Direct-current submerged arc furnace and detection system

Abstract

The invention relates to the technical field of chemical detection data processing, provides a direct-current submerged arc furnace and a detection system, and aims to solve the problem that an existing direct-current submerged arc furnace cannot detect chemical substances in the furnace in real time, so that the state of equipment cannot be pre-judged and pre-warned in time. The data acquisition unit acquires concentration data and other basic data of chemical substances in the furnace in real time, the detection unit performs preprocessing and feature extraction on the data, and the data processing unit constructs a knowledge base, retrieves fault data to generate chemical detection substances and detects the chemical substances in the furnace. And according to the detection result, real-time electrical equipment adjustment and maintenance data is generated, and the operation state of the equipment is optimized. According to the invention, the real-time detection of the chemical substances in the direct current submerged arc furnace is realized, the pre-judgment and early warning capabilities of the equipment state are improved, and the pertinence of fault processing is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical detection data processing, in particular to a DC ore-fired furnace and a detection system. Background Art

[0002] A DC electric arc furnace is a type of smelting equipment that utilizes direct current (DC) for heating. Its operating principle is to convert electrical energy into heat, melting ore or metal raw materials through the high temperature of the electric arc. Compared to AC electric arc furnaces, DC electric arc furnaces utilize DC power. This DC supply stabilizes the arc, facilitating process control and improving smelting efficiency. DC electric arc furnaces are widely used in fields such as ore-thermal smelting and steelmaking, and are an essential piece of equipment in the modern metallurgical industry.

[0003] Existing DC submerged arc furnaces typically use AC power, which results in low energy efficiency. Furthermore, due to the large arc fluctuations generated by AC power, additional cooling, detection, and monitoring devices are required to dissipate the heat generated by the arc. These multiple devices monitor and adjust the DC submerged arc furnace to ensure stable operation. However, existing DC submerged arc furnaces are unable to detect chemical substances within them during actual use, making it impossible to provide timely predictions and warnings about the status of various DC submerged arc furnace devices. To address this technical problem, the present invention provides a DC submerged arc furnace and detection system. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the present invention provides a DC ore-burning furnace and a detection system to solve the problem that the existing DC ore-burning furnace cannot detect the chemical substances in the DC ore-burning furnace during actual use, resulting in the inability to timely predict and warn the conditions of various equipment in the DC ore-burning furnace.

[0005] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows: In a first aspect, the present invention provides a DC ore-fired furnace, comprising: Data acquisition unit, used to obtain basic data of DC submerged arc furnace; A detection unit, used for preprocessing the basic data of the DC submerged arc furnace to obtain a basic data feature group of the DC submerged arc furnace; A state detection unit is used to establish an operation state detection model of the DC submerged arc furnace based on the basic data feature group of the DC submerged arc furnace, extract the features of the DC submerged arc furnace collected in real time, and input them into the operation state detection model of the DC submerged arc furnace to generate the DC submerged arc furnace state detection results. The DC submerged arc furnace state detection results include the DC submerged arc furnace state normal state, the DC submerged arc furnace state warning state, and the DC submerged arc furnace state fault state; The early warning unit is used to trigger the DC submerged arc furnace early warning process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state early warning state, and to trigger the DC submerged arc furnace fault detection process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state fault state; The data processing unit is used to obtain the historical operation data of the DC submerged arc furnace, build a DC submerged arc furnace knowledge base, retrieve the fault data corresponding to the state and fault state of the DC submerged arc furnace, obtain real-time fault data, extract data features of the real-time fault data, match the obtained real-time fault feature data in the DC submerged arc furnace knowledge base, obtain real-time fault feature association data, the real-time fault feature association data includes historical chemical substance detection data and historical electrical equipment adjustment and maintenance data, generate chemical detection substances based on the historical chemical substance detection data, use the chemical detection substances to detect the DC submerged arc furnace, obtain the chemical detection results of the DC submerged arc furnace, match the DC submerged arc furnace chemical detection results with the historical electrical equipment adjustment and maintenance data, obtain real-time electrical equipment adjustment and maintenance data, and send the real-time electrical equipment adjustment and maintenance data to the device end of the DC submerged arc furnace.

[0006] Furthermore, the data acquisition unit of the DC ore-fired furnace of the present invention is further used for: The data acquisition unit initiates a data acquisition request to the sensor in the DC submerged arc furnace, requesting real-time data of the chemical substances in the furnace. The request message body carries the data acquisition unit address, sensor address, time range of the requested data, and requested data type. The requested data type includes temperature, pressure, and chemical component concentration. The sensor responds to the request and returns the original detection data of the DC submerged arc furnace.

[0007] Furthermore, the detection unit of the DC ore-fired furnace of the present invention is also used for: The detection unit receives the original detection data from the sensor, removes duplicate, invalid or erroneous data, calibrates the cleaned data, and labels the temperature data, pressure data, and chemical composition concentration data according to the characteristics of the detection data.

[0008] Furthermore, the data processing unit of the DC ore-fired furnace of the present invention is also used for: Collect chemical substance data in the furnace and identify the type, concentration and change trend of chemical substances in the furnace through a preset algorithm model; Receive and set thresholds, build an anomaly detection model, use the anomaly detection model to detect the type, concentration and change trend of chemical substances in the furnace, obtain abnormal detection results of chemical substances in the furnace, and alarm for data that exceeds the normal range, indicating existing equipment failures or safety hazards.

[0009] Based on historical and current data, predict future trends in the chemical state of the furnace and equipment operation.

[0010] Furthermore, the data processing unit of the DC ore-fired furnace of the present invention is also used for: Retrieving historical chemical substance detection data from a DC ore-fired furnace knowledge base, preprocessing the acquired historical chemical substance detection data, and obtaining preprocessed historical chemical substance detection data; Use a preset algorithm model to analyze the pre-processed historical chemical substance detection data to identify the chemical substances in the DC submerged arc furnace and the characteristics of the DC submerged arc furnace; According to the chemical substances in the DC ore-forming furnace and the characteristics of the DC ore-forming furnace, matching is performed in the DC ore-forming furnace knowledge base to obtain the corresponding chemical detection substances.

[0011] Apply chemical test substances to the parts to be tested of the DC ore-fired furnace, including the furnace interior and the furnace wall of the DC ore-fired furnace, and collect reaction data between the chemical test substances and the chemical substances in the DC ore-fired furnace; Perform data feature extraction on the reaction data between the chemical test substance and the chemical substances in the DC submerged arc furnace to obtain the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace. Compare and analyze the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace with the preset chemical test standards to obtain the chemical test results; Collect the operating data after the device end of the DC submerged arc furnace executes the real-time electrical equipment adjustment and maintenance data to obtain the adjusted operating data, and compare the adjusted operating data with the preset DC submerged arc furnace standard data. If the adjusted operating data is not in the preset DC submerged arc furnace standard data, use the distillation algorithm to optimize the real-time electrical equipment adjustment and maintenance data, and send the optimized real-time electrical equipment adjustment and maintenance data to the DC submerged arc furnace device end.

[0012] In a second aspect, the present invention provides a DC ore-bearing furnace detection system, which is applied to a DC ore-bearing furnace as described above, and is characterized in that it includes: a server end, a data acquisition end, and a DC ore-bearing furnace equipment end, wherein the server end establishes communication connections with the data acquisition end and the DC ore-bearing furnace equipment end respectively; The server side includes: Data acquisition unit, used to obtain basic data of DC submerged arc furnace; A detection unit, used for preprocessing the basic data of the DC submerged arc furnace to obtain a basic data feature group of the DC submerged arc furnace; A state detection unit is used to establish an operation state detection model of the DC submerged arc furnace based on the basic data feature group of the DC submerged arc furnace, extract the features of the DC submerged arc furnace collected in real time, and input them into the operation state detection model of the DC submerged arc furnace to generate the DC submerged arc furnace state detection results. The DC submerged arc furnace state detection results include the DC submerged arc furnace state normal state, the DC submerged arc furnace state warning state, and the DC submerged arc furnace state fault state; The early warning unit is used to trigger the DC submerged arc furnace early warning process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state early warning state, and to trigger the DC submerged arc furnace fault detection process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state fault state; The data processing unit is used to obtain the historical operation data of the DC submerged arc furnace, build a DC submerged arc furnace knowledge base, retrieve the fault data corresponding to the state and fault state of the DC submerged arc furnace, obtain real-time fault data, extract data features of the real-time fault data, match the obtained real-time fault feature data in the DC submerged arc furnace knowledge base, obtain real-time fault feature association data, the real-time fault feature association data includes historical chemical substance detection data and historical electrical equipment adjustment and maintenance data, generate chemical detection substances based on the historical chemical substance detection data, use the chemical detection substances to detect the DC submerged arc furnace, obtain the chemical detection results of the DC submerged arc furnace, match the DC submerged arc furnace chemical detection results with the historical electrical equipment adjustment and maintenance data, obtain real-time electrical equipment adjustment and maintenance data, and send the real-time electrical equipment adjustment and maintenance data to the device end of the DC submerged arc furnace.

[0013] Furthermore, the data acquisition unit of the DC ore-forming furnace detection system of the present invention is also used for: The data acquisition unit initiates a data acquisition request to the sensor in the DC submerged arc furnace, requesting real-time data of the chemical substances in the furnace. The request message body carries the data acquisition unit address, sensor address, time range of the requested data, and requested data type. The requested data type includes temperature, pressure, and chemical component concentration. The sensor responds to the request and returns the original detection data of the DC submerged arc furnace.

[0014] Furthermore, the DC ore-forming furnace detection system and the detection unit of the present invention are also used for: The detection unit receives the original detection data from the sensor, removes duplicate, invalid or erroneous data, calibrates the cleaned data, and labels the temperature data, pressure data, and chemical composition concentration data according to the characteristics of the detection data.

[0015] Furthermore, the data processing unit of the DC ore-forming furnace detection system of the present invention is also used for: Collect chemical substance data in the furnace and identify the type, concentration and change trend of chemical substances in the furnace through a preset algorithm model; Receive and set thresholds, build an anomaly detection model, use the anomaly detection model to detect the type, concentration and change trend of chemical substances in the furnace, obtain abnormal detection results of chemical substances in the furnace, and alarm for data that exceeds the normal range, indicating existing equipment failures or safety hazards.

[0016] Based on historical and current data, predict future trends in the chemical state of the furnace and equipment operation.

[0017] Furthermore, the data processing unit of the DC ore-forming furnace detection system of the present invention is also used for: Retrieving historical chemical substance detection data from a DC ore-fired furnace knowledge base, preprocessing the acquired historical chemical substance detection data, and obtaining preprocessed historical chemical substance detection data; Use a preset algorithm model to analyze the pre-processed historical chemical substance detection data to identify the chemical substances in the DC submerged arc furnace and the characteristics of the DC submerged arc furnace; According to the chemical substances in the DC ore-forming furnace and the characteristics of the DC ore-forming furnace, matching is performed in the DC ore-forming furnace knowledge base to obtain the corresponding chemical detection substances.

[0018] Apply chemical test substances to the parts to be tested of the DC ore-fired furnace, including the furnace interior and the furnace wall of the DC ore-fired furnace, and collect reaction data between the chemical test substances and the chemical substances in the DC ore-fired furnace; Perform data feature extraction on the reaction data between the chemical test substance and the chemical substances in the DC submerged arc furnace to obtain the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace. Compare and analyze the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace with the preset chemical test standards to obtain the chemical test results; Collect the operating data after the device end of the DC submerged arc furnace executes the real-time electrical equipment adjustment and maintenance data to obtain the adjusted operating data, and compare the adjusted operating data with the preset DC submerged arc furnace standard data. If the adjusted operating data is not in the preset DC submerged arc furnace standard data, use the distillation algorithm to optimize the real-time electrical equipment adjustment and maintenance data, and send the optimized real-time electrical equipment adjustment and maintenance data to the DC submerged arc furnace device end.

[0019] Beneficial effects of the present invention: This invention enables real-time detection of chemical substances within a DC ore-fired furnace. Existing DC ore-fired furnaces cannot directly detect chemical substances within the furnace, resulting in an inability to accurately assess the chemical environment within the furnace. This invention, through a data acquisition unit and a detection unit, can collect and process real-time concentration data of chemical substances within the furnace, including temperature, pressure, and chemical component concentrations. Real-time detection of chemical substances within the furnace allows operators to promptly understand the furnace's conditions, providing a data foundation for subsequent fault prediction and early warning.

[0020] This invention improves the ability to predict and warn of equipment status. Existing DC ore-fired furnaces are unable to predict and warn of equipment status due to their inability to detect chemical substances within the furnace. This invention utilizes a status detection unit to establish and run an operating status detection model based on real-time data collected from the chemical substances within the furnace and other basic data. This model accurately identifies the operating status of the DC ore-fired furnace. When an abnormality or impending abnormality in the equipment status occurs, the warning unit promptly triggers a warning process, alerting the operator to take action and effectively preventing the occurrence or escalation of the fault.

[0021] This invention enhances the targeted nature of equipment fault handling. Existing DC submerged arc furnaces often lack targeted fault handling, resulting in inefficient handling. This invention utilizes a data processing unit to construct a DC submerged arc furnace knowledge base, storing historical fault data and adjustment and maintenance data. When a device fault occurs, the data processing unit accesses relevant knowledge and generates real-time electrical equipment adjustment and maintenance data, optimizing the device's operating status. This fault handling approach, based on both historical and real-time data, makes handling measures more targeted and effective, improving equipment reliability.

[0022] This invention improves production efficiency and reduces maintenance costs: Due to the inability to predict and warn equipment status, existing DC submerged arc furnaces often require shutdown for maintenance when a fault occurs, resulting in low production efficiency. This invention, through real-time monitoring and early warning mechanisms, enables timely action before equipment status anomalies occur, avoiding unnecessary downtime. Furthermore, through targeted troubleshooting measures, the frequency of equipment failures and maintenance costs are reduced, thereby improving overall production efficiency.

[0023] This invention introduces advanced data detection and analysis technology to the DC submerged arc furnace industry, driving technological innovation in this field. Through real-time monitoring and data analysis, it provides strong support for the optimized design and intelligent control of DC submerged arc furnaces, promoting the upgrading and development of the entire industry.

[0024] In summary, the present invention's benefits primarily include enabling real-time detection of chemical substances within DC submerged arc furnaces, improving the ability to predict and warn equipment status, enhancing the targeted and effective troubleshooting of equipment failures, improving production efficiency and reducing maintenance costs, and promoting technological innovation and industrial upgrading. These benefits not only address the problems of existing DC submerged arc furnaces but also provide more reliable and efficient technical support for their application in the metallurgical industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without any creative work.

[0026] Figure 1 A schematic diagram of the functional modules of a DC ore-fired furnace provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present invention and corresponding drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention. The technical solutions provided by each embodiment of the present invention are described in detail below in conjunction with the drawings.

[0028] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0029] The present invention realizes real-time detection of chemical substances in a DC ore-fired furnace through the following steps: Data Acquisition Unit: This unit initiates data acquisition requests to sensors within the DC submerged arc furnace, requesting real-time data on chemical substances within the furnace. The request message carries the data acquisition unit address, sensor address, the time range for the requested data, and the requested data type (including temperature, pressure, and chemical concentration). The sensors respond to the request and return the raw detection data from the DC submerged arc furnace.

[0030] Detection unit: The detection unit receives raw detection data from the sensor, removes duplicate, invalid or erroneous data, and calibrates the cleaned data to ensure data accuracy and consistency.

[0031] According to the characteristics of the test data, the temperature data, pressure data, and chemical composition concentration data are marked to facilitate subsequent data processing and analysis.

[0032] Data feature extraction: The detection unit extracts data features from the preprocessed data to obtain a basic data feature group of the DC submerged arc furnace, which includes chemical substance concentration features, etc.

[0033] Data processing unit: The data processing unit collects data on chemical substances within the furnace and uses a pre-set algorithm model to identify the type, concentration, and changing trends of these chemicals. An anomaly detection model is constructed to monitor the type, concentration, and changing trends of these chemicals within the furnace. When data exceeds the normal range, an alarm is generated, indicating an equipment failure or safety hazard. Through these steps, the present invention enables real-time detection of chemical substances within a DC submerged arc furnace, providing accurate data support for subsequent fault prediction, early warning, and resolution.

[0034] The early warning unit does not directly establish an operating status detection model based on the real-time collected chemical substance data in the furnace. Instead, the status detection unit is responsible for this task. The following is the logical reasoning process: Function of the status detection unit: The status detection unit is responsible for establishing and running the DC submerged arc furnace operation status detection model based on the DC submerged arc furnace basic data feature group (including electrical parameter characteristics, temperature characteristics, arc stability characteristics, chemical substance concentration characteristics, etc.).

[0035] The data acquisition unit collects real-time concentration data of chemical substances in the furnace and other basic data (such as electrical parameter data, temperature data, etc.). The detection unit preprocesses and extracts features from this data to obtain real-time DC submerged arc furnace characteristic data.

[0036] Establishing an operating status detection model: The status detection unit uses the DC submerged arc furnace's basic data feature set and corresponding operating status labels from the historical dataset to train an operating status detection model using machine learning or deep learning algorithms (such as support vector machines, random forests, and neural networks). This model can identify the DC submerged arc furnace's normal, warning, and fault states.

[0037] Real-time status detection: The DC submerged arc furnace characteristic data collected in real time is input into the operation status detection model. The model generates the DC submerged arc furnace status detection results based on the input characteristic data.

[0038] Triggering the early warning unit: When the status detection unit outputs a warning state or fault state, the early warning unit is triggered. Based on the warning or fault state, the early warning unit triggers the corresponding early warning process or fault detection process, such as sending a warning signal, recording warning information, and starting a troubleshooting program.

[0039] Therefore, the early warning unit does not directly establish an operating status detection model, but relies on the model established by the status detection unit through historical data and real-time data to identify the operating status of the DC submerged arc furnace and trigger the corresponding early warning or fault handling process when necessary.

[0040] In a first aspect, the present invention provides a DC ore-fired furnace, comprising: A data acquisition unit is used to acquire basic data of the DC submerged arc furnace, including electrical parameter data, temperature data, arc stability data, chemical substance concentration data, cooling system data, and equipment status data; Data acquisition unit for DC submerged arc furnace Scope of basic data acquisition: The data acquisition unit is used to obtain basic data of DC submerged arc furnaces, including but not limited to electrical parameter data (such as current, voltage, power factor, etc.), temperature data (such as furnace temperature, cooling water temperature, etc.), arc stability data (such as arc length, arc voltage fluctuation, etc.), chemical substance concentration data (such as furnace gas component concentration, melt component concentration, etc.), cooling system data (such as cooling water flow, pressure, etc.) and equipment status data (such as electrode position, furnace door opening and closing status, etc.).

[0041] To accurately acquire the above data, the data acquisition unit needs to interact with various sensor types. For example, electrical parameter data can be acquired through current transformers and voltage sensors; temperature data can be obtained through thermocouples or infrared thermometers; arc stability data can be obtained through arc sensors or image recognition technology; chemical concentration data can be obtained through gas analyzers or spectrometers; cooling system data can be obtained through flow meters and pressure sensors; and equipment status data can be obtained through limit switches and proximity switches. The sensor layout should fully consider the furnace environment and measurement requirements.

[0042] The data acquisition unit must establish a stable communication connection with the sensor, using either standard industrial communication protocols (such as Modbus and Profibus) or custom protocols for data transmission. Furthermore, to accommodate differences in data formats and transmission rates among different sensors, the data acquisition unit must include data format conversion and buffering mechanisms.

[0043] During the data acquisition process, sensor failures, communication interference, and other factors can lead to incomplete or erroneous data. Therefore, the data acquisition unit verifies data integrity through internal, pre-set data verification and error handling mechanisms. These mechanisms include redundant sensors and data comparison and verification to improve data accuracy.

[0044] As the data source for the detection system, the data acquisition unit must collaborate closely with other units (such as the detection unit, status detection unit, early warning unit, and data processing unit). For example, the detection unit must obtain preprocessed data from the data acquisition unit for feature extraction; the status detection unit must build and update the operating status detection model based on the data provided by the data acquisition unit; and the early warning unit and data processing unit must use the data provided by the data acquisition unit to make fault warnings and make adjustment and maintenance decisions.

[0045] The entire detection system forms a closed-loop feedback mechanism, from data acquisition to status detection, early warning, and processing. The data acquisition unit serves as the foundation, enabling data input for subsequent steps. The detection unit and status detection unit process and analyze the data, providing a basis for early warning and processing. The early warning unit and data processing unit then take appropriate measures based on the analysis results to ensure the safe and stable operation of the DC submerged arc furnace.

[0046] a detection unit, configured to preprocess the DC submerged arc furnace basic data to obtain preprocessed DC submerged arc furnace basic data, extract data features from the preprocessed DC submerged arc furnace basic data to obtain a DC submerged arc furnace basic data feature group, the DC submerged arc furnace basic data feature group including electrical parameter features, temperature features, arc stability features, chemical substance concentration features, cooling system features, and equipment status features; Data cleaning: The detection unit first receives basic data from the data acquisition unit. The data cleaning process includes removing duplicate data, handling missing values (such as through interpolation), and identifying and correcting outliers (such as setting thresholds based on statistical methods or domain knowledge).

[0047] Data calibration: To address the accuracy differences between different sensors, data calibration is performed to ensure that all data is compared under the same dimensions and standards. For example, temperature data is compensated for temperature and electrical parameter data is corrected for zero drift.

[0048] Data normalization / standardization: To eliminate the dimensional differences between different features and improve the efficiency of subsequent feature extraction and model training, the data is normalized or standardized.

[0049] Data normalization / standardization not only helps improve model training efficiency, but also enhances the model's generalization ability.

[0050] Based on domain knowledge and statistical analysis, features that significantly influence the operating status of the submerged arc furnace are selected from the preprocessed data. For example, current, voltage, and power factor are selected as electrical parameter features, furnace temperature and cooling water temperature are selected as temperature features, and arc length and arc voltage fluctuation are selected as arc stability features.

[0051] Feature dimensionality reduction: For high-dimensional data, methods such as principal component analysis (PCA) and linear discriminant analysis (LDA) are used to reduce feature dimensionality, reducing computational complexity while retaining key information.

[0052] Feature Engineering: New features need to be constructed based on specific application scenarios. For example, the arc power fluctuation rate can be calculated as a new feature of arc stability, or the heat flux distribution in the furnace can be calculated based on the temperature gradient.

[0053] The detection unit serves as a bridge between the data acquisition unit and subsequent analysis units (such as the status detection unit). By preprocessing and extracting features from raw data, it provides data support for subsequent status monitoring and fault warning.

[0054] After the data acquisition unit provides raw data, the detection unit first performs preprocessing. Next, it performs feature extraction, converting the high-dimensional, complex raw data into low-dimensional, easily analyzable feature vectors. These feature vectors are then input into the status detection unit, which builds and updates the operating status detection model, enabling real-time status monitoring and fault warnings for the submerged arc furnace. This entire process forms a closed-loop feedback mechanism from data acquisition to status monitoring.

[0055] a state detection unit, configured to establish an operating state detection model of the DC submerged arc furnace based on a basic data feature group of the DC submerged arc furnace, wherein the operating state detection model of the DC submerged arc furnace is configured to identify the operating state of the DC submerged arc furnace, extract features of the DC submerged arc furnace collected in real time, obtain real-time DC submerged arc furnace feature data, input the real-time DC submerged arc furnace feature data into the operating state detection model of the DC submerged arc furnace, and the operating state detection model of the DC submerged arc furnace generates a DC submerged arc furnace state detection result, wherein the DC submerged arc furnace state detection result includes a normal DC submerged arc furnace state, a warning DC submerged arc furnace state, and a fault DC submerged arc furnace state; Based on the DC submerged arc furnace's operating characteristics and monitoring requirements, select an appropriate machine learning or deep learning algorithm as the basis for the operating status detection model. For example, algorithms such as support vector machines (SVMs), random forests, and neural networks can be used.

[0056] Feature engineering: Before building the model, further feature selection, feature transformation, or feature construction is performed on the basic data feature group of the DC submerged arc furnace to extract the most useful information for operating status identification.

[0057] Model training: Using the basic data feature set of DC submerged arc furnaces and the corresponding operating status labels from the historical dataset, the selected algorithm is trained to obtain an operating status detection model. During the training process, methods such as cross-validation and grid search can be used to optimize model parameters and improve model performance.

[0058] The real-time collected DC submerged arc furnace data is preprocessed and feature extracted to obtain real-time DC submerged arc furnace feature data.

[0059] Model input and output: Real-time DC submerged arc furnace characteristic data is input into the operating status detection model. The model generates DC submerged arc furnace status detection results based on the input characteristic data. The detection results can include DC submerged arc furnace status, warning status, or fault status.

[0060] Result processing and feedback: The state detection results generated by the model are processed and analyzed, such as triggering corresponding alarm mechanisms or maintenance measures based on early warning or fault conditions. At the same time, the detection results are fed back to system users or operators so that they can promptly understand the operating status of the DC submerged arc furnace.

[0061] Normal state: When all operating parameters of the DC ore-fired furnace are within the normal range and there is no abnormal fluctuation or trend, it is judged to be in normal state.

[0062] Warning state: When certain operating parameters of the DC submerged arc furnace show slight anomalies or trend changes, but have not yet reached the failure threshold, it is judged to be in warning state. The warning state is intended to detect potential problems in advance so that preventive measures can be taken.

[0063] Fault status: When one or more operating parameters of the DC submerged arc furnace exceed the normal range, or show obvious abnormal fluctuations or trends, it is determined to be a fault state. Fault status requires immediate maintenance or shutdown measures to avoid more serious consequences.

[0064] The normal state is the ideal state for the operation of a DC ore-fired furnace and is also the goal pursued by the state detection system.

[0065] Early warning status is one of the important functions of the status detection system. By detecting potential problems in advance, the occurrence and impact of failures can be reduced.

[0066] Fault status is the most undesirable situation in the operation of a DC ore-fired furnace, but the status detection system can detect and alarm in time, which helps to reduce the losses caused by the fault.

[0067] As the core part of the DC submerged arc furnace operation status detection system, the status detection unit is responsible for establishing and running the operation status detection model to achieve real-time and accurate identification of the DC submerged arc furnace operation status.

[0068] First, the basic data of the DC ore-fired furnace is acquired and preprocessed through the data acquisition unit and the detection unit, and the characteristic data is extracted.

[0069] Then, the feature data is input into the running state detection model in the state detection unit, and the model generates a state detection result based on the input data.

[0070] Finally, the status detection results are processed and analyzed, and corresponding measures are taken or feedback is given to the user to ensure the safe and stable operation of the DC submerged arc furnace.

[0071] The entire process forms a closed-loop system from data acquisition to status detection, which provides strong guarantees for production efficiency and safety by accurately identifying the operating status of the DC submerged arc furnace in real time.

[0072] An early warning unit is used to trigger a DC submerged arc furnace early warning process when the DC submerged arc furnace status detection result is a DC submerged arc furnace status early warning state or a DC submerged arc furnace status fault state. When the DC submerged arc furnace status detection result is a DC submerged arc furnace status early warning state, the DC submerged arc furnace fault detection process is triggered when the DC submerged arc furnace status detection result is a DC submerged arc furnace status fault state. Based on the operating characteristics and safety standards of the DC blast furnace, specific conditions for triggering the early warning state are pre-set. These conditions may include abnormal fluctuations in certain key operating parameters, exceeding preset ranges, or trend changes.

[0073] Early warning process triggering: When the status detection unit outputs a warning status result for the DC submerged arc furnace, the early warning unit immediately triggers the early warning process. The early warning process includes sending an early warning signal to the operator or control system, recording relevant information such as the warning time and warning reason, and taking appropriate preventive measures based on the warning level.

[0074] Warning signal transmission: Warning signals can be transmitted to operators or control systems in a variety of ways, such as sound alarms, light indications, SMS notifications, system pop-ups, etc., to ensure that operators can receive warning information and respond in a timely manner.

[0075] The establishment of early warning conditions is fundamental to the early warning unit's operation. Appropriate early warning conditions ensure timely triggering of the early warning process when potential equipment issues arise. Triggering the early warning process is the unit's core function. By delivering early warning information promptly and accurately, it prompts operators to take preventative measures to avoid the occurrence or escalation of equipment failures. The method for delivering early warning signals should be selected based on actual usage scenarios and needs to ensure that warning information is conveyed quickly and accurately to relevant personnel.

[0076] Fault condition setting: Similarly, specific conditions that trigger a fault state are pre-set based on the DC blast furnace's operating characteristics and safety standards. These conditions are typically more stringent than warning conditions, such as when a key operating parameter significantly exceeds its normal range or when the equipment shows obvious signs of failure.

[0077] Fault detection process triggering: When the status detection unit outputs a faulty DC submerged arc furnace status detection result, the early warning unit immediately triggers the fault detection process. The fault detection process may include stopping equipment operation, sending a fault alarm signal, initiating a troubleshooting program, recording relevant information such as the fault time and fault type, and taking appropriate maintenance or emergency repair measures based on the fault level.

[0078] Fault alarm and handling: Fault alarm signals should be transmitted to operators or control systems in a prominent manner, such as emergency shutdown instructions, sound and light alarms, system emergency pop-up windows, etc., to ensure that operators can immediately realize the severity of the fault and take appropriate handling measures.

[0079] Setting fault conditions is key to ensuring timely equipment shutdown and triggering the fault handling process when a serious problem occurs. Triggering the fault detection process is the core function of the early warning unit when equipment fails. By quickly and accurately transmitting fault information, it can prompt operators to take timely maintenance or emergency repair measures, thereby reducing losses caused by the failure.

[0080] As part of the DC submerged arc furnace operation status detection and early warning system, the early warning unit is responsible for identifying the DC submerged arc furnace's early warning and fault status and triggering the corresponding processing flow.

[0081] First, the status detection unit detects the operating status of the DC ore-fired furnace in real time and outputs the status detection result.

[0082] Then, the early warning unit identifies the early warning or fault state of the DC ore-fired furnace according to the state detection result, and triggers the corresponding early warning or fault detection process.

[0083] Next, the early warning or fault detection process ensures the safe and stable operation of the equipment by sending alarm signals, recording relevant information, and taking preventive measures or maintenance measures.

[0084] Finally, through process connection, information sharing and process optimization, the efficiency and accuracy of early warning and fault handling will be continuously improved.

[0085] The entire process forms a closed-loop system from status detection to early warning or fault handling. By accurately identifying and processing the operating status of the DC submerged arc furnace in real time, it provides a strong guarantee for the safe and stable operation of the equipment.

[0086] The data processing unit is used to obtain the operation history data of the DC ore-heating furnace, build a knowledge base of the DC ore-heating furnace, retrieve the fault data corresponding to the state fault state of the DC ore-heating furnace, use the fault data corresponding to the state fault state of the DC ore-heating furnace as real-time fault data, extract data features from the real-time fault data to obtain real-time fault feature data, match the real-time fault feature data with the DC ore-heating furnace knowledge base to obtain real-time fault feature associated data, the real-time fault feature associated data includes historical chemical substance detection data and historical electrical equipment adjustment and maintenance data, generate chemical detection substances based on the historical chemical substance detection data, use the chemical detection substances to detect the DC ore-heating furnace, and obtain the DC ore-heating furnace. The chemical test results of the DC submerged arc furnace are matched with the historical electrical equipment adjustment and maintenance data to obtain real-time electrical equipment adjustment and maintenance data, and the real-time electrical equipment adjustment and maintenance data are sent to the device end of the DC submerged arc furnace. The operating data of the device end of the DC submerged arc furnace after executing the real-time electrical equipment adjustment and maintenance data is collected to obtain the adjusted operating data, and the operating data of the adjusted operating data is compared with the preset DC submerged arc furnace standard data. If the adjusted operating data is not in the preset DC submerged arc furnace standard data, the distillation algorithm is used to optimize the real-time electrical equipment adjustment and maintenance data, and the optimized real-time electrical equipment adjustment and maintenance data is sent to the DC submerged arc furnace device end.

[0087] The data processing unit is connected to the control system or sensor network of the DC ore-fired furnace to obtain the operating history data of the DC ore-fired furnace in real time or periodically, including but not limited to temperature, current, voltage, chemical content, equipment operating status, etc.

[0088] Knowledge base construction: Based on the acquired operation history data, the data processing unit constructs a DC submerged arc furnace knowledge base, which contains normal operation data, historical fault data, fault handling solutions, chemical substance detection data, electrical equipment adjustment and maintenance data, etc., providing data support for subsequent fault analysis and processing.

[0089] Fault data retrieval: When a DC ore-fired furnace fails, the data processing unit retrieves the fault data corresponding to the fault status from the knowledge base, including historical fault cases, fault phenomena, fault causes, etc.

[0090] Data feature extraction: Perform data feature extraction on the retrieved fault data, such as statistical analysis, trend analysis, and spectrum analysis, to obtain real-time fault feature data for subsequent fault matching and processing.

[0091] Fault feature matching: Real-time fault feature data is matched against the DC ore-fired furnace knowledge base to find similar historical fault cases and obtain real-time fault feature-related data, including historical chemical substance detection data and historical electrical equipment adjustment and maintenance data.

[0092] Chemical test substance generation: Based on historical chemical substance test data, a list or formula of chemical test substances is generated for chemical testing of DC submerged arc furnaces to verify or determine the cause of faults.

[0093] Obtaining chemical test results: Use the generated chemical test substances to test the DC submerged arc furnace to obtain the chemical test results of the DC submerged arc furnace, such as chemical substance content, reaction products, etc.

[0094] Fault signature matching bridges the gap between the current fault and historical failure cases, enabling the identification of similar fault resolution solutions or experiences. The generation and testing of chemical test substances is an important means of verifying the cause of a fault, especially when the fault is chemically related.

[0095] Match the chemical test results of the DC submerged arc furnace with the historical electrical equipment adjustment and maintenance data, find similar adjustment and maintenance cases, and obtain real-time electrical equipment adjustment and maintenance data.

[0096] Adjustment and maintenance data transmission: Real-time electrical equipment adjustment and maintenance data is sent to the equipment end of the DC submerged arc furnace to guide the equipment to make corresponding adjustments or maintenance.

[0097] Collection of post-adjustment operation data: Collect the operation data after the DC ore-fired furnace equipment performs real-time electrical equipment adjustment and maintenance data to obtain the post-adjustment operation data.

[0098] Data comparison and optimization: The adjusted operating data is compared with the preset DC submerged arc furnace standard data. If the adjusted operating data is not within the preset range, the distillation algorithm is used to optimize the real-time electrical equipment adjustment and maintenance data to improve the accuracy and effectiveness of the adjustment and maintenance.

[0099] Optimized data transmission: The optimized real-time electrical equipment adjustment and maintenance data is sent to the DC submerged arc furnace equipment end to guide the equipment to be readjusted or maintained.

[0100] Matching adjustment and maintenance data is key to achieving targeted adjustments or maintenance measures. This matching process helps identify similar adjustment and maintenance experiences or solutions. The transmission and execution of adjustment and maintenance data are the actual operational steps in troubleshooting, ensuring data accuracy and timely execution. The collection and comparison of post-adjustment operational data is crucial for verifying the effectiveness of adjustment and maintenance measures, and through this comparison, the effectiveness of adjustment and maintenance measures can be assessed.

[0101] The application of distillation algorithm is an effective means to optimize adjustment and maintenance data, which can improve the accuracy and efficiency of adjustment and maintenance.

[0102] As the core part of the DC submerged arc furnace data processing and fault handling system, the data processing unit is responsible for acquiring and processing the DC submerged arc furnace's operating history data, building a knowledge base, and retrieving relevant data for fault analysis, detection, and processing in the event of a fault, providing strong support for the stable operation of the DC submerged arc furnace.

[0103] Firstly, the data processing unit obtains the operation history data of the DC ore-fired furnace in real time or periodically through a data acquisition method and builds a knowledge base.

[0104] Then, in a fault state, the data processing unit retrieves fault data from the knowledge base and performs data feature extraction to obtain real-time fault feature data.

[0105] Next, the data processing unit matches the real-time fault feature data in the knowledge base to obtain real-time fault feature association data, and generates chemical detection substances based on historical chemical substance detection data to detect the DC ore-fired furnace.

[0106] Subsequently, the data processing unit matches the chemical test results with the historical electrical equipment adjustment and maintenance data to obtain real-time electrical equipment adjustment and maintenance data, and sends it to the DC submerged arc furnace equipment end for execution.

[0107] Finally, the data processing unit collects the adjusted operating data and compares it with the preset DC electric arc furnace standard data. If it does not meet expectations, the distillation algorithm is used to optimize the real-time electrical equipment adjustment and maintenance data and send it to the device end for re-execution.

[0108] The entire process forms a closed-loop system from data acquisition to fault handling. By accurately processing and analyzing DC submerged arc furnace operating data in real time, this system provides a strong guarantee for stable equipment operation. Furthermore, by continuously optimizing and adjusting maintenance data, the efficiency and accuracy of fault handling are improved.

[0109] Specifically, the data acquisition unit of the DC ore-fired furnace of the present invention is further used for: The data acquisition unit initiates a data acquisition request to the sensor in the DC submerged arc furnace, requesting real-time data of the chemical substances in the furnace. The request message body carries the data acquisition unit address, sensor address, time range of the requested data, and requested data type. The requested data type includes temperature, pressure, and chemical component concentration. The sensor responds to the request and returns the original detection data of the DC submerged arc furnace.

[0110] Data acquisition unit address: used to identify the data acquisition unit that initiates the request so that the sensor or other parts of the system can identify the source of the request.

[0111] Sensor address: Clearly specifies the sensor that needs to respond to the request so that data can be obtained from the correct sensor.

[0112] Time range of requested data: Specify the time period of data to be collected, which can be real-time data or a time range of historical data.

[0113] Request data type: Specify the data type that needs to be collected, including but not limited to temperature, pressure, chemical composition concentration, etc., to meet data requirements in different scenarios.

[0114] After receiving the data collection request, the sensor performs the corresponding data collection task according to the information in the request message body.

[0115] After the sensor collects the data, it packages the data into a response message and returns it to the data acquisition unit that initiated the request.

[0116] The returned raw test data includes but is not limited to temperature values, pressure values, chemical component concentration values, etc., which are determined by the requested data type.

[0117] Specifically, the DC ore-fired furnace detection unit of the present invention is further used for: The detection unit receives the original detection data from the sensor, removes duplicate, invalid or erroneous data, calibrates the cleaned data, and labels the temperature data, pressure data, and chemical composition concentration data according to the characteristics of the detection data.

[0118] The detection unit is responsible for receiving raw detection data from sensors in real time or periodically. The data types received include but are not limited to temperature data, pressure data, chemical composition concentration data, etc.

[0119] The detection unit has a built-in data cleaning module that removes duplicate, invalid, and malformed data from the raw detection data it receives. Duplicate data can be caused by repeated sensor transmissions or data transmission errors; invalid data can be caused by sensor failure or environmental interference; and malformed data can be caused by encoding or decoding errors during data transmission. The data cleaning module automatically identifies and removes this bad data using pre-set algorithms or rules, ensuring the accuracy of subsequent processing.

[0120] The detection unit is equipped with a data calibration module, which is responsible for calibrating the cleaned data.

[0121] The calibration process includes linear calibration, nonlinear calibration, temperature compensation, pressure compensation, etc., which is determined according to the characteristics of the sensor and the nature of the detection data.

[0122] The data calibration module processes the cleaned data through preset calibration parameters or algorithms to ensure the accuracy and consistency of the data.

[0123] The detection unit has a built-in data annotation module, which is responsible for annotating the calibrated data.

[0124] The annotation content includes but is not limited to the data type (such as temperature, pressure, chemical component concentration, etc.), collection time, sensor number, etc.

[0125] The data annotation module automatically adds annotation information to the calibrated data through preset annotation rules or algorithms to facilitate subsequent data analysis and processing.

[0126] Specifically, the DC ore-fired furnace and data processing unit of the present invention are further used for: Collect chemical substance data in the furnace and identify the type, concentration and change trend of chemical substances in the furnace through a preset algorithm model; Receive and set thresholds, build an anomaly detection model, use the anomaly detection model to detect the type, concentration and change trend of chemical substances in the furnace, obtain abnormal detection results of chemical substances in the furnace, and alarm for data that exceeds the normal range, indicating existing equipment failures or safety hazards.

[0127] Based on historical and current data, predict future trends in the chemical state of the furnace and equipment operation.

[0128] The data processing unit is responsible for collecting the raw data of the chemical substances in the furnace in real time or periodically, including but not limited to temperature, pressure, chemical component concentration, etc.

[0129] The data processing unit has a built-in preset algorithm model, which is based on machine learning, deep learning or statistical analysis methods and can identify the types, concentrations and change trends of chemical substances in the furnace.

[0130] The algorithm model learns the characteristic patterns of chemical substances by training historical data, so as to accurately identify newly collected data.

[0131] By identifying the types, concentrations, and changing trends of chemical substances, we can timely understand the chemical environment inside the furnace, providing a basis for subsequent abnormality detection and prediction.

[0132] The data processing unit is provided with a threshold setting module, which is responsible for receiving and setting the normal range thresholds of the types, concentrations and change trends of the chemical substances in the furnace.

[0133] Based on the set thresholds, the data processing unit constructs an anomaly detection model that can monitor the status of the chemical substances in the furnace in real time and compare it with the thresholds.

[0134] When it is detected that the state of the chemical substances in the furnace is outside the normal range, the anomaly detection model triggers the alarm mechanism, indicating the existence of equipment failure or safety hazards.

[0135] The data processing unit incorporates a built-in prediction model. Based on historical and current data, this model uses time series analysis, machine learning, or deep learning to predict future trends in the chemical state within the furnace and equipment operation. This prediction model considers multiple factors, such as changes in furnace temperature, pressure, and chemical concentration, as well as equipment operating time and load, to enhance prediction accuracy. Prediction results can be displayed in charts and reports, enabling operators to intuitively understand future trends in the chemical state within the furnace and equipment operation.

[0136] The data processing unit first collects raw data on the chemical substances within the furnace. It then uses a pre-set algorithm model to identify the chemical types, concentrations, and changing trends. Based on pre-set thresholds, the data processing unit constructs an anomaly detection model to monitor the chemical status within the furnace in real time, promptly detecting and alerting users to abnormal conditions. Using a predictive model, the data processing unit uses historical and current data to predict future trends in the chemical status within the furnace and equipment operation, providing decision support for operators.

[0137] In summary, the data processing unit and data processing method of a DC ore-fired furnace described in the present invention realize comprehensive monitoring of the chemical substance status and equipment operation status in the DC ore-fired furnace through the steps of collecting, identifying, detecting and predicting chemical substance data.

[0138] Specifically, the DC ore-fired furnace and data processing unit of the present invention are further used for: Retrieving historical chemical substance detection data from a DC ore-fired furnace knowledge base, preprocessing the acquired historical chemical substance detection data, and obtaining preprocessed historical chemical substance detection data; Use a preset algorithm model to analyze the pre-processed historical chemical substance detection data to identify the chemical substances in the DC submerged arc furnace and the characteristics of the DC submerged arc furnace; According to the chemical substances in the DC ore-forming furnace and the characteristics of the DC ore-forming furnace, matching is performed in the DC ore-forming furnace knowledge base to obtain the corresponding chemical detection substances.

[0139] Apply chemical test substances to the parts to be tested of the DC ore-fired furnace, including the furnace interior and the furnace wall of the DC ore-fired furnace, and collect reaction data between the chemical test substances and the chemical substances in the DC ore-fired furnace; Data features are extracted from the reaction data between the chemical detection substances and the chemical substances in the DC submerged arc furnace to obtain the types of chemical substances in the DC submerged arc furnace, the concentrations of chemical substances in the DC submerged arc furnace and the color data of chemical substances in the DC submerged arc furnace. The types of chemical substances in the DC submerged arc furnace, the concentrations of chemical substances in the DC submerged arc furnace and the color data of chemical substances in the DC submerged arc furnace are compared and analyzed with the preset chemical detection standards to obtain the chemical detection results.

[0140] The data processing unit retrieves historical chemical substance test data from the DC submerged arc furnace knowledge base. This knowledge base includes past test records, chemical substance types, concentrations, colors, and reaction characteristics. The retrieved historical chemical substance test data undergoes preprocessing operations such as cleaning, denoising, and normalization to eliminate outliers, missing values, and noise, resulting in preprocessed historical chemical substance test data.

[0141] The data processing unit uses a preset algorithm model, such as a machine learning algorithm, a statistical analysis method, or a pattern recognition technology, to perform data analysis on the pre-processed historical chemical substance detection data.

[0142] Through data analysis, the types of chemical substances in the DC ore-fired furnace, their concentration distribution, and the operating characteristics of the DC ore-fired furnace, such as temperature changes and pressure fluctuations, can be identified.

[0143] Based on the identified chemical substances and characteristics of the DC submerged arc furnace, the data processing unit matches the DC submerged arc furnace knowledge base and selects or prepares the corresponding chemical test substances. The selection or preparation of chemical test substances requires consideration of factors such as their reactivity with the target chemical substances, detection sensitivity, and safety.

[0144] The matching and preparation of chemical test substances are key steps in chemical testing, directly impacting the accuracy and reliability of the test. The matching process within the knowledge base needs to be based on accurate chemical substance and characteristic information to ensure that the selected or prepared chemical test substances are suitable for use.

[0145] The prepared chemical detection substances are applied to the parts to be detected of the DC ore-fired furnace, including the furnace interior and furnace walls.

[0146] The data processing unit collects the reaction data of the chemical detection substance and the chemical substances in the DC ore-fired furnace in real time or periodically, including the type, concentration, color and reaction time of the reaction product.

[0147] The data processing unit includes a built-in data feature extraction module, which extracts features from the collected data on the reaction between chemical test substances and the chemical substances within the DC submerged arc furnace. This feature extraction process provides data on the type, concentration, and color of the chemical substances within the DC submerged arc furnace. This extracted data is then compared and analyzed with pre-set chemical test standards to determine chemical test results, including any abnormalities in the chemical substances, any concentration exceeding the standard, and any color changes.

[0148] Data feature extraction is a key step in chemical testing, requiring the extraction of useful information from complex reaction data. Chemical testing standards must be established based on factors such as industry regulations, equipment requirements, and safety standards to ensure the accuracy and reliability of test results.

[0149] The data processing unit first retrieves historical chemical substance detection data and performs preprocessing to provide accurate baseline data for subsequent data analysis. Preprocessed data is analyzed using a preset algorithm model to accurately identify the chemical substances and characteristics within the DC submerged arc furnace. Based on the identified chemical substances and characteristics, a matching is performed within the knowledge base, and suitable chemical test substances are selected or prepared. The chemical test substances are applied to the areas to be tested, and reaction data is collected in real time or periodically to ensure data timeliness and accuracy. Feature extraction is performed on the reaction data and compared with preset standards to produce accurate chemical test results.

[0150] In summary, the data processing unit and chemical detection method for a DC electric arc furnace described in the present invention realize comprehensive detection of chemical substances in a DC electric arc furnace through the steps of historical data retrieval, data analysis, preparation of chemical detection substances, application and reaction data collection, and data feature extraction.

[0151] In a second aspect, the present invention provides a DC ore-bearing furnace detection system, which is applied to a DC ore-bearing furnace as described above, comprising: a server end, a data acquisition end, and a DC ore-bearing furnace equipment end, wherein the server end establishes communication connections with the data acquisition end and the DC ore-bearing furnace equipment end respectively; The server side includes: A data acquisition unit acquires basic data of a DC submerged arc furnace, including electrical parameter data, temperature data, arc stability data, chemical substance concentration data, cooling system data, and equipment status data; The detection unit preprocesses the basic data of the DC submerged arc furnace to obtain preprocessed basic data of the DC submerged arc furnace, and extracts data features from the preprocessed basic data of the DC submerged arc furnace to obtain a basic data feature group of the DC submerged arc furnace. The basic data feature group of the DC submerged arc furnace includes electrical parameter features, temperature features, arc stability features, chemical substance concentration features, cooling system features, and equipment status features. A state detection unit establishes an operation state detection model of the DC submerged arc furnace based on a basic data feature group of the DC submerged arc furnace. The operation state detection model of the DC submerged arc furnace is used to identify the operation state of the DC submerged arc furnace, extract features of the DC submerged arc furnace collected in real time, obtain real-time DC submerged arc furnace feature data, input the real-time DC submerged arc furnace feature data into the operation state detection model of the DC submerged arc furnace, and the operation state detection model of the DC submerged arc furnace generates a DC submerged arc furnace state detection result. The DC submerged arc furnace state detection result includes a normal state of the DC submerged arc furnace, a warning state of the DC submerged arc furnace, and a fault state of the DC submerged arc furnace; The early warning unit triggers the DC submerged arc furnace early warning process when the DC submerged arc furnace status detection result is the DC submerged arc furnace status early warning state or the DC submerged arc furnace status fault state; and triggers the DC submerged arc furnace fault detection process when the DC submerged arc furnace status detection result is the DC submerged arc furnace status fault state; The data processing unit obtains the operation history data of the DC ore-heating furnace, builds the knowledge base of the DC ore-heating furnace, retrieves the fault data corresponding to the state fault state of the DC ore-heating furnace, uses the fault data corresponding to the state fault state of the DC ore-heating furnace as the real-time fault data, extracts data features from the real-time fault data, obtains real-time fault feature data, matches the real-time fault feature data with the DC ore-heating furnace knowledge base, obtains real-time fault feature correlation data, the real-time fault feature correlation data includes historical chemical substance detection data and historical electrical equipment adjustment and maintenance data, generates chemical detection substances based on the historical chemical substance detection data, uses the chemical detection substances to detect the DC ore-heating furnace, and obtains the DC ore-heating furnace. Chemical test results of the DC submerged arc furnace are matched with historical electrical equipment adjustment and maintenance data to obtain real-time electrical equipment adjustment and maintenance data, and the real-time electrical equipment adjustment and maintenance data are sent to the device end of the DC submerged arc furnace. The operating data after the device end of the DC submerged arc furnace executes the real-time electrical equipment adjustment and maintenance data is collected to obtain the adjusted operating data. The operating data of the adjusted operating data is compared with the preset DC submerged arc furnace standard data. If the adjusted operating data is not in the preset DC submerged arc furnace standard data, the distillation algorithm is used to optimize the real-time electrical equipment adjustment and maintenance data, and the optimized real-time electrical equipment adjustment and maintenance data is sent to the device end of the DC submerged arc furnace. Specifically, the data acquisition unit of the DC ore-forming furnace detection system of the present invention is further used for: The data acquisition unit initiates a data acquisition request to the sensor in the DC submerged arc furnace, requesting real-time data of the chemical substances in the furnace. The request message body carries the data acquisition unit address, sensor address, time range of the requested data, and requested data type. The requested data type includes temperature, pressure, and chemical component concentration. The sensor responds to the request and returns the original detection data of the DC submerged arc furnace.

[0152] Specifically, the DC ore-forming furnace detection system and detection unit described in the present invention are further used to: The detection unit receives the original detection data from the sensor, removes duplicate, invalid or erroneous data, calibrates the cleaned data, and labels the temperature data, pressure data, and chemical composition concentration data according to the characteristics of the detection data.

[0153] Specifically, the DC ore-bearing furnace detection system and data processing unit of the present invention are further used for: Collect chemical substance data in the furnace and identify the type, concentration and change trend of chemical substances in the furnace through a preset algorithm model; Receive and set thresholds, build an anomaly detection model, use the anomaly detection model to detect the type, concentration and change trend of chemical substances in the furnace, obtain abnormal detection results of chemical substances in the furnace, and alarm for data that exceeds the normal range, indicating existing equipment failures or safety hazards.

[0154] Based on historical and current data, predict future trends in the chemical state of the furnace and equipment operation.

[0155] Specifically, the DC ore-bearing furnace detection system and data processing unit of the present invention are further used for: Retrieving historical chemical substance detection data from a DC ore-fired furnace knowledge base, preprocessing the acquired historical chemical substance detection data, and obtaining preprocessed historical chemical substance detection data; Use a preset algorithm model to analyze the pre-processed historical chemical substance detection data to identify the chemical substances in the DC submerged arc furnace and the characteristics of the DC submerged arc furnace; According to the chemical substances in the DC ore-forming furnace and the characteristics of the DC ore-forming furnace, matching is performed in the DC ore-forming furnace knowledge base to obtain the corresponding chemical detection substances.

[0156] Apply chemical test substances to the parts to be tested of the DC ore-fired furnace, including the furnace interior and the furnace wall of the DC ore-fired furnace, and collect reaction data between the chemical test substances and the chemical substances in the DC ore-fired furnace; Data features are extracted from the reaction data between the chemical detection substances and the chemical substances in the DC submerged arc furnace to obtain the types of chemical substances in the DC submerged arc furnace, the concentrations of chemical substances in the DC submerged arc furnace and the color data of chemical substances in the DC submerged arc furnace. The types of chemical substances in the DC submerged arc furnace, the concentrations of chemical substances in the DC submerged arc furnace and the color data of chemical substances in the DC submerged arc furnace are compared and analyzed with the preset chemical detection standards to obtain the chemical detection results.

[0157] The technical solution of the present invention effectively solves the problem that the existing DC ore-fired furnace cannot detect the chemical substances in the furnace during actual use, and thus cannot timely predict and warn the status of each device.

[0158] Existing DC submerged arc furnaces lack the means to directly monitor the chemical substances within the furnace in real time, making it impossible to accurately understand the chemical environment within the furnace. This inability to monitor the chemical substances within the furnace makes it impossible to predict the equipment's operating status based on chemical changes, nor can it provide timely warnings when the equipment's status deviates.

[0159] The technical solution of the present invention is as follows: The data acquisition unit initiates data collection requests from sensors within the DC submerged arc furnace, acquiring various basic data, including chemical concentrations. The detection unit preprocesses and extracts features from this acquired basic data, including chemical concentration characteristics. Through data cleaning, calibration, and feature extraction, noise and invalid data are removed, key features are extracted, and high-quality data is provided for subsequent status detection. Based on the extracted feature data, the status detection unit establishes and runs an operating status detection model to identify the operating status of the DC submerged arc furnace. This model, using machine learning or deep learning algorithms, automatically learns the relationship between characteristics such as chemical concentration within the furnace and the equipment's operating status, enabling accurate status identification.

[0160] When the status detection unit identifies a warning or fault condition, it triggers the corresponding warning or fault detection process. By monitoring key indicators such as chemical concentrations within the furnace in real time, the warning unit can issue a timely warning before the equipment status becomes abnormal, providing operators with sufficient time to take action and prevent the occurrence or escalation of the fault.

[0161] A knowledge base for DC submerged arc furnaces was built to store historical chemical substance test data. When equipment malfunctions, the fault data is retrieved, chemical test substances are generated, and the chemicals within the furnace are tested. Based on the test results and historical data, real-time electrical equipment adjustment and maintenance data is generated to optimize equipment operation. This knowledge base provides rich historical data support for data analysis. The use of chemical test substances allows for direct detection of the type and concentration of chemicals within the furnace, enabling accurate assessment of equipment status. The resulting adjustment and maintenance data can be used to address equipment issues, improving equipment reliability and stability.

[0162] The technical solution of this invention achieves real-time detection of chemical substances within a DC submerged arc furnace and accurate prediction of the equipment's status through the close collaboration of multiple steps, including data acquisition, detection, status identification, early warning, and data processing. This solution not only solves the problem of existing DC submerged arc furnaces being unable to detect chemical substances within the furnace, but also ensures safe and stable operation of the equipment through early warning and adjustment of maintenance procedures, effectively improving production efficiency and reducing maintenance costs.

[0163] The model, construction steps, uses and model parameters involved in the present invention are as follows: Running status detection model: Build steps: Data collection: Collect basic data of DC submerged arc furnace, including electrical parameters, temperature, arc stability, chemical concentration, cooling system data and equipment status data.

[0164] Data preprocessing: Clean, calibrate and extract features from the collected data to form a feature data group.

[0165] Model selection: Choose an appropriate machine learning or deep learning algorithm (such as support vector machine, random forest, neural network, etc.) as the basis of the model.

[0166] Feature engineering: further perform feature selection, transformation, or construction on the feature data set to extract the most useful information for operating status identification.

[0167] Model training: Use the feature data and corresponding operating status labels in the historical dataset to train the model, and optimize the model parameters through methods such as cross-validation and grid search.

[0168] Use of the operating status detection model: It is used to identify the operating status of the DC ore-fired furnace in real time, including normal status, warning status and fault status.

[0169] Model parameters for the running status detection model: Specific parameters of the machine learning or deep learning algorithm (such as the number of neural network layers, number of nodes, learning rate, etc.). Parameters used in feature selection, transformation, or construction.

[0170] Anomaly detection model building steps: Data collection: Collect data on the types, concentrations and changing trends of chemical substances in the furnace.

[0171] Threshold setting: Based on historical data and actual operating experience, set the normal range threshold of the chemical substance data in the furnace.

[0172] Anomaly detection model construction: Build an anomaly detection model based on the set threshold.

[0173] Use of anomaly detection model: Used to monitor the status of chemical substances in the furnace in real time and compare it with the threshold value. When the data exceeds the normal range, an alarm is triggered to indicate the existence of equipment failure or safety hazards.

[0174] Model parameters of the anomaly detection model: normal range thresholds for furnace chemical substance data.

[0175] Prediction model building steps: Data collection: Collect historical and current data, including changes in furnace temperature, pressure, chemical composition concentration, equipment operating time, load, etc.

[0176] Model selection: Choose an appropriate time series analysis, machine learning, or deep learning algorithm as the basis for the forecasting model.

[0177] Model training: Use historical and current data to train the forecasting model and optimize the model parameters to improve the forecast accuracy.

[0178] Use of the prediction model: It is used to predict the future trends of the chemical state in the furnace and the operation of the equipment, and provide decision support for operators.

[0179] Forecast model parameters: Specific parameters of time series analysis, machine learning, or deep learning algorithms.

[0180] Historical and current data used to train the model.

[0181] When using the distillation algorithm to optimize real-time electrical equipment adjustment and maintenance data, the specific steps can be summarized as follows: Determine the goal of knowledge distillation: Clarify the goal of the distillation algorithm, which is to improve the accuracy and effectiveness of real-time electrical equipment adjustment and maintenance data, making it closer to the preset DC submerged arc furnace standard data.

[0182] Build a teacher model: Select or build a high-performance teacher model. In this scenario, the teacher model can be based on historically optimal electrical equipment adjustment and maintenance data and the operating status of the DC submerged arc furnace. It can accurately predict and adjust equipment status. The teacher model should have high accuracy and generalization capabilities, and serve as a reference standard for optimizing real-time electrical equipment adjustment and maintenance data.

[0183] Prepare the student model: The student model is the real-time electrical equipment adjustment and maintenance data model that needs to be optimized.

[0184] Initialize the parameters of the student model, which are derived from preliminary tuning maintenance data or a simple rule-based model.

[0185] Data preparation: Prepare training data for distillation, including historical optimal electrical equipment adjustment and maintenance data, corresponding DC submerged arc furnace operating status data, and current real-time electrical equipment adjustment and maintenance data.

[0186] Knowledge distillation process: Prediction stage: First, let the teacher model predict the training data and obtain the prediction results of the teacher model.

[0187] Learning phase: The student model is then asked to make predictions on the training data and its predictions are compared with those of the teacher model.

[0188] Loss calculation: Calculate the difference between the student model prediction and the teacher model prediction, that is, the distillation loss. At the same time, you can also calculate the standard loss in combination with the true label of the original data.

[0189] Backpropagation and optimization: Through the backpropagation algorithm, the parameters of the student model are jointly optimized according to the distillation loss and the standard loss.

[0190] Model evaluation and tuning: During the distillation process, the performance of the student model is regularly evaluated to ensure that it is optimizing in the right direction.

[0191] Based on the evaluation results, adjust the hyperparameters in the distillation process, such as the weight of distillation loss and standard loss, learning rate, etc.

[0192] Outputting the optimized model: When the performance of the student model reaches expectations, the distillation process is stopped and the optimized real-time electrical equipment adjustment and maintenance data model is output.

[0193] Deployment and Application: The optimized model is deployed to the DC submerged arc furnace equipment to guide actual electrical equipment adjustment and maintenance work.

[0194] Through these steps, the distillation algorithm can effectively transfer the knowledge of the teacher model to the student model, improve the accuracy and effectiveness of real-time electrical equipment adjustment and maintenance data, and thus optimize the operating status of the DC submerged arc furnace.

Claims

1. A DC ore-fired furnace, characterized in that: include: Data acquisition unit, used to obtain basic data of DC submerged arc furnace; A detection unit, used for preprocessing the basic data of the DC submerged arc furnace to obtain a basic data feature group of the DC submerged arc furnace; A state detection unit is used to establish an operation state detection model of the DC submerged arc furnace based on the basic data feature group of the DC submerged arc furnace, extract the features of the DC submerged arc furnace collected in real time, and input them into the operation state detection model of the DC submerged arc furnace to generate the DC submerged arc furnace state detection results. The DC submerged arc furnace state detection results include the DC submerged arc furnace state normal state, the DC submerged arc furnace state warning state, and the DC submerged arc furnace state fault state; The early warning unit is used to trigger the DC submerged arc furnace early warning process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state early warning state, and to trigger the DC submerged arc furnace fault detection process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state fault state; The data processing unit is used to obtain the historical operation data of the DC submerged arc furnace, build a DC submerged arc furnace knowledge base, retrieve the fault data corresponding to the state and fault state of the DC submerged arc furnace, obtain real-time fault data, extract data features of the real-time fault data, match the obtained real-time fault feature data in the DC submerged arc furnace knowledge base, obtain real-time fault feature association data, the real-time fault feature association data includes historical chemical substance detection data and historical electrical equipment adjustment and maintenance data, generate chemical detection substances based on the historical chemical substance detection data, use the chemical detection substances to detect the DC submerged arc furnace, obtain the chemical detection results of the DC submerged arc furnace, match the DC submerged arc furnace chemical detection results with the historical electrical equipment adjustment and maintenance data, obtain real-time electrical equipment adjustment and maintenance data, and send the real-time electrical equipment adjustment and maintenance data to the device end of the DC submerged arc furnace.

2. A DC ore-fired furnace according to claim 1, characterized in that: The data acquisition unit is also used to: The data acquisition unit initiates a data acquisition request to the sensor in the DC submerged arc furnace, requesting real-time data of the chemical substances in the furnace. The request message body carries the data acquisition unit address, sensor address, time range of the requested data, and requested data type. The requested data type includes temperature, pressure, and chemical component concentration. The sensor responds to the request and returns the original detection data of the DC submerged arc furnace.

3. A DC ore-fired furnace according to claim 1, characterized in that: The detection unit is also used to: The detection unit receives the original detection data from the sensor, removes duplicate, invalid or erroneous data, calibrates the cleaned data, and labels the temperature data, pressure data, and chemical composition concentration data according to the characteristics of the detection data.

4. A DC ore-fired furnace according to claim 1, characterized in that: The data processing unit is also used to: Collect chemical substance data in the furnace and identify the type, concentration and change trend of chemical substances in the furnace through a preset algorithm model; Receive and set thresholds, build an anomaly detection model, use the anomaly detection model to detect the type, concentration and change trend of chemical substances in the furnace, obtain abnormal detection results of chemical substances in the furnace, and alarm for data that exceeds the normal range, indicating existing equipment failures or safety hazards. Based on historical and current data, predict future trends in the chemical state of the furnace and equipment operation.

5. The DC ore-fired furnace according to claim 1, characterized in that: The data processing unit is also used to: Retrieving historical chemical substance detection data from a DC ore-fired furnace knowledge base, preprocessing the acquired historical chemical substance detection data, and obtaining preprocessed historical chemical substance detection data; Use a preset algorithm model to analyze the pre-processed historical chemical substance detection data to identify the chemical substances in the DC submerged arc furnace and the characteristics of the DC submerged arc furnace; According to the chemical substances in the DC ore-forming furnace and the characteristics of the DC ore-forming furnace, matching is performed in the DC ore-forming furnace knowledge base to obtain the corresponding chemical detection substances. Apply chemical test substances to the parts to be tested of the DC ore-fired furnace, including the furnace interior and the furnace wall of the DC ore-fired furnace, and collect reaction data between the chemical test substances and the chemical substances in the DC ore-fired furnace; Perform data feature extraction on the reaction data between the chemical test substance and the chemical substances in the DC submerged arc furnace to obtain the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace. Compare and analyze the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace with the preset chemical test standards to obtain the chemical test results; Collect the operating data after the device end of the DC submerged arc furnace executes the real-time electrical equipment adjustment and maintenance data to obtain the adjusted operating data, and compare the adjusted operating data with the preset DC submerged arc furnace standard data. If the adjusted operating data is not in the preset DC submerged arc furnace standard data, use the distillation algorithm to optimize the real-time electrical equipment adjustment and maintenance data, and send the optimized real-time electrical equipment adjustment and maintenance data to the DC submerged arc furnace device end.

6. A DC ore-bearing furnace detection system, applied to a DC ore-bearing furnace according to any one of claims 1 to 5, characterized in that: include: The server side, the data acquisition side and the DC submerged arc furnace equipment side, the server side establishes communication connections with the data acquisition side and the DC submerged arc furnace equipment side respectively; The server side includes: Data acquisition unit, used to obtain basic data of DC submerged arc furnace; A detection unit, used for preprocessing the basic data of the DC submerged arc furnace to obtain a basic data feature group of the DC submerged arc furnace; A state detection unit is used to establish an operation state detection model of the DC submerged arc furnace based on the basic data feature group of the DC submerged arc furnace, extract the features of the DC submerged arc furnace collected in real time, and input them into the operation state detection model of the DC submerged arc furnace to generate the DC submerged arc furnace state detection results. The DC submerged arc furnace state detection results include the DC submerged arc furnace state normal state, the DC submerged arc furnace state warning state, and the DC submerged arc furnace state fault state; The early warning unit is used to trigger the DC submerged arc furnace early warning process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state early warning state, and to trigger the DC submerged arc furnace fault detection process when the DC submerged arc furnace state detection result is the DC submerged arc furnace state fault state; The data processing unit is used to obtain the historical operation data of the DC submerged arc furnace, build a DC submerged arc furnace knowledge base, retrieve the fault data corresponding to the state and fault state of the DC submerged arc furnace, obtain real-time fault data, extract data features of the real-time fault data, match the obtained real-time fault feature data in the DC submerged arc furnace knowledge base, obtain real-time fault feature association data, the real-time fault feature association data includes historical chemical substance detection data and historical electrical equipment adjustment and maintenance data, generate chemical detection substances based on the historical chemical substance detection data, use the chemical detection substances to detect the DC submerged arc furnace, obtain the chemical detection results of the DC submerged arc furnace, match the DC submerged arc furnace chemical detection results with the historical electrical equipment adjustment and maintenance data, obtain real-time electrical equipment adjustment and maintenance data, and send the real-time electrical equipment adjustment and maintenance data to the device end of the DC submerged arc furnace.

7. A DC ore furnace detection system according to claim 6, characterized in that: The data acquisition unit is also used to: The data acquisition unit initiates a data acquisition request to the sensor in the DC submerged arc furnace, requesting real-time data of the chemical substances in the furnace. The request message body carries the data acquisition unit address, sensor address, time range of the requested data, and requested data type. The requested data type includes temperature, pressure, and chemical component concentration. The sensor responds to the request and returns the original detection data of the DC submerged arc furnace.

8. A DC ore furnace detection system according to claim 6, characterized in that: The detection unit is also used to: The detection unit receives the original detection data from the sensor, removes duplicate, invalid or erroneous data, calibrates the cleaned data, and labels the temperature data, pressure data, and chemical composition concentration data according to the characteristics of the detection data.

9. A DC ore-bearing furnace detection system according to claim 6, characterized in that: The data processing unit is also used to: Collect chemical substance data in the furnace and identify the type, concentration and change trend of chemical substances in the furnace through a preset algorithm model; Receive and set thresholds, build an anomaly detection model, use the anomaly detection model to detect the type, concentration and change trend of chemical substances in the furnace, obtain abnormal detection results of chemical substances in the furnace, and alarm for data that exceeds the normal range, indicating existing equipment failures or safety hazards. Based on historical and current data, predict future trends in the chemical state of the furnace and equipment operation.

10. A DC ore furnace detection system according to claim 6, characterized in that: The data processing unit is also used to: Retrieving historical chemical substance detection data from a DC ore-fired furnace knowledge base, preprocessing the acquired historical chemical substance detection data, and obtaining preprocessed historical chemical substance detection data; Use a preset algorithm model to analyze the pre-processed historical chemical substance detection data to identify the chemical substances in the DC submerged arc furnace and the characteristics of the DC submerged arc furnace; According to the chemical substances in the DC ore-forming furnace and the characteristics of the DC ore-forming furnace, matching is performed in the DC ore-forming furnace knowledge base to obtain the corresponding chemical detection substances. Apply chemical test substances to the parts to be tested of the DC ore-fired furnace, including the furnace interior and the furnace wall of the DC ore-fired furnace, and collect reaction data between the chemical test substances and the chemical substances in the DC ore-fired furnace; Perform data feature extraction on the reaction data between the chemical test substance and the chemical substances in the DC submerged arc furnace to obtain the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace. Compare and analyze the type of chemical substances in the DC submerged arc furnace, the concentration of chemical substances in the DC submerged arc furnace, and the color data of chemical substances in the DC submerged arc furnace with the preset chemical test standards to obtain the chemical test results; Collect the operating data after the device end of the DC submerged arc furnace executes the real-time electrical equipment adjustment and maintenance data to obtain the adjusted operating data, and compare the adjusted operating data with the preset DC submerged arc furnace standard data. If the adjusted operating data is not in the preset DC submerged arc furnace standard data, use the distillation algorithm to optimize the real-time electrical equipment adjustment and maintenance data, and send the optimized real-time electrical equipment adjustment and maintenance data to the DC submerged arc furnace device end.

Images