A method and apparatus for measuring the length of the working end of an electrode in a calcium carbide furnace.
By dynamically adjusting the electrode working end length measurement method using intelligent sensors and computational models, the problems of low measurement accuracy and high safety risks in existing technologies are solved, realizing real-time, accurate measurement and safe automated control of the electrode working end length.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 聊城研聚新材料有限公司
- Filing Date
- 2026-03-02
- Publication Date
- 2026-06-02
Smart Images
Figure CN122130023A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of calcium carbide furnace electrodes, and in particular to a method and apparatus for measuring the length of the working end of a calcium carbide furnace electrode. Background Technology
[0002] In the smelting process of electric arc furnaces (such as calcium carbide furnaces), the working end length of the self-baking electrode is a key process parameter that determines the furnace reaction efficiency, electrode consumption rate, and overall energy consumption level. Accurate measurement of the electrode working end length is of great significance for achieving stable furnace condition control, improving energy utilization efficiency, and reducing production costs. If the electrode length is out of control, it can easily lead to excessive electrode consumption, increased furnace temperature fluctuations, and even production safety accidents.
[0003] Currently, the industry mainly relies on manual experience or offline indirect measurement methods to measure the length of the working end of the electrode. The common practice is for operators to estimate based on experience based on parameters such as electrode pressure and discharge amount and paste column height, or to perform contact measurement using special tools after the machine is stopped. Such methods are not only cumbersome to operate, but also take a long time for a single measurement (usually more than 10 minutes), which seriously affects the continuity of production.
[0004] The existing technology has obvious defects: First, the measurement accuracy is low, and the error of human estimation generally exceeds 10%; second, the data lag is serious, and real-time monitoring and feedback control cannot be achieved; third, the operator needs to be close to the high-temperature furnace body during the measurement process, which poses a high safety risk; in addition, because the density of the electrode paste changes significantly in different temperature ranges, the measurement results are seriously inaccurate when the electrode length changes or the furnace temperature fluctuates. Summary of the Invention
[0005] This application provides a method and apparatus for measuring the length of the working end of an electrode in a calcium carbide furnace, which can at least partially solve the above-mentioned technical problems.
[0006] Firstly, this application provides a method for measuring the length of the working end of an electrode in a calcium carbide furnace, which adopts the following technical solution: A method for measuring the length of the working end of an electrode in a calcium carbide furnace includes the following steps: Data acquisition: Real-time acquisition of the total weight of the holder, the height of the paste column, the furnace temperature, and the weight of the holder itself; Coefficient determination: Call the set paste column weight coefficient and electrode weight coefficient; Length calculation: Based on the total weight, paste column height, paste column weight coefficient, holder weight and electrode weight coefficient, a calculation model is established, and the length of the working end of the electrode is calculated through the calculation model.
[0007] By adopting the above technical solution, the total weight of the holder, real-time paste column height, and real-time furnace temperature data are collected through intelligent or mechanical sensors. The holder automatically calls the set paste column weight coefficient and electrode weight coefficient, and then inputs this information into the calculation model. The calculation model calculates the electrode working end length. By acquiring multi-source data in real time and combining it with the calculation model, real-time performance and accuracy are improved. This transforms the measurement from relying on manual experience estimation or offline contact measurement to online, automated, and contactless measurement based on physical quantities. This is the most fundamental advancement, replacing traditional manual estimation (error > 10%) and achieving contactless measurement. Online measurement improves accuracy by over 98%, reducing single measurement time from over 10 minutes to real-time. The measurement process is abstracted into a clear mathematical formula and procedure, allowing for precise description, replication, and verification, laying the foundation for all subsequent optimizations. Safety is enhanced by reducing operator proximity to the high-temperature furnace, lowering safety risks, and increasing automation. Results are output directly without human intervention, providing core parameters for stable furnace control. Furthermore, electrode length measurement can be performed at any time without shutdown, unlike existing weighing technologies, which require no waiting for electrode cooling or shutdown, minimizing impact on the calcium carbide furnace operation.
[0008] Optionally, the coefficient determination step further includes establishing a relational database containing measured density values of electrode paste in different temperature ranges, and retrieving the set paste column weight coefficient and electrode weight coefficient from the relational database, as follows: The real-time collected furnace temperature is matched and interpolated with the database to output the paste column density value and electrode density value at the current temperature, and then converted into the paste column weight coefficient and electrode weight coefficient.
[0009] The original scheme used fixed paste column weight coefficients and electrode weight coefficients, resulting in some inaccuracies in the calculation results. By adopting the above technical solution, a relational database is introduced into the coefficient determination step. This database stores the measured density values of the electrode paste at different temperature ranges (e.g., 200-1500℃). After real-time acquisition of the furnace temperature, matching and interpolation calculations (e.g., linear interpolation) are used to output the paste column density and electrode density at the current temperature from the database, and then convert them into weight coefficients (e.g., the paste column weight coefficient is dynamically adjusted with temperature). Example: When the furnace temperature is 1200℃, a density of 1.5-2.0 is used. g / cm³, converted to a weight coefficient of 2.8-3.0 tons / meter; the measurement is upgraded from using a fixed empirical coefficient to using a dynamic coefficient that varies with the main operating condition parameter (temperature), making the model closer to the real physicochemical process. Through dynamic accuracy optimization, the coefficient is automatically adjusted according to the furnace temperature change, solving the measurement inaccuracy problem caused by the temperature fluctuation of electrode paste density; and data-driven reliability, based on 300 sets of measured data, ensures that the coefficient is adaptable to multiple scenarios in the range of 1100-1600℃, with an error ≤±1%. Combined with the original solution, the robustness of the model under high temperature and variable operating conditions is further improved, and the measurement accuracy is further optimized from ±1%.
[0010] Optionally, the establishment of the relational database also links the correspondence between the glue column height and the glue column weight coefficient, so that the glue column weight coefficient can be adjusted according to the real-time glue column height.
[0011] By adopting the above technical solution, the paste column itself is not a homogeneous medium; its compaction degree and sintering state may differ in different height ranges, resulting in its "average density" (i.e., weight coefficient) not being constant. This solution solves this error source ignored by traditional methods, achieving self-adaptation to production process fluctuations. Electrodes are continuously consumed and released during production, and the paste column height is a continuously changing process parameter. A relational database is used to link the paste column height with its weight coefficient (e.g., a height of 3.5-5 meters corresponds to a coefficient of 2.6-3.0 tons / meter). Based on real-time... The weight coefficient of the glued column is dynamically adjusted according to its height; for example, the coefficient increases accordingly as the height increases. When the glued column height is maintained between 4.0 and 4.2 meters, the coefficient remains stable at 2.8 tons / meter. A height compensation mechanism is introduced to overcome the nonlinear effect of changes in glued column height on weight and reduce systematic errors. Full working condition coverage is achieved, expanding the applicability of the model and ensuring calculation accuracy when the glued column height fluctuates (e.g., between 3.5 and 5 meters). Combined with temperature compensation, the model tracks this change and self-adjusts, achieving bivariate optimization of "temperature-height," improving model adaptability by more than 20%.
[0012] Optionally, the relationship between the height of the glue column and the weight coefficient of the glue column is as follows: the weight coefficient of the glue column increases with the increase of the height of the glue column and decreases with the decrease of the height of the glue column, with an increase or decrease of 5% per meter.
[0013] By adopting the above technical solution and through dual dynamic compensation of temperature and altitude, systematic errors caused by changes in operating conditions are greatly eliminated. This is a significant enhancement of the effect of claim 1, upgrading the measurement model from a "static empirical formula" to a "dynamic adaptive intelligent model," ensuring that the measurement accuracy remains stable within ±1% under a wider range of process parameter fluctuations.
[0014] With full-condition adaptability, this model effectively covers the two core variables that inevitably exist in the production process: temperature fluctuations and continuous changes in paste column height. Whether it is temperature changes caused by furnace start-up, furnace shutdown, or load adjustment, or height changes caused by normal pressing and consumption, the system can automatically adapt, solving the pain point that single-variable models are prone to inaccuracy under boundary conditions.
[0015] The fuzzy experience of operators ("higher temperature means higher density," "the calculation is different when the column is taller") is transformed into a precise, quantifiable mathematical algorithm that can be recognized and executed by the control system. This not only improves the reliability of the results but also fundamentally changes the replicability and scalability of the technology, as the operation no longer relies on the experience of "veteran operators." The synergy between temperature compensation and height compensation produces a 1+1>2 effect. Temperature compensation alone still has deviations when the height changes drastically, and height compensation alone is also inaccurate when the temperature changes drastically. The combination of the two forms a three-dimensional compensation network that approximates the real physical world. This is a core innovation and significant advancement in the combination of patent applications.
[0016] Optionally, the correspondence between the height of the glue column and the weight coefficient of the glue column is established by assigning different weight coefficients to historical data for different height ranges, specifically including: Weighting steps: When establishing the relational database, the measured data collected within the range of 4.0-4.2 meters in height of the glued column are assigned the highest weight; Weighted calculation steps: When calling the weight coefficient of the glue column, the system prioritizes matching the coefficient value corresponding to the high weight interval based on the real-time glue column height. If it is at the interval boundary, the weighted average algorithm is used to calculate the final coefficient.
[0017] By adopting the above technical solutions, and assigning the highest weight to the high-frequency, high-stability working condition range (column height 4.0-4.2 meters), the system can prioritize the most reliable and common data foundation when calling data. This directly improves the accuracy and representativeness of coefficient determination from the data source. The weighted average algorithm is used to handle the boundary conditions of the range, avoiding coefficient jumps caused by small height fluctuations, making coefficient changes continuous and smooth, thus ensuring the stability of the final length calculation result and preventing drastic oscillations in the output value. The constructed "bivariate dynamic compensation model" has been "optimized and refined." When determining the basic coefficients through temperature, not only temperature but also different heights at that temperature are considered. The reliability of the temperature data is enhanced; for example, at 1200℃, the system prioritizes density data measured within the 4.0-4.2 meter height range to establish the temperature-coefficient relationship, making the base coefficient itself more accurate; the "one-size-fits-all" linear compensation rule (±5% per meter) is supplemented with a "confidence level" mechanism. When the real-time height falls within the most commonly used 4.0-4.2 meter range, the system has the highest confidence level in the linear rule; when the height deviates from this range, the system will fine-tune the fixed compensation rate by weighted averaging and comprehensively referencing data from other height ranges; this is equivalent to adding a "smart calibrator" to the simple linear model, enabling it to remain robust even under unconventional operating conditions.
[0018] Optionally, it also includes a prediction step, which includes a data preprocessing step, a rate of change calculation step, a furnace temperature trend analysis step, a prediction model establishment step, and a prediction result output step. Data preprocessing: Based on the continuously collected column height data within a historical time period, outliers are removed, and the data sequence is smoothed using the moving average method; Rate of change calculation: Time series analysis is performed on the smoothed column height data series to calculate the rate of change of the column height. The rate of change is obtained by linear regression or difference method, and the unit is meters per hour. Furnace temperature trend analysis: Real-time collection of furnace temperature data, combined with historical furnace temperature data, and the use of trend fitting methods to predict the furnace temperature change trend within a set future time period; Prediction model establishment: Using the rate of change and furnace temperature trend as input variables, establish a multiple linear regression model or a time series prediction model to output the predicted value of the change in electrode working end length within a set future time period; Prediction output: The predicted values are integrated into the submerged arc furnace control system for real-time adjustment of electrode operating parameters.
[0019] By adopting the above technical solutions, a leap from real-time monitoring to forward-looking prediction has been achieved: from "knowing how long it is now" to "knowing how long it will be in the future," providing a time window for proactive control; by integrating the two key influencing factors of the paste column height change rate and furnace temperature trend, the prediction model is closer to the actual physical process, and the prediction results are more scientific; this is a qualitative change from "perception" to "cognition," which together ensures that the measurement of the "current state" is extremely accurate. The accuracy of any prediction model is highly dependent on the quality of the input data. The calculated current electrode length value is the absolute benchmark for the prediction model calculation. If this benchmark error is large, the prediction value will be meaningless; the "paste column height change rate" required by the prediction model depends on the continuous and accurate height sequence data provided by the current electrode length value; if the height data itself fluctuates greatly or is inaccurate, the calculated change rate will also be distorted; the high-precision and high-stability real-time measurement achieved is the fundamental premise for the establishment and effectiveness of the prediction function; this combination makes prediction no longer a theoretical conjecture, but a scientific inference based on reliable data, thus enabling it to be truly used to guide production adjustments.
[0020] Optionally, it also includes an early warning step, which specifically includes: Set a minimum value for the working end length of an electrode, and compare the predicted value obtained in the prediction step with the minimum value; If the predicted value is greater than the minimum value, calculate the critical time for the predicted value to drop to the minimum value, and initiate different levels of tiered early warning based on the length of the critical time. If the predicted value is less than or equal to the minimum value, the highest level of emergency warning will be triggered immediately.
[0021] By adopting the above technical solutions, a single alarm is transformed into a tiered early warning system, enabling operators to take different measures according to the level of urgency, thus optimizing human resources and emergency response processes. By calculating the "critical time," potential downtime risks are quantified and visualized, realizing the transformation from "alarm after failure" to "early warning before failure," a key link in the "prediction-decision-action" closed-loop control. It provides information on "what will happen" (predicted values) and defines the rules for "what we should do about it" (early warning strategies).
[0022] Without accurate predictions, early warnings are like water without a source. By establishing highly reliable prediction models, early warnings can be trusted and implemented by production personnel, thereby truly reducing the number of downtimes caused by electrode abnormalities from an average of 3 times per month to 0 times. This combination transforms technical insights into tangible productivity and safety.
[0023] Optionally, it also includes a calibration step, which includes a periodic acquisition step, a data fitting step, a coefficient optimization step, and a verification step; Regular data collection: Regularly collect multiple sets of total weight data of the holder under different working conditions, including different paste column heights, different furnace temperatures, and different electrode consumption stages; Data fitting: The least squares method or nonlinear regression algorithm is used to fit the collected total weight data of the gripper with the measured working end length of the electrode, and the paste column weight coefficient and electrode weight coefficient are solved in reverse. Coefficient optimization: The fitted coefficients are compared with the original coefficients, the coefficient values are automatically adjusted according to the error threshold, and the results are updated in the relational database. Verification and iteration: Through three batches of process verification, ensure that the error of the optimized coefficients is stable within ±1% in actual measurement; otherwise, repeat the periodic data collection and data fitting steps.
[0024] By adopting the above technical solutions, the problem of decreased measurement accuracy caused by long-term factors such as changes in sensor characteristics and batch differences in electrode paste raw materials is solved. The system can automatically optimize core parameters (paste column / electrode weight coefficient) through data fitting, and has the ability to learn and optimize itself, ensuring the "long-term robustness" of the precision measurement system. Through the above solutions, the basic coefficients of the dynamic model that has been running for a long time are adjusted, so that the calculation results remain accurate.
[0025] This creates a virtuous cycle: the initial model guides production, production accumulates new data, the model is calibrated and optimized regularly, and the optimized model guides production more accurately. This combination ensures that the technology not only performs well during project acceptance but also plays a continuous and stable role in front-line production for several years, greatly enhancing the technology's life cycle and promotional value.
[0026] Optionally, in the coefficient determination step, an attenuation factor based on the electrode consumption stage is introduced into the electrode weight coefficient, specifically including: Phase division steps: Based on the cumulative usage time or cumulative pressure discharge of the electrode, the electrode life cycle is divided into the initial stage, the stable stage, and the later stage; Coefficient decay step: In the later stage of electrode use, the electrode weight coefficient is multiplied by a decay factor less than 1 to reflect the slight decrease in average density of the lower part of the electrode due to excessive sintering and erosion.
[0027] By adopting the above technical solutions, full life-cycle accuracy management is achieved. By introducing an attenuation factor, the slight changes in the physical properties (such as average density) of the electrodes caused by long-term high-temperature sintering and arc erosion are compensated, ensuring the accuracy of measurements in the later stages of electrode use. This allows the measurement model to more closely align with the actual physical evolution of the production equipment. In the "dynamic compensation model for operating conditions," compensation for the "time dimension" is added, enabling the model to respond not only to instantaneous temperature and altitude changes but also to slow aging processes lasting several weeks or months. In the calibration process, the system can not only optimize the basic coefficients but also verify and adjust the "later stage" classification criteria and the specific values of the "attenuation factor." Together, these two aspects construct a full-time-space-scale adaptive measurement system that covers both "short-term operating condition fluctuations" and "long-term equipment aging."
[0028] Secondly, the device provided in this application adopts the following technical solution: An apparatus comprising the following modules: The weighing module is used to collect the total weight of the electrode holder in real time. The height detection module is used to collect the height of the glue column in real time; The data processing module has its input end connected to the output end of the weighing module and the output end of the height detection module, and is configured to execute the method for measuring the length of the working end of the calcium carbide furnace electrode as described in any one of claims 1-9, and to calculate the length of the working end of the electrode; The output module has its input end connected to the output end of the data processing module and is used to output and display the length of the working end.
[0029] By adopting the above technical solution and setting up various modules of the device, information data can be collected, processed, calculated, and output, thereby achieving automated measurement.
[0030] In summary, this application includes at least one of the following beneficial technical effects: 1. By abstracting the measurement process into a clear mathematical formula and procedure, the technology can be accurately described, replicated, and verified, laying the foundation for all subsequent optimizations; safety is enhanced by reducing the need for operators to approach the high-temperature furnace body, lowering safety risks, improving automation levels, and providing direct output results without human intervention, thus providing core parameters for stable furnace control; moreover, it does not require shutdown and electrode length measurement can be performed at any time. Compared with the existing weighing measurement technology, it does not require waiting for electrode cooling or shutdown, reducing the impact on the operation of the calcium carbide furnace; 2. By assigning the highest weight to the high-frequency, high-stability operating range (column height 4.0-4.2 meters), the system can prioritize the most reliable and common data base when calling data, which directly improves the accuracy and representativeness of coefficient determination from the data source. The weighted average algorithm is used to handle the boundary cases of the range, avoiding coefficient jumps caused by small fluctuations in height, making the coefficient changes continuous and smooth, thereby ensuring the stability of the final length calculation result and preventing drastic oscillations in the output value. 3. During the calibration process, the system can not only optimize the basic coefficients, but also verify and adjust the classification criteria for the "later stage" and the specific value of the "attenuation factor"; the two work together to build an adaptive measurement system covering both "short-term operating condition fluctuations" and "long-term equipment aging" across all time and space scales. Attached Figure Description
[0031] Figure 1 This is a flowchart of the measurement method in Embodiment 1 of this application; Figure 2 This is a flowchart of the measurement method in Embodiment 4 of this application. Detailed Implementation
[0032] The following combination Figures 1 to 2 This application will be described in further detail.
[0033] This embodiment discloses a method for measuring the length of the working end of an electrode in a calcium carbide furnace.
[0034] Example 1: Refer to Figure 1 The method for measuring the length of the working end of an electrode in a calcium carbide furnace includes the following steps: Data acquisition: Real-time acquisition of the total weight of the holder, the height of the paste column, the furnace temperature, and the weight of the holder itself; Specifically, the data acquisition step aims to obtain key parameters related to the electrode state in real time using smart or mechanical sensors, including the total weight of the holder, the height of the paste column, furnace temperature data, and the holder's own weight. The specific implementation is as follows: Total weight of the holder: measured in real time using a high-precision weight sensor (such as a strain gauge or piezoelectric sensor) mounted on the holder's support structure. The sensor should be heat-resistant, dustproof, and interference-resistant to adapt to the high-temperature environment near the submerged arc furnace. A data acquisition frequency of 1-10 times per second is recommended to ensure real-time performance. The weight sensor output signal is converted into a digital signal by a transmitter and transmitted to the control system.
[0035] The height of the paste column is measured using non-contact measuring equipment, such as a laser rangefinder or ultrasonic sensor, which is perpendicular to the surface of the electrode paste. The sensor is installed in a fixed position above the holder to avoid the influence of high temperature and dust in the furnace. Since the surface of the electrode paste may fluctuate, the accuracy can be improved by taking the average value of multiple measurements (e.g., 5 measurements per second and taking the median value). The measurement data is transmitted to the control system in real time.
[0036] Furnace temperature data: The furnace temperature is measured using multiple thermocouples or infrared thermometers placed around the electrodes. The thermocouples should be inserted to an appropriate depth inside the furnace (e.g., outside the electrode shell) to avoid direct contact with the electrodes, and the average value of multiple measurement points is taken as the representative temperature. The data acquisition frequency is synchronized with the weight acquisition, typically once per second; the temperature data is used to compensate for density changes in the electrode paste and the working end of the electrodes.
[0037] Holder weight: This is a constant value, provided by the equipment manufacturer or obtained through calibration before installation; during calibration, the weight of the holder is measured under no-load conditions (without electrode paste and electrode working ends) and recorded as follows. In actual operation, As a constant input control system.
[0038] Coefficient determination: Call the set paste column weight coefficient and electrode weight coefficient; Length calculation: Based on the total weight, paste column height, paste column weight coefficient, holder weight and electrode weight coefficient, a calculation model is established, and the length of the working end of the electrode is calculated through the calculation model.
[0039] The calculation model is as follows: ,in, Indicates the total weight. Indicates the height of the paste column. Indicates the weight coefficient of the paste column. Indicates the weight of the gripper. This represents the electrode weight coefficient.
[0040] Specifically, the coefficient determination step involves the paste column weight coefficient. and electrode weight coefficient The calculation or invocation of these coefficients is related to the density of the electrode paste and the working end of the electrode, and since the density varies with temperature, it needs to be dynamically adjusted based on the furnace temperature data. The specific implementation method is as follows: Paste column weight coefficient This coefficient represents the weight (kg / m) of electrode paste per unit height. The calculation formula is: ,in, This indicates the density of the electrode paste (kg / m³). This indicates the volume of electrode paste per meter at the working end of the electrode. The density of the electrode paste is a fixed value, i.e. .
[0041] Electrode weight coefficient This coefficient represents the weight per unit length of the electrode's working end (kg / m). The calculation formula is: ,in, This indicates the volume per meter of the electrode's working end. This represents the density per cubic meter at the working end of the electrode; the density per cubic meter at the working end of the electrode is a fixed value. ; In this embodiment, and Take fixed values, respectively , ; The coefficient determination process is completed automatically by the control system, based on real-time data acquisition. retrieval and This ensures that the furnace temperature can be adapted to fluctuations.
[0042] Length Calculation: The length calculation step is based on the collected data and determined coefficients, and the length of the electrode working end is solved in real time through a calculation model. The calculation model is as follows: ,in, Indicates the total weight. Indicates the height of the paste column. Indicates the weight coefficient of the paste column. Indicates the weight of the gripper. This represents the electrode weight coefficient.
[0043] For example, a large calcium carbide furnace is in a stable smelting stage. Operators need to monitor the working end length of a self-baking electrode in real time from the control room to ensure stable furnace conditions. This system integrates the measurement method described in Example 1.
[0044] Given system parameters: electrode paste column diameter 1500mm; holder weight: 21000kg; Paste column weight coefficient and electrode weight coefficient Take fixed values, respectively , ; Real-time data acquisition: at each specific moment The system data acquisition module obtained the following set of real-time data: gross weight : 40100 kg (read by the weight sensor below the grip); Column height 4.1m (measured by the laser rangefinder at the top); Furnace temperature 1200℃; Substitute into the calculation model, .
[0045] Embodiment 1 of this application automatically detects the height of the paste column, establishes a calculation model, and completes the measurement of the electrode working end length without shutting down the machine. The entire measurement process requires no human intervention, the sensor remotely collects data, and the operator can monitor all information in the central control room. This fundamentally eliminates occupational health risks such as high-temperature burns and fume inhalation, meeting the safety development requirements of modern industrial automation and unmanned operation. It avoids production interruptions caused by machine shutdown for measurement, improving equipment utilization and overall production efficiency. Online real-time measurement eliminates the need to shut down the furnace, ensuring continuous and stable production operation. It helps maintain optimal furnace conditions, thereby improving product quality, reducing excessive electrode consumption and energy loss, and directly reducing production costs.
[0046] Example 2: The difference from Example 1 is that the coefficient determination step further includes establishing a relational database. The relational database contains measured density values of the electrode paste at different temperature ranges. The set paste column weight coefficient and electrode weight coefficient are retrieved from the relational database, as follows: The real-time collected furnace temperature is matched and interpolated with the database to output the paste column density value and electrode density value at the current temperature, and then converted into the paste column weight coefficient and electrode weight coefficient.
[0047] The establishment of the relational database also establishes a correspondence between the height of the glue column and the weight coefficient of the glue column, enabling the weight coefficient of the glue column to be adjusted according to the real-time height of the glue column.
[0048] The relationship between the height of the glue column and the weight coefficient of the glue column is as follows: the weight coefficient of the glue column increases with the increase of the height of the glue column and decreases with the decrease of the height of the glue column, with an increase or decrease of 5% per meter.
[0049] Specifically, in the relationship between the length range of the glue column and the weight coefficient of the glue column, the glue column height is 4.0-4.2m, and the corresponding glue column weight coefficient is 2.8 tons / m. When the glue column height is less than 4.0m, the weight coefficient decreases by 5%, and when it is greater than 4.2m, the glue column weight coefficient increases by 5%.
[0050] This represents the density per cubic meter at the electrode's working end, which varies with temperature, increasing as temperature rises; for example, an empirical formula can also be established. ,in, and These are material constants, which are determined through laboratory testing.
[0051] This represents the density per cubic meter at the working end of the electrode, which varies with temperature; the density increases with increasing temperature. For example... The determination of can also be established using empirical formulas, with the calculation method being the same as... The calculation.
[0052] Specifically, the coefficient determination step achieves dynamic compensation through a relational database, and the specific implementation method is as follows: Relational database establishment: The relational database adopts an embedded SQLite database and contains the following core data tables: Based on a column height of 4.0-4.2m, a temperature-density relationship table was created, as shown in Table 1 below: Table 1 Temperature-Density Relationship
[0053] When the temperature is below 1100℃, the electrode paste is not fully sintered and has a low density; as the temperature increases, the degree of sintering increases and the density increases accordingly. The baseline value of electrode density remains unchanged.
[0054] The relationship between the height of the paste column and the correction factor is shown in Table 2 below: Table 2 Relationship between Paste Column Height and Correction Factor
[0055] Update the paste column weight coefficient to ,in Indicates the height of the paste column is Correction factor at that time.
[0056] For example, a large calcium carbide furnace is in a stable smelting stage. Operators need to monitor the working end length of a self-baking electrode in real time from the control room to ensure stable furnace conditions. This system integrates the measurement method described in Example 1.
[0057] Given system parameters: electrode paste column diameter: 1500mm; electrode diameter: 1506mm; holder weight: 21000kg; Paste column weight coefficient and electrode weight coefficient The determination is based on the temperature and the height of the paste column; Real-time data acquisition: at each specific moment The system data acquisition module obtained the following set of real-time data: gross weight 38,000 kg (read by the weight sensor below the grip); Column height 4.1m (measured by the laser rangefinder at the top); Holder weight 21,000 kg; Furnace temperature 1000℃; Coefficient determination: temperature and density selection, Temperature below 1100℃, within the 800-1100℃ range, obtained from the database, paste column density. Electrode density ; Height correction factor determined: The value falls within the 4.0-4.2m range, with a correction factor of [missing value]. ; Calculation: Substitute into the calculation model, .
[0058] Example 3: The difference between this example and Example 2 is that the correspondence between the height of the glue column and the weight coefficient of the glue column is established by assigning different weight coefficients to historical data for different height ranges, specifically including: Weighting steps: When establishing the relational database, the measured data collected within the range of 4.0-4.2 meters in height of the glued column are assigned the highest weight; Weighted calculation steps: When calling the weight coefficient of the glue column, the system prioritizes matching the coefficient value corresponding to the high weight interval based on the real-time glue column height. If it is at the interval boundary, the weighted average algorithm is used to calculate the final coefficient.
[0059] Specifically, the weighting steps are as follows: When establishing the relational database, different weight coefficients are assigned to historical data in different height ranges, and a column height weighting allocation table is established, as shown in Table 3 below: Table 3 Weighting of Paste Column Height
[0060] Weight allocation is based on the following criteria: the 4.0-4.2 meter range is the optimal process range with the most stable data, and is therefore assigned the highest weight of 1.0. The weights of adjacent ranges decrease sequentially based on their distance from the optimal range. The weight coefficients are determined based on long-term production data statistical analysis.
[0061] Weighted calculation steps: When the real-time gluing column height is within the interval boundary, the weighted average algorithm is used to calculate the final gluing column weight coefficient: Weighted average calculation formula: ,in, This is the final calculated paste column weight coefficient. Let be the weight coefficient for the j-th height interval. This is the baseline value for the paste column weight coefficient corresponding to the j-th height interval. This represents the number of height intervals involved in the calculation.
[0062] For example, during the adjustment of the production process of a certain calcium carbide furnace, the height of the paste column fluctuates around the boundary of 4.0-4.2 meters, requiring precise calculation of the length of the electrode working end. This system integrates the weight allocation and weighted calculation mechanism described in Example 3.
[0063] Given system parameters: electrode paste column diameter: 1500mm; electrode diameter: 1506mm; holder weight: 21000kg; Paste column weight coefficient and electrode weight coefficient The determination is based on the temperature and the height of the paste column; Real-time data acquisition: at each specific moment The system data acquisition module obtained the following set of real-time data: gross weight : 42000 kg (read by the weight sensor below the grip); Column height 4.15m (measured by the laser rangefinder at the top); Holder weight 21,000 kg; Furnace temperature 1250℃; Coefficient determination: temperature and density selection, Temperature above 1100℃, within the 1100-1600℃ range, obtained from the database, paste column density. Electrode density ; Height correction factor determined: The value falls within the 4.0-4.2m range, with a correction factor of [missing value]. ; Due to its proximity to the 4.2-meter boundary, and considering the influence of the 4.2-4.4-meter range, the weight coefficients are obtained from the weight allocation table: 4.0-4.2-meter range: weight W1 = 1.0; 4.2-4.4-meter range: weight W2 = 0.9; ; Calculation: Substitute into the calculation model, .
[0064] Example 4: Reference Figure 2 The difference between this embodiment and embodiment 1 is that it also includes a prediction step, which includes a data preprocessing step, a rate of change calculation step, a furnace temperature trend analysis step, a prediction model establishment step, and a prediction result output step. Data preprocessing: Based on the continuously collected column height data within a historical time period, outliers are removed, and the data sequence is smoothed using the moving average method; Rate of change calculation: Time series analysis is performed on the smoothed column height data series to calculate the rate of change of the column height. The rate of change is obtained by linear regression or difference method, and the unit is meters per hour. Furnace temperature trend analysis: Real-time collection of furnace temperature data, combined with historical furnace temperature data, and the use of trend fitting methods to predict the furnace temperature change trend within a set future time period; Prediction model establishment: Using the rate of change and furnace temperature trend as input variables, establish a multiple linear regression model or a time series prediction model to output the predicted value of the change in electrode working end length within a set future time period; Prediction output: The predicted values are integrated into the submerged arc furnace control system for real-time adjustment of electrode operating parameters.
[0065] It also includes an early warning step, which specifically includes: Set a minimum value for the working end length of an electrode, and compare the predicted value obtained in the prediction step with the minimum value; If the predicted value is greater than the minimum value, calculate the critical time for the predicted value to drop to the minimum value, and initiate different levels of tiered early warning based on the length of the critical time. If the predicted value is less than or equal to the minimum value, the highest level of emergency warning will be triggered immediately.
[0066] Specifically, based on Example 1, Example 4 adds a systematic collection and storage of historical data: Extended data acquisition: In addition to real-time data, the system continuously records historical time-series data, including paste column height, furnace temperature, electrode working end length, etc., with a sampling interval of 1 minute; Data quality assurance: A dual verification mechanism is adopted, including range verification (e.g., the height of the glued column is between 2.0 and 6.0 meters) and rate of change verification (e.g., a single change does not exceed 0.1 meters). Prediction steps: Data preprocessing steps: Cleaning and smoothing the continuously collected column height data within the historical time period: Outlier removal formula: Moving average smoothing formula: Where N=10 (10-minute window). The average height of the glued column. Standard deviation It is the height of the paste column detected at time t; This indicates the height of the paste column after cleaning; Indicates the height of the paste column after smoothing; Steps for calculating the rate of change: Perform rate of change analysis on the smoothed data: Linear regression method: Where n=30 (30-minute data window), n represents the number of data points for regression analysis; Difference method: ,in (5-minute intervals) indicate the predicted time span; Furnace temperature trend analysis steps: Trend prediction based on historical furnace temperature data: Exponential smoothing prediction model: ,in, Represents the smoothing coefficient. k = 1, 2, ..., 12 (representing the next 1-12 hours); This represents the predicted temperature at time k in the future; Steps for building a prediction model: Establish a multiple linear regression prediction model: ,in, , , , , These are the regression coefficients (obtained through training with historical data); Indicates the trend of temperature change; (Forecast period); This represents the predicted electrode length at the future time Δt2.
[0067] Prediction result output steps: Integrate the prediction results into the control system: It displays the electrode working end length change curve in real time over the next 4 hours, provides a prediction confidence interval (95% confidence level), and generates operation suggestions (such as timing and amount of compression and release). Early warning steps: Early warning rule setting: Set the minimum length of the electrode working end to be... ; Early warning logic implementation: Critical time calculation: ,in, This indicates the rate of change of the length of the working end of the motor; Tiered Alert: Green Alert: Normal monitoring; Yellow alert: Pay attention; Orange alert: Adjustments are recommended; Red alert: Immediate intervention; emergency warning: Emergency shutdown.
[0068] For example, during continuous operation, a large calcium carbide furnace needs to use a predictive system to identify the risk of insufficient electrode working end length in advance to avoid unplanned shutdowns.
[0069] System parameters: Electrode diameter: 1500 mm; Minimum electrode working end length is [value missing]. ; Current data: Real-time electrode length: Height of the paste column: Furnace temperature 1250℃; Total weight of the gripper 45,000 kg, weight of the gripper 21,000 kg; Predictive Analysis: Data Processing: Collect the height data of the past 30 minutes, remove outliers, and smooth using a 10-minute moving average. ; Rate of change calculation: calculated using linear regression method: (Rate of decrease in height of the paste column); Furnace temperature trend analysis: Prediction based on exponential smoothing model: , ; Predictive model building: Substitute the trained regression model (coefficients trained based on historical data): Considering the confidence interval, the actual prediction range is 3.2-3.5 meters. Early warning judgment: Calculate the critical time: current electrode consumption rate ; .
[0070] Warning output: Green warning, normal monitoring; Warning information: The electrode working end is expected to approach the safety limit in 7.5 hours, and it is recommended to arrange electrode pressing and releasing after 4.5 hours; Operation suggestion: Pressing and releasing amount 0.3 meters, which is expected to extend the running time by 12 hours.
[0071] In other embodiments, a calibration step is also included, which includes a periodic acquisition step, a data fitting step, a coefficient optimization step, and a verification step. Regular data collection: Regularly collect multiple sets of total weight data of the holder under different working conditions, including different paste column heights, different furnace temperatures, and different electrode consumption stages; Data fitting: The least squares method or nonlinear regression algorithm is used to fit the collected total weight data of the gripper with the measured working end length of the electrode, and the paste column weight coefficient and electrode weight coefficient are solved in reverse. Coefficient optimization: The fitted coefficients are compared with the original coefficients, the coefficient values are automatically adjusted according to the error threshold, and the results are updated in the relational database. Verification and iteration: Through three batches of process verification, ensure that the error of the optimized coefficients is stable within ±1% in actual measurement; otherwise, repeat the periodic data collection and data fitting steps.
[0072] Specifically, the data collection plan is as follows: Collection cycle: a full calibration is performed quarterly, and a partial calibration is performed monthly; Operating condition coverage: ensure that the collected data covers the following typical operating condition combinations, as shown in Table 4 below: Table 4 Typical Operating Conditions Operating conditions Column height range (m) Furnace temperature range (°C) Electrode consumption stage Normal operating conditions 4.0-4.2 1100-1300 Medium term (2-4 meters) Low paste column 3.5-3.8 1000-1200 Initial stage (>4 meters) High paste column 4.3-4.6 1200-1400 Late stage (<2 meters) low temperature 4.0-4.2 800-1000 Mid-term high temperature 4.0-4.2 1400-1600 Mid-term Data acquisition specifications: Each set of data includes: total weight of the holder, height of the paste column, furnace temperature, measured length of the working end of the electrode, timestamp, and operating condition identifier; Data quality control: A triple verification mechanism is adopted: Range verification: 3.0 ≤ height of the glued column ≤5.0, 500≤furnace temperature ≤1800; Consistency check: Coefficient of variation (CV) < 5% under the same operating conditions; Trend check: Data changes conform to process patterns; Data fitting steps: Least squares fitting: Establishing an overdetermined system of equations: ; Matrix representation ; in, ; ; ; Least squares solution: ; Nonlinear regression: When the data exhibits obvious nonlinear characteristics, nonlinear regression is used. ; The Levenberg-Marquardt algorithm is used to solve for the parameters. , , , .
[0073] Coefficient optimization steps: Error analysis: Calculate the relative error between the fitted coefficients and the original coefficients. ; ; Optimize decision rules: Small-scale optimization: δ < 5%, directly update coefficients; Medium-scale optimization: 5% ≤ δ < 15%, trigger verification steps; Large-scale change: δ ≥ 15%, alarm and manual confirmation required.
[0074] Coefficient smoothing update: A weighted average is used to achieve a smooth transition. ; ; Where α1 represents the update weight, α1=0.7.
[0075] Verification and iteration: Through three batches of process verification, ensure that the error of the optimized coefficients is stable within ±1% in actual measurement; otherwise, repeat the periodic data collection and data fitting steps.
[0076] Example 5: This example differs from Example 2 in that, in the coefficient determination step, an attenuation factor based on the electrode consumption stage is introduced into the electrode weight coefficient, specifically including: Phase division steps: Based on the cumulative usage time or cumulative pressure discharge of the electrode, the electrode life cycle is divided into the initial stage, the stable stage, and the later stage; Coefficient decay step: In the later stage of electrode use, the electrode weight coefficient is multiplied by a decay factor less than 1 to reflect the slight decrease in average density of the lower part of the electrode due to excessive sintering and erosion.
[0077] Specifically, the electrode lifecycle is divided into three stages based on the cumulative pressure discharge of the electrode: The electrode life cycle stages are divided into stages as shown in Table 5 below: Table 5 Electrode Life Cycle Stage Division Table stage Cumulative release amount ∑Δ (m) Stage characteristics Attenuation factor γ Early stage 0-2.0 The electrode sintering degree is insufficient, and the density is slightly low. 1.00 Stable period 2.0-8.0 The electrodes are fully sintered and have stable density. 1.00 Later ＞8.0 The lower part of the electrode is excessively sintered and eroded, resulting in a slight decrease in density. 0.95-0.98 Coefficient attenuation step: In the later stage of electrode use, an attenuation factor is introduced to dynamically adjust the electrode weight coefficient: Attenuation factor calculation model: γ=1.0-k*(∑ΔL-8.0), where k=0,005 is the attenuation coefficient, and the value range of γ is limited to [0.95, 1.0].
[0078] This application also discloses an apparatus comprising the following modules: The weighing module is used to collect the total weight of the electrode holder in real time. Specifically, the weighing module is responsible for collecting the total weight data of the electrode holder in real time, providing core input parameters for calculating the length of the working end of the electrode.
[0079] Specific implementation method: Sensor selection: High temperature resistant strain gauge weight sensor is adopted, with a measuring range of 0-60 tons, an accuracy class of 0.5%, and an operating temperature range of -20℃ to +300℃; Installation position: The sensor is installed at the stress key points of the gripper support structure, with a total of 4 measuring points set up in a 90° symmetrical distribution. Signal processing: Each sensor outputs a mV-level analog signal, which is converted into a standard industrial signal by a 4-20mA transmitter; a Σ-Δ type 24-bit ADC is used for high-precision analog-to-digital conversion; Data acquisition: Sampling frequency: 10 Hz; Digital filtering: Using moving average filtering to eliminate mechanical vibration interference; Temperature compensation: Built-in temperature sensor to compensate for temperature drift error in real time. The height detection module is used to collect the height of the glue column in real time; Specifically, the height detection module is used to measure the height of the electrode paste column in real time, providing key data for calculating the weight of the paste column.
[0080] Specific implementation method: Sensor selection: Explosion-proof laser rangefinder sensor is adopted. Measurement range: 0.5-10 m, accuracy: ±1 mm, laser class: Class 2, wavelength: 650 nm; Installation configuration: A fixed bracket is installed 3 meters above the holder, pointing vertically towards the surface of the electrode paste, and equipped with a compressed air blowing device to keep the optical window clean; Anti-interference measures: The Time-of-Flight (ToF) measurement principle is adopted, which has strong anti-dust interference capability. A measurement confidence threshold is set to eliminate data with poor reflection. Mechanical anti-vibration design reduces the impact of furnace vibration.
[0081] The data processing module has its input end connected to the output end of the weighing module and the output end of the height detection module, and is configured to execute the method for measuring the length of the working end of the calcium carbide furnace electrode and calculate the length of the working end of the electrode. Specifically, the data processing module is the core of the device, responsible for executing the algorithm for calculating the length of the working end of the electrode, and integrating all the functions of embodiments 1-5.
[0082] Specific implementation method: Hardware platform: industrial-grade PLC (such as Siemens S7-1500 series), CPU: multi-core processor, main frequency 1.5 GHz, memory: 4 GB DDR4, storage: 32 GB SSD for historical data storage; Software architecture: Real-time operating system: VxWorks or similar RTOS; Programming environment: IEC 61131-3 standard; Algorithm library: Contains all computational models from Examples 1-5. The output module, whose input end is connected to the output end of the data processing module, is used to output and display the working length, predicted length, and warning status.
[0083] Specifically, the output module is responsible for displaying the calculation results in various forms and providing a human-computer interaction interface.
[0084] Specific implementation: Display unit: Industrial touch screen: 15.6 inches, resolution 1920×1080, high-brightness LED display, suitable for strong light environments; Redundant display: Synchronous display on a large screen in the control room; Output content: Real-time data: Current electrode working end length, paste column height, and total weight; Trend display: Length change trend graph (4 hours, 24 hours, 7 days); Prediction information: Length prediction curve for the next 4 hours; Warning information: Color-coded warning status; System status: Indicators of the operating status of each module; Communication interfaces: PROFINET: integrated with DCS system; OPC UA: data upload to MES system; 4-20mA analog output: backup output channel; RS-485: connection to local display device.
[0085] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. A method for measuring the length of the working end of an electrode in a calcium carbide furnace, characterized in that: Includes the following steps: Data acquisition: Real-time acquisition of the total weight of the holder, the height of the paste column, the furnace temperature, and the weight of the holder itself; Coefficient determination: Call the set paste column weight coefficient and electrode weight coefficient; Length calculation: Based on the total weight, paste column height, paste column weight coefficient, holder weight and electrode weight coefficient, a calculation model is established, and the length of the working end of the electrode is calculated through the calculation model.
2. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to claim 1, characterized in that: The coefficient determination step also includes establishing a relational database containing measured density values of electrode paste at different temperature ranges. The set paste column weight coefficient and electrode weight coefficient are retrieved from this relational database, as follows: The real-time collected furnace temperature is matched and interpolated with the database to output the paste column density value and electrode density value at the current temperature, and then converted into the paste column weight coefficient and electrode weight coefficient.
3. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to claim 2, characterized in that: The establishment of the relational database also establishes a correspondence between the height of the glue column and the weight coefficient of the glue column, enabling the weight coefficient of the glue column to be adjusted according to the real-time height of the glue column.
4. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to claim 3, characterized in that: The relationship between the height of the glue column and the weight coefficient of the glue column is as follows: the weight coefficient of the glue column increases with the increase of the height of the glue column and decreases with the decrease of the height of the glue column, with an increase or decrease of 5% per meter.
5. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to claim 4, characterized in that: The relationship between the height of the glue column and the weight coefficient of the glue column is established by assigning different weight coefficients to historical data for different height ranges, specifically including: Weighting steps: When establishing the relational database, the measured data collected within the range of 4.0-4.2 meters in height of the glued column are assigned the highest weight; Weighted calculation steps: When calling the weight coefficient of the glue column, the system prioritizes matching the coefficient value corresponding to the high weight interval based on the real-time glue column height. If it is at the interval boundary, the weighted average algorithm is used to calculate the final coefficient.
6. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to any one of claims 1-5, characterized in that: It also includes a prediction step, which includes a data preprocessing step, a rate of change calculation step, a furnace temperature trend analysis step, a prediction model establishment step, and a prediction result output step. Data preprocessing: Based on the continuously collected column height data within a historical time period, outliers are removed, and the data sequence is smoothed using the moving average method; Rate of change calculation: Time series analysis is performed on the smoothed column height data series to calculate the rate of change of the column height. The rate of change is obtained by linear regression or difference method, and the unit is meters per hour. Furnace temperature trend analysis: Real-time collection of furnace temperature data, combined with historical furnace temperature data, and the use of trend fitting methods to predict the furnace temperature change trend within a set future time period; Prediction model establishment: Using the rate of change and furnace temperature trend as input variables, establish a multiple linear regression model or a time series prediction model to output the predicted value of the change in electrode working end length within a set future time period; Prediction output: The predicted values are integrated into the submerged arc furnace control system for real-time adjustment of electrode operating parameters.
7. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to claim 6, characterized in that: It also includes an early warning step, which specifically includes: Set a minimum value for the working end length of an electrode, and compare the predicted value obtained in the prediction step with the minimum value; If the predicted value is greater than the minimum value, calculate the critical time for the predicted value to drop to the minimum value, and initiate different levels of tiered early warning based on the length of the critical time. If the predicted value is less than or equal to the minimum value, the highest level of emergency warning will be triggered immediately.
8. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to any one of claims 1-5, characterized in that: It also includes a calibration step, which includes a periodic data acquisition step, a data fitting step, a coefficient optimization step, and a verification step. Regular data collection: Regularly collect multiple sets of total weight data of the holder under different working conditions, including different paste column heights, different furnace temperatures, and different electrode consumption stages; Data fitting: The least squares method or nonlinear regression algorithm is used to fit the collected total weight data of the gripper with the measured working end length of the electrode, and the paste column weight coefficient and electrode weight coefficient are solved in reverse. Coefficient optimization: The fitted coefficients are compared with the original coefficients, the coefficient values are automatically adjusted according to the error threshold, and the results are updated in the relational database. Verification and iteration: Through three batches of process verification, ensure that the error of the optimized coefficients is stable within ±1% in actual measurement; otherwise, repeat the periodic data collection and data fitting steps.
9. The method for measuring the length of the working end of an electrode in a calcium carbide furnace according to any one of claims 1-5, characterized in that: In the coefficient determination step, an attenuation factor based on the electrode consumption stage is introduced into the electrode weight coefficient, specifically including: Phase division: Based on the cumulative usage time or cumulative pressure discharge of the electrode, the electrode life cycle is divided into the initial stage, the stable stage, and the later stage; Coefficient decay: In the later stages of electrode use, the electrode weight coefficient is multiplied by a decay factor less than 1 to reflect the slight decrease in average density at the bottom of the electrode due to excessive sintering and erosion.
10. An apparatus, characterized in that: Includes the following modules: The weighing module is used to collect the total weight of the electrode holder in real time. The height detection module is used to collect the height of the glue column in real time; The data processing module has its input end connected to the output end of the weighing module and the output end of the height detection module, and is configured to execute the method for measuring the length of the working end of the calcium carbide furnace electrode as described in any one of claims 1-9, and to calculate the length of the working end of the electrode; The output module has its input end connected to the output end of the data processing module and is used to output and display the length of the working end.