracted from blast furnace ironmaking data. Smelting events are identified based on anomalies or specific patterns in historical data. For example, equipment failure events, production anomaly events, and energy waste events. Each event is labeled and classified to facilitate subsequent analysis. The standardized smelting features are analyzed using the PCA method to identify the main features affecting smelting events. Through PCA dimensionality reduction, the set of main smelting features corresponding to each smelting event is determined. For example, blast furnace failure events may be mainly related to furnace top pressure and hot blast temperature. According to the definition of smelting operating condition levels, the occurrence of each smelting event under different operating conditions is statistically analyzed. For example, the frequency of occurrence of blast furnace failure events under normal, warning, and failure conditions is statistically analyzed. The differences in smelting features under different operating conditions are analyzed to identify key features leading to state changes. For example, under failure conditions, furnace top pressure is significantly higher than under normal conditions. Based on the differences in characteristics under different operating conditions, the thresholds in the smelting feature association rules are adjusted. For example, under fault conditions, the threshold for furnace top pressure is adjusted from 200 kPa under normal conditions to 220 kPa. The weights of features are adjusted according to their importance under different operating conditions. For example, under fault conditions, the weight of hot blast temperature may be increased to reflect its greater impact on the event. Combining the adjusted thresholds and weights, the smelting feature association rules are updated to form rule sets for different operating conditions. Based on the adjusted smelting feature association rules, templates are configured for each smelting event under different operating conditions. Templates include feature sets, thresholds, weights, and logical relationships. For example, the template for a blast furnace fault event under fault conditions might include rules such as furnace top pressure > 220 kPa and hot blast temperature < 1000℃.

[0053] Step S102: Obtain the smelting plant data for the current period of time, expand the smelting plant data in the form of a time axis to obtain the smelting plant data axis, and confirm all smelting event types.

[0054] In this embodiment, the time range to be analyzed is determined, such as the most recent week, month, or quarter, to ensure sufficient and representative data volume. Data is collected in real time using a data acquisition system (such as SCADA or DCS) and stored in a database or data lake for subsequent processing and analysis. Data from different sources is aligned on the time axis to address issues such as inconsistent timestamps or different data sampling frequencies. For example, equipment sensor data may be collected once per second, while production management system data may be updated once per minute. Data from various smelting plants is expanded chronologically along the horizontal axis to form a smelting plant data axis. For example, temperature, pressure, and output data are arranged chronologically to form time series data. Data visualization tools (such as Matplotlib or Tableau) are used to visualize the data axis, facilitating intuitive observation of data trends and patterns. For example, a curve showing temperature changes over time is plotted. Based on the smelting process and business needs, possible smelting event types are defined, such as: Equipment malfunction events: such as equipment overheating, abnormal pressure, excessive vibration, etc.

[0055] Abnormal production events: such as excessive smelting time, deviation in raw material ratio, insufficient output, etc.

[0056] Energy waste incidents include excessive electricity consumption and low coal utilization.

[0057] Quality non-compliance incidents: such as excessive product ingredients or substandard mechanical properties.

[0058] Step S103: Match the current smelting event on the smelting plant data axis using the smelting event template, and count the attributes of each current smelting event.

[0059] In some embodiments of this application, matching the current smelting event on the smelting plant data axis using a smelting event template includes, Determine the range of smelting operating condition status levels within the current period. Within the range of smelting operating condition status levels, match the smelting events on the smelting plant data axis using the matching degree according to the template of each smelting event under each smelting operating condition status level, and identify the current smelting event. The current smelting events include confirmed smelting events and suspected smelting events.

[0060] In this embodiment, pre-configured smelting event templates are loaded from a database or configuration file. Each template corresponds to a smelting event type and includes feature sets, thresholds, weights, and logical relationships under different operating condition levels.

[0061] Match the smelter plant data axis (time series data) with the smelter event template. The specific steps are as follows: Data segmentation: The data axis of the smelting plant is segmented according to time windows, with each time window corresponding to a potential event occurrence period.

[0062] Feature extraction: Extract smelting features such as temperature, pressure, and output from each time window.

[0063] Template matching: The extracted features are matched with smelting event templates, and the matching degree is calculated. The matching degree can be calculated using methods such as feature similarity, threshold satisfaction, and weighted sum.

[0064] For each smelting event template, calculate its matching degree with the current data window. For example: Feature similarity: Calculate the similarity between the current feature and the template feature (such as Euclidean distance, cosine similarity).

[0065] Threshold satisfaction: Check whether the current feature meets the threshold conditions defined in the template (e.g., temperature > 1200℃).

[0066] Weighted sum: Calculate the weighted matching score based on the weights defined in the template.

[0067] Based on the matching results, identify the current smelting event: The smelting event has been identified: the matching degree is higher than the preset threshold (e.g., 90%), and the features and thresholds fully meet the template definition.

[0068] Suspected smelting event: The matching degree is within the preset range (e.g., 70%-90%), but some features or thresholds do not fully meet the template definition and need further confirmation.

[0069] Attribute statistics are performed on the identified current smelting events (including confirmed and suspected events), including but not limited to: Event types: such as equipment overheating, excessive smelting time, excessive power consumption, and excessive iron composition.

[0070] Time of occurrence: The specific point in time or period during which the event occurred.

[0071] Duration: The length of time from the start to the end of an event.

[0072] Scope of impact: The equipment, production line, or product quality affected by the incident.

[0073] Matching score: The score of how well the event matches the template.

[0074] Event attributes are recorded in a database or log system to generate an event report. The report may include an event description, the time of occurrence, the duration, the scope of impact, and recommended actions.

[0075] In some embodiments of this application, after statistically analyzing the attributes of each current smelting event, the method further includes, Analyze the relationships between all different smelting events, output pre-selected smelting depth information based on the relationships between different smelting events, and construct a relationship network between different smelting events and smelting depth information output relationship.

[0076] Event Correlation Analysis Time series analysis: Using time series analysis methods (such as sliding window and time window clustering), the chronological order or simultaneous occurrence of smelting events can be identified. For example, an equipment overheating event may be immediately followed by an event of excessively long smelting time.

[0077] Causal relationship analysis: Using causal inference methods (such as Granger causality test, Bayesian network) to analyze the causal relationships between smelting events. For example, deviations in raw material proportions may lead to product quality defects.

[0078] Statistical correlation analysis: This involves calculating the correlation between smelting events using statistical methods (such as chi-square test and mutual information). For example, equipment failure events and energy waste events may be significantly correlated. Representing the relationships between smelting events as a graph structure or matrix facilitates subsequent analysis and application. For instance, a smelting event correlation graph can be constructed, where nodes represent events and edges represent relationships.

[0079] Deeper information in smelting refers to deeper production problems or optimization opportunities reflected through the correlation of smelting events, such as: Equipment health status: Frequent equipment failures may reflect equipment aging or insufficient maintenance.

[0080] Production process optimization: Excessive smelting time and deviations in raw material ratios may indicate that the production process needs adjustment.

[0081] Energy management optimization: The correlation between energy waste events and equipment failure events may indicate that the energy management system needs to be optimized.

[0082] Based on the correlation of smelting events, extract in-depth information for pre-selected smelting processes. For example: If equipment overheating events and excessively long smelting times occur frequently and simultaneously, "low efficiency of the blast furnace cooling system" may be extracted as deeper information.

[0083] If the deviation in raw material ratio is related to the product quality failure, "lax raw material quality control" may be extracted as deeper information.

[0084] The structure of the smelting event-smelting deep information network is usually designed using graph structures (such as Bayesian networks and knowledge graphs) or matrix forms.

[0085] Nodes represent smelting events or deep smelting information, and edges represent the relationships between events and deep information.

[0086] Step S104: Based on the current smelting event type and attributes, all current smelting events are correlated, integrated and complemented to obtain in-depth smelting information, and the in-depth smelting information is visualized.

[0087] In this embodiment, confirmed smelting events and suspected smelting events are combined. Suspected smelting events indicate the possible occurrence of an event, which requires further analysis. The correlation and complementarity of different smelting events are considered to output deeper information.

[0088] Choose a visualization tool: Select a suitable visualization tool based on your needs, such as Tableau, Power BI, Matplotlib, D3.js, etc.

[0089] Design the visualization layout: Determine the layout and structure of the visualization, such as timelines, event diagrams, dashboards, etc.

[0090] Event Timeline: Displays the occurrence time and duration of smelting events in a timeline format, making it easy to observe the time distribution and trends of events.

[0091] Event Relationship Graph: A graph structure is used to show the relationships between smelting events. Nodes represent events and edges represent relationships, making it easy to observe the complex relationships between events.

[0092] Deep Information Dashboard: Displays deep information about smelting in the form of a dashboard, such as equipment health status scores, production process optimization suggestions, and energy management improvement measures, making it easy to quickly understand the production status and optimization direction.

[0093] Interactive feature design: Add interactive features to the visualization, such as zooming, filtering, drill-down, etc., to facilitate users to explore and analyze the data in depth.

[0094] Dynamic updates: Supports dynamic updates for visual display, reflecting changes in smelting events and in-depth information in real time.

[0095] In some embodiments of this application, all current smelting events are correlated, integrated, and complemented based on the current smelting event type and attributes to obtain deeper smelting information, including: Mark the current smelting event type and attributes on the correlation network between smelting events and smelting depth information, and use different tags to mark confirmed smelting events and suspected smelting events; The state transition probability of each suspected smelting event is calculated using a Markov chain. Based on the state transition probability, the corresponding suspected smelting event is transformed into a confirmed smelting event. On the basis of the correlation network between smelting events and smelting depth information, multiple candidate smelting depth information is output. Based on the attributes of the determined smelting events, the correlation between the determined smelting events, and the state transition probability, the credibility of each candidate smelting depth information is calculated, and the final smelting depth information is output.

[0096] In this embodiment, the current smelting event type and attributes are marked on the network of associations between smelting events and smelting depth information.

[0097] Confirmed smelting events: Represented using a marker (such as a green node).

[0098] Suspected smelting event: Represented using an alternative marker (such as a yellow node).

[0099] Based on the correlation between smelting events and the deep information of smelting, a correlation network between smelting events and deep information of smelting is constructed.

[0100] Nodes represent smelting events or deep smelting information, and edges represent the relationships between events or between events and deep information.

[0101] Markov chains are used to model the state transitions of suspected smelting events.

[0102] The state is defined as whether a suspected smelting event is a confirmed smelting event (i.e., a template for a suspected smelting event). The state transition probability of each suspected smelting event is calculated, i.e., the probability of transitioning from a suspected state to a confirmed state. A state transition probability matrix is ​​calculated based on historical data or expert experience. For example, if the state transition probability of suspected smelting event A is 0.8, it means there is an 80% probability of it transforming into a confirmed smelting event. Based on the state transition probability, the suspected smelting event is transformed into a confirmed smelting event. For example, if the state transition probability of suspected smelting event A is 0.8, it is transformed into a confirmed smelting event. When the state transition probability is high, it indicates that the suspected smelting event may transform or evolve into the confirmed smelting event in the future. Based on the network of connections between smelting events and deep smelting information, the combinations of smelting events under different paths are analyzed. Multiple confirmed smelting events in the same group may output multiple candidate deep smelting information items due to different paths. Multiple candidate deep smelting information items are generated based on the path analysis results. For example, path 1: event A → event B → deep information C; path 2: event A → event D → deep information E.

[0103] The credibility of each candidate smelting deep information is calculated based on the attributes of the determined smelting events, the correlation between the determined smelting events, and the state transition probability. The calculation formula is as follows: ; The confirmed smelting events are divided into first smelting events and second smelting events. The first smelting event is the original confirmed smelting event, and the second smelting event is a confirmed smelting event transformed from a suspected smelting event. It is the first The reliability of candidate deep smelting information is determined by the correlation network between smelting events and deep smelting information. For the same group of multiple determined smelting events, different paths may result in the output of multiple candidate deep smelting information pieces. , These are the reliable conversion coefficients for the first and second smelting events, respectively. , These represent the quantities of the first and second smelting events, respectively. , The first The first smelting event and the first The combined weight of each second smelting event, , The first The first smelting event and the first The degree of impact of each secondary smelting event is determined by the sum of its individual attributes. For the first The state transition probability of the second smelting event. For the first The adjustment coefficient for the second smelting event (representing the correction of the state transition probability to the degree of influence). Indicates by the first The adjustment coefficient is obtained by mapping the state transition probability of the second smelting event. For the first The correlation between the determined smelting events corresponding to the paths of the candidate smelting deep information. For the first The fourth constant of the candidate smelting depth information The fourth constant is a correction to the sum of the influence of the correlation between the determined smelting events corresponding to the path on the first and second smelting events.

[0104] Understandably, key attributes are extracted from each smelting event, such as event occurrence time, duration, scope of impact, severity, and economic loss. Since different attributes may have different dimensions and ranges, they need to be standardized for subsequent calculations. A weighted summation is then used to determine the degree of event impact. Correlation represents the closeness of the connection between smelting events and can be measured by temporal correlation, causal correlation, or statistical correlation. Within the network of connections between smelting events and deeper smelting information, a path for a candidate piece of deeper smelting information is extracted. The path consists of a series of defined smelting events, for example: Event A → Event B → Deep Information C. The correlation degree between each pair of adjacent smelting events in the path is calculated, and these correlation degrees are then combined to obtain the path correlation degree.

[0105] Correspondingly, this application also provides a data visualization management system for intelligent smelting plants, such as... Figure 2 As shown, including, The first module is used to obtain historical smelting plant data for all smelting events, define smelting operating condition status levels, and configure a template for each smelting event based on historical smelting plant data and smelting operating condition status levels. The second module is used to obtain the smelting plant data within the current time period, expand the smelting plant data in the form of a time axis to obtain the smelting plant data axis, and confirm all smelting event types. The third module is used to match the current smelting event on the smelting plant data axis based on the smelting event template, and to count the attributes of each current smelting event. The fourth module is used to associate, integrate, and complement all current smelting events based on the current smelting event type and attributes to obtain in-depth smelting information, and then visualize this in-depth smelting information.

[0106] Compared with the prior art, the beneficial effects of this invention are as follows: 1. Define smelting condition status levels to comprehensively classify smelting conditions into detailed levels, providing a reliable foundation for configuring templates for subsequent smelting events. Configure templates for each smelting event based on historical smelting plant data and smelting condition status levels, adjusting the rules for each smelting event determined by historical data according to the smelting condition status levels.

[0107] 2. By using smelting event templates to match current smelting events on the smelting plant's data axis, smelting data is transformed into specific smelting events, reducing errors caused by data redundancy. Furthermore, matching smelting events using templates improves the accuracy of smelting event identification. Based on the current smelting event type and attributes, all current smelting events are correlated, integrated, and complementary to obtain deeper smelting information. Integrating smelting events outputs more in-depth smelting information, improving the accuracy and adaptability of smelting data visualization. By merging surface-level smelting data and outputting deeper information, it meets the complex and ever-changing smelting needs, better providing decision support and straightforward management for managers.

[0108] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0109] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0110] Those skilled in the art will understand that the modules in the system of the implementation scenario can be distributed throughout the system of the implementation scenario as described, or they can be modified to reside in one or more systems different from this implementation scenario. The modules of the above-mentioned implementation scenario can be merged into one module, or they can be further divided into multiple sub-modules.

[0111] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A data visualization management method for intelligent smelting plants, characterized in that, include, Obtain historical smelting plant data for all smelting events, define smelting condition status levels, and configure a template for each smelting event based on the historical smelting plant data and smelting condition status levels. Obtain the smelting plant data for the current period, expand the smelting plant data in the form of a time axis to obtain the smelting plant data axis, and confirm all smelting event types; Use smelting event templates to match the current smelting events on the smelting plant's data axis and count the attributes of each current smelting event; Based on the current smelting event type and attributes, all current smelting events are correlated, integrated, and complemented to obtain in-depth smelting information, which is then visualized.

2. The intelligent smelting plant data visualization management method according to claim 1, characterized in that, Smelting plant data includes smelting equipment data, smelting production data, smelting energy data, and smelting quality data.

3. The intelligent smelting plant data visualization management method according to claim 2, characterized in that, Define the smelting operating condition levels, including: The fluctuation period of each parameter in the smelting equipment data, smelting production data, smelting energy data, and smelting quality data is determined based on the factors involved in each of these data. The historical smelting plant data is divided according to the fluctuation cycle of each parameter in the smelting equipment data, smelting production data, smelting energy data, and smelting quality data. Multiple sample data are generated for each parameter, and the average range and rate of change of each parameter are determined based on the multiple sample data. The smelting condition status value of each parameter is obtained by evaluating and analyzing the average range of each parameter, and the smelting condition status level is defined based on the smelting condition status value and the rate of change.

4. The intelligent smelting plant data visualization management method according to claim 3, characterized in that, The fluctuation period of each parameter in the smelting equipment data, smelting production data, smelting energy data, and smelting quality data is determined based on the factors involved in each of these data, including: Statistical analysis was conducted on historical smelting plant data for smelting equipment, smelting production, smelting energy, and smelting quality respectively to obtain the first fluctuation period of each parameter. Collect the factors involved in each parameter from smelting equipment data, smelting production data, smelting energy data, and smelting quality data, and comprehensively analyze the factors to obtain the second fluctuation period of each parameter; The fluctuation period of each parameter is determined based on the first and second fluctuation periods of each parameter.

5. The intelligent smelting plant data visualization management method according to claim 1, characterized in that, Configure a template for each smelting event based on historical smelting plant data and smelting operating condition levels, including: Smelting features and events are extracted from historical smelting plant data. Principal component analysis is used to determine the set of smelting features matched for each smelting event. The Apriori algorithm is used to mine the basic association rules of smelting features for each smelting event. All smelting operating condition status levels under each smelting event are statistically analyzed. The smelting feature threshold, smelting feature category, and weight of each smelting feature in the basic association rules of smelting features are adjusted based on all smelting condition status levels under a single smelting event, thereby configuring the template for each smelting event under each smelting condition status level.

6. The intelligent smelting plant data visualization management method according to claim 5, characterized in that, Use smelting event templates to match current smelting events on the smelting plant's data axis, including: Determine the range of smelting operating condition status levels within the current period. Within the range of smelting operating condition status levels, match the smelting events on the smelting plant data axis using the matching degree according to the template of each smelting event under each smelting operating condition status level, and identify the current smelting event. The current smelting events include confirmed smelting events and suspected smelting events.

7. The intelligent smelting plant data visualization management method according to claim 6, characterized in that, After analyzing the attributes of each current smelting event, the method further includes, Analyze the relationships between all different smelting events, output pre-selected smelting depth information based on the relationships between different smelting events, and construct a relationship network between different smelting events and smelting depth information output relationship.

8. The intelligent smelting plant data visualization management method according to claim 7, characterized in that, Based on the current smelting event type and attributes, all current smelting events are correlated, integrated, and complemented to obtain deeper smelting information, including: Mark the current smelting event type and attributes on the correlation network between smelting events and smelting depth information, and use different tags to mark confirmed smelting events and suspected smelting events; The state transition probability of each suspected smelting event is calculated using a Markov chain. Based on the state transition probability, the corresponding suspected smelting event is transformed into a confirmed smelting event. On the basis of the correlation network between smelting events and smelting depth information, multiple candidate smelting depth information is output. Based on the attributes of the determined smelting events, the correlation between the determined smelting events, and the state transition probability, the credibility of each candidate smelting depth information is calculated, and the final smelting depth information is output.

9. A data visualization management system for intelligent smelting plants, characterized in that, include, The first module is used to obtain historical smelting plant data for all smelting events, define smelting operating condition status levels, and configure a template for each smelting event based on historical smelting plant data and smelting operating condition status levels. The second module is used to obtain the smelting plant data within the current time period, expand the smelting plant data in the form of a time axis to obtain the smelting plant data axis, and confirm all smelting event types. The third module is used to match the current smelting event on the smelting plant data axis based on the smelting event template, and to count the attributes of each current smelting event. The fourth module is used to associate, integrate, and complement all current smelting events based on the current smelting event type and attributes to obtain in-depth smelting information, and then visualize this in-depth smelting information.

Images