A smart control system for electrode melting rate based on start-stop control

Abstract

This invention discloses an intelligent control system for electrode melting rate based on start-stop control, belonging to the field of electrode melting rate control technology. First, based on quality constraint analysis, this invention obtains candidate melting rate ranges for the electrode. Then, based on environmental constraints, it filters these candidate melting rate ranges to obtain the optimal melting rate range for the electrode. Finally, based on the optimal melting rate range, it analyzes the electrode deviation correction rate. During deviation correction, the deviation correction time is compared with the time interval between time points to determine whether to perform deviation correction or stop the operation. This invention not only avoids the limitations of a single quality constraint or a single environmental constraint, but also avoids the target weakening problem caused by traditional weight settings. Furthermore, it avoids steel quality fluctuations caused by incomplete correction and continued deviation from the range during deviation correction, ensuring product quality stability.

Description

Technical Field

[0001] This invention relates to the field of electrode melting rate control technology, and more specifically to an intelligent control system for electrode melting rate based on start-stop control. Background Technology

[0002] In the smelting process of metal materials, electrodes generate electric arcs or resistance heat by passing electricity through them, converting electrical energy into heat energy, so that the metal material reaches the melting temperature and forms a uniform molten steel. The melting rate of the electrodes directly affects the degree of melting of the metal raw materials, the composition of the molten steel, and the temperature. Controlling the melting rate of the electrodes can not only ensure the quality of metal products, but also reduce equipment safety risks. Therefore, controlling the melting rate of the electrodes is crucial.

[0003] Existing technologies for controlling electrode melting rate have some shortcomings, specifically in the following aspects: Most existing technologies correlate electrode melting rate with product quality or equipment safety, and then control the electrode melting rate accordingly. However, they do not consider that the total amount of harmful substances generated varies greatly when the electrode melting rate is different. Therefore, the current technology is not the best for controlling electrode melting rate. At the same time, the current technology does not consider the impact of deviation adjustment rate and deviation adjustment time on product quality when adjusting the deviation of electrode melting rate, making it difficult to adapt to melting scenarios with high product requirements. Summary of the Invention

[0004] To address the aforementioned technical shortcomings, the present invention aims to provide an intelligent control system for electrode melting rate based on start-stop control.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides an intelligent control system for electrode melting rate based on start-stop control, including: an information acquisition module: used to acquire equipment information, process conditions, environmental information and production information in the target factory from the management center.

[0006] Melting rate analysis module: Used to perform melting tests on various raw steel materials of the target plant, and train an electrode melting rate decision model based on the test results. Then, analyze and obtain the electrode candidate melting rate range corresponding to the target raw steel. Next, conduct environmental protection tests on the target raw steel based on the electrode candidate melting rate range to obtain the optimal melting rate range of the electrode corresponding to the target raw steel. Finally, obtain the optimal deviation correction rate of the target raw steel based on the optimal melting rate range test.

[0007] Detection module: Used to monitor the electrode melting rate in real time through sensing devices during the electrode melting process.

[0008] Control module: Used to acquire electrode melting rate data at several consecutive historical time points, then predict the electrode melting rate at the next time point, and send control commands to the execution terminal based on the prediction results.

[0009] Execution module: Used to receive instructions from the control module and execute corresponding control operations.

[0010] The beneficial effects of this invention are as follows: 1. This invention first trains a decision model for the electrode melting rate of the target worker based on various information of the target factory, and then analyzes and obtains the electrode melting rate range that is suitable for production requirements. Then, based on environmental protection requirements, the electrode melting rate range that is suitable for quality requirements is screened to obtain the optimal melting rate range of the electrode. This avoids the limitations caused by a single quality constraint or a single environmental protection requirement constraint, and also avoids the problem of neglecting one aspect for another caused by traditional weight setting. At the same time, it ensures the unique priority of quality constraints. Finally, based on the optimal melting rate range, the optimal deviation correction rate is obtained through targeted analysis and is specifically adapted to the optimal melting rate range. This improves the correction accuracy by nearly 90% and avoids the production risks such as electrode wear caused by a universal correction rate.

[0011] 2. This invention obtains the correction time for melting rate deviation by analyzing the deviation correction rate. By comparing the correction time with the time interval, it determines whether to perform correction or stop the operation, thus avoiding the risk of ineffective correction. When the correction time is insufficient, forcibly adjusting the electrode rate may cause system oscillation and electrode impact, and aggravate equipment wear or failure. In this case, the machine is stopped directly, which effectively avoids such mechanical wear and extends the service life of the equipment. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Reference Figure 1 As shown, the first aspect of the present invention provides an intelligent control system for electrode melting rate based on start-stop control, comprising: an information acquisition module for acquiring equipment information, process conditions, environmental information and production information within the target factory from the management center.

[0016] In a specific example, the equipment information includes smelting equipment information and auxiliary equipment information; the process conditions include the melting process type and slag characteristics; the environmental information includes physical environmental parameters and environmental interference parameters; and the production information includes raw material steel information and production plan information.

[0017] It should be noted that the smelting equipment information includes the dimensions, capacity, and operating parameters of the electric arc furnace and electroslag furnace in the target plant, the type, material, and dimensions of the electrodes used in the target plant, and the operating parameters of the power supply equipment; the auxiliary equipment information includes the models and parameters of cooling equipment, ventilation equipment, and monitoring equipment; the physical environment parameters include the temperature range, humidity range, air pressure range, and dust concentration range in the workshop; the environmental interference parameters include the voltage fluctuation interference range and the vibration interference range generated by adjacent equipment; the raw material steel information includes the raw material steel type and the dimensions and weight of the steel ingots; and the production plan information includes historical production plan information and projected production plan information.

[0018] It should be noted that the production plan information includes the dimensions, shape, weight and permissible deviations, technical requirements, test methods, inspection rules, packaging, marking and quality certificate requirements of the target mold steel to be produced.

[0019] Melting rate analysis module: Used to perform melting tests on various raw steel materials of the target plant, and train an electrode melting rate decision model based on the test results. Then, analyze and obtain the electrode candidate melting rate range corresponding to the target raw steel. Next, conduct environmental protection tests on the target raw steel based on the electrode candidate melting rate range to obtain the optimal melting rate range of the electrode corresponding to the target raw steel. Finally, obtain the optimal deviation correction rate of the target raw steel based on the optimal melting rate range test.

[0020] In a specific example, the melting test of each raw steel in the target factory is carried out as follows: Several raw steels of any type are obtained and designated as test raw steels. Corresponding consumable electrodes are obtained according to the type of raw steel and designated as test electrodes. Based on historical production information, the equipment information, process conditions, and environmental information of the raw steel type during historical production are obtained. Based on this, the test conditions for each test raw steel are set and kept constant. At the same time, the melting rate is set according to a preset gradient. Each test raw steel is melted at each melting rate. After a preset test duration, the quality data of the corresponding molten steel sample at each melting rate is detected. The raw steel type, test conditions, melting rates, and the quality data of the corresponding molten steel sample at each melting rate are collectively recorded as test data.

[0021] It should be noted that the tests for each melting rate were repeated three times, and the test data were the average of the three repeated tests.

[0022] It should be noted that the quality data includes the chemical composition content, non-metallic inclusion qualification level, grain size, microstructure qualification level, ultrasonic testing results, and number of surface defects in the mold steel formed by casting the molten steel sample.

[0023] It should be noted that the above testing methods can be used to test various types of raw steel in the target factory.

[0024] In a specific example, the electrode melting rate decision model trained based on test results is trained as follows: First, the test data are classified and coded according to the type of raw steel. A three-stage architecture consisting of a feature extraction layer, a target prediction layer, and a rate decision layer is used as the basic architecture of the electrode melting rate decision model. A two-layer fully connected network is set in the feature extraction layer to capture the nonlinear correlation between the test data. An output branch is set in the target prediction layer to output quality data. The rate decision layer is used to filter rate intervals that match the set quality data intervals, which are then used as the output electrode candidate melting rate intervals. After preprocessing the test data, the basic architecture of the electrode melting rate decision model is trained and validated to obtain the electrode melting rate decision model for the target plant.

[0025] It should be noted that the quality data range is the critical value for assessing whether the quality of steel meets the standard. The higher the requirement for the quality data range, the higher the quality requirement for the steel. The specific value of the quality data range is obtained from the steel production requirements.

[0026] It should be noted that preprocessing includes outlier removal, feature selection, data normalization, and dataset partitioning. Outlier removal, feature selection, data normalization, and dataset partitioning are all existing technologies and will not be elaborated here.

[0027] In a specific example, the analysis obtains the electrode candidate melting rate range corresponding to the target raw material steel. The specific process is as follows: obtain the current production plan based on the production information, and obtain the model and production requirements of the target raw material steel corresponding to the target mold steel based on the current production plan. Input the model and production requirements into the electrode melting rate decision model, and output the electrode candidate melting rate range of the target raw material steel through the electrode melting rate decision model.

[0028] In a specific example, environmental protection tests are conducted on the target raw steel based on the alternative melting rate ranges of the electrodes to obtain the optimal melting rate range of the electrodes corresponding to the target raw steel. The specific process is as follows: Several identical raw steels are selected based on the model of the target raw steel and denoted as each test raw steel. Each test rate is set according to a preset gradient based on the alternative melting rate range corresponding to the target raw steel. Each test raw steel is melted according to each test rate until it is melted into molten steel. The total amount of harmful substances generated at each test rate is detected. The test rates with the lowest total amount of harmful substances are combined to obtain the optimal melting rate range of the electrodes corresponding to the target raw steel.

[0029] It should be noted that the specific gradient value of the preset gradient is set by the relevant personnel themselves, and no specific restrictions are imposed here. The smaller the preset gradient, the more accurate the optimal melting rate range obtained from the test will be.

[0030] It should be noted that when the total amount of harmful substances is at its lowest and there is only one test rate, the gradient is set with that test rate as the midpoint, and then each test rate is obtained again, and each test raw material steel is tested again until the optimal melting rate range of the electrode is obtained.

[0031] In a specific example, the optimal deviation correction rate for the target raw steel is obtained based on the optimal melting rate range test. The specific analysis process is as follows: Select various deviation test raw steels that are exactly the same as the target raw steel, and divide each deviation test raw steel into control groups. Each control group contains the same number of deviation test raw steels. According to the optimal melting rate range corresponding to the target raw steel, set the deviation correction rate corresponding to each melting rate deviation value according to the preset gradient. The melting rate deviation values ​​of each deviation test raw steel in the same control group are the same. Perform deviation correction on each deviation test raw steel in each control group according to each deviation correction rate. After the correction is completed, detect the index information of the molten steel sample corresponding to each deviation correction rate, and obtain the average value of each deviation correction rate corresponding to the molten steel sample that best meets the standard index information in each control group. This value is recorded as the optimal deviation correction rate for the target raw steel.

[0032] It should be noted that the indicator information includes melting uniformity, composition consistency and grain refinement, etc., and the standard indicator information refers to the various quality data requirements specified in the production requirements of the target raw material steel.

[0033] Detection module: Used to monitor the electrode melting rate in real time through sensing devices during the electrode melting process.

[0034] It should be noted that the initial weight of the electrode is measured by a pressure sensor, and the remaining weight of the electrode is collected in real time during the electrode's operation. The melting time is also recorded. The melting rate of the electrode at the current time point is calculated by using the initial weight, remaining weight, operating time, and melting rate at each historical time point.

[0035] Control module: Used to acquire electrode melting rate data at several consecutive historical time points, then predict the electrode melting rate at the next time point, and send control commands to the execution terminal based on the prediction results.

[0036] In a specific example, electrode melting rate data from several consecutive historical time points are obtained, and then the electrode melting rate at the next time point is predicted. The specific process is as follows: the electrode melting rate monitored at several consecutive historical time points is obtained from the data center, the changing characteristics of the electrode melting rate are captured by the time series prediction algorithm, and then the melting rate of the electrode at the next moment is predicted.

[0037] It should be noted that the historical consecutive time points refer to the consecutive time points before the current time point, including the current time point.

[0038] It should be noted that time series prediction algorithms include LSTM, ARIMA, and XGBoost, among which LSTM, ARIMA, and XGBoost are existing algorithm technologies, so they will not be described in detail.

[0039] For example, after preprocessing the electrode melting rate data within several consecutive historical time points, the data is input into the LSTM model. The model parameters are adjusted using RMSE and MAE as indicators, and the model is trained and optimized accordingly. The model parameters include the number of hidden layer nodes of the LSTM. After calculation, the model directly outputs the predicted value of the electrode melting rate for the next time point. For each new actual monitoring value, the model is incrementally trained to ensure the timeliness of the prediction.

[0040] In a specific example, control commands are sent to the execution terminal based on the prediction results. The specific process is as follows: The melting rate of the electrode at the next time point is compared with the optimal melting rate range and the alternative melting rate range of the electrode. If the electrode melting rate at the next time point belongs to the optimal melting rate range, control command 00 is sent to the execution terminal. If the electrode melting rate at the next time point does not belong to the alternative melting rate range, control command 11 is sent to the execution terminal. If the electrode melting rate at the next time point does not belong to the optimal melting rate range but belongs to the alternative melting rate range, the deviation correction time is calculated based on the optimal deviation correction rate and the deviation amount at the next time point. The deviation correction time is compared with the time interval. If the deviation correction time is less than or equal to the time interval, control command 010 is sent to the execution terminal; otherwise, command 011 is sent to the execution terminal.

[0041] Execution terminal: Used to receive instructions from the control module and execute corresponding control operations.

[0042] In a specific instance, the control module receives control commands and executes corresponding control operations. The specific process is as follows: when the execution terminal receives control command 00, no operation is performed.

[0043] When the execution terminal receives control command 11, it performs a stop operation.

[0044] When the execution terminal receives control command 010, it performs deviation correction according to the optimal deviation correction rate; when the execution terminal receives control command 011, it performs a stop operation.

[0045] It should be noted that when the predicted electrode melting rate at the next time point deviates, the deviation is corrected immediately at the current time point until the predicted electrode melting rate at the next time point does not deviate.

[0046] It should be noted that the present invention obtains the correction time for melting rate deviation by analyzing the deviation correction rate. When the correction time is longer than the time interval, it means that the deviation correction cannot be completed before the next monitoring. At this time, the stop operation is performed to avoid the fluctuation of molten steel quality caused by incomplete correction and deviation from the range. At the same time, it reduces the ineffective resource consumption such as power and electrode wear during the correction process.

[0047] It should be noted that after the stop operation is performed, relevant personnel will check the cause of the anomaly, including the accuracy of the prediction model, electrode condition, raw material characteristics and equipment failure, and then formulate a targeted handling plan.

[0048] This invention provides an intelligent control system for electrode melting rate based on start-stop control. First, candidate melting rate ranges for the electrode are obtained based on quality constraint analysis. Then, these ranges are screened based on environmental constraints to obtain the optimal melting rate range for the electrode. Finally, the deviation correction rate for the electrode is obtained based on analysis of the optimal melting rate range. During deviation correction, the deviation correction time is compared with the time interval between time points to determine whether to perform deviation correction or stop the operation. This invention not only avoids the limitations of a single quality constraint or a single environmental constraint, but also avoids the target weakening problem caused by traditional weight settings. Furthermore, it avoids steel quality fluctuations caused by incomplete correction and deviation from the range during deviation correction, ensuring product quality stability.

[0049] The examples described in this invention are not limited to the specific embodiments listed above. These examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications or equivalent substitutions made within the spirit and principles of this invention should be included within the scope of protection.

[0050] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A start-stop control based intelligent control system for electrode melting rate, characterized in that, Comprise the following modules: Information acquisition module: for obtaining equipment information, process conditions, environmental information and production information in the target factory from the management center; Melting rate analysis module: for performing melting tests on each raw steel of the target factory, and training an electrode melting rate decision model according to the test results, and then analyzing to obtain the electrode candidate melting rate interval corresponding to the target raw steel, and then performing environmental test on the target raw steel according to the electrode candidate melting rate interval, obtaining the electrode optimal melting rate interval corresponding to the target raw steel, and finally testing the optimal deviation correction rate of the target raw steel according to the optimal melting rate interval; Detection module: for real-time monitoring of electrode melting rate during electrode melting process through sensing equipment; Control module: for obtaining electrode melting rate data at a plurality of consecutive time points in history, and then predicting the electrode melting rate at the next time point, and sending control instructions to the execution terminal according to the prediction results; The specific process of sending control instructions to the execution terminal according to the prediction results is as follows: The melting rate of the electrode at the next time point is compared with the optimal melting rate interval and the candidate melting rate interval of the electrode. When the electrode melting rate at the next time point belongs to the optimal melting rate interval, control instruction 00 is sent to the execution terminal. When the electrode melting rate at the next time point does not belong to the candidate melting rate interval, control instruction 11 is sent to the execution terminal. When the electrode melting rate at the next time point does not belong to the optimal melting rate interval but belongs to the candidate melting rate interval, the deviation correction time is calculated according to the optimal deviation correction rate and the deviation amount at the next time point. The deviation correction time is compared with the time point interval time. If the deviation correction time is less than or equal to the time point interval time, control instruction 010 is sent to the execution terminal. Otherwise, control instruction 011 is sent to the execution terminal; Execution terminal: for receiving the control instructions of the control module and executing the corresponding control operation; The specific process of receiving the control instructions of the control module and executing the corresponding control operation is as follows: When the execution terminal receives control instruction 00, no operation is performed; When the execution terminal receives control instruction 11, stop operation is performed; When the execution terminal receives control instruction 010, deviation correction is performed according to the optimal deviation correction rate. When the execution terminal receives control instruction 011, stop operation is performed.

2. The intelligent control system for electrode melting rate based on start-stop control according to claim 1, characterized in that, The equipment information includes smelting equipment information and auxiliary equipment information; the process conditions include melting process type and slag characteristics; the environmental information includes physical environmental parameters and environmental interference parameters; the production information includes raw steel information and production plan information.

3. The intelligent electrode melting rate control system based on start-stop control according to claim 2, characterized in that, The specific test process of the melting test on each raw steel of the target factory is as follows: Obtain several raw steels of any type, denoted as test raw steels, and obtain corresponding consumable electrodes according to the type of the raw steels, denoted as test electrodes, obtain the equipment information, process conditions and environmental information of the historical production of the type of raw steel according to the historical production information, set the test conditions of each test raw steel according to the information, and keep the test conditions of each test raw steel unchanged, and set the melting rate according to the preset gradient, melt each test raw steel according to each melting rate, test for a preset time, detect the quality data of the corresponding liquid steel sample under each melting rate, and comprehensively record the type of raw steel, test conditions, each melting rate and the quality data of the corresponding liquid steel sample under each melting rate as test data.

4. The intelligent electrode melting rate control system based on start-stop control according to claim 3, characterized in that, The electrode melting rate decision model is trained according to the test results, and the specific training process is as follows: First, each test data is classified and coded according to the type of raw steel, and a three-section architecture of feature extraction layer, target prediction layer and rate decision layer is used as the basic architecture of the electrode melting rate decision model, 2 layers of full connection network are set in the feature extraction layer to capture the nonlinear correlation between the test data, an output branch is set in the target prediction layer to output the quality data, and the rate decision layer is used to select the rate interval of the quality data that meets the set quality data interval as the output electrode candidate melting rate interval, and the basic architecture of the electrode melting rate decision model is trained and verified after the test data is preprocessed, and the electrode melting rate decision model of the target factory is obtained.

5. The intelligent electrode melting rate control system based on start-stop control according to claim 4, characterized in that, The electrode candidate melting rate interval corresponding to the target raw steel is obtained by analyzing, and the specific process is as follows: According to the production information, obtain the current production plan, and obtain the type and production requirements of the target raw steel corresponding to the target die steel according to the current production plan, input the type and production requirements into the electrode melting rate decision model, and output the electrode candidate melting rate interval of the target raw steel through the electrode melting rate decision model.

6. The intelligent electrode melting rate control system based on start-stop control according to claim 5, wherein, The target raw steel is tested according to the electrode candidate melting rate interval to obtain the electrode best melting rate interval corresponding to the target raw steel, and the specific process is as follows: Based on the type of the target raw steel, several completely same raw steels are selected, denoted as test raw steels, and the test rate is set according to the preset gradient according to the candidate melting rate interval corresponding to the target raw steel, and the test raw steels are melted according to the test rate, until the test raw steels are melted into liquid steel, the total amount of harmful substances generated by each test rate is detected, and the test rate with the lowest total amount of harmful substances is combined to obtain the electrode best melting rate interval corresponding to the target raw steel.

7. The intelligent electrode melting rate control system based on start-stop control according to claim 6, characterized in that, The best deviation correction rate of the target raw steel is tested according to the best melting rate interval, and the specific analysis process is as follows: The same deviation test raw material steel as the target raw material steel is selected, and the deviation test raw material steels are divided into control groups, and the same number of deviation test raw material steels exist in each control group; according to the best melting rate interval corresponding to the target raw material steel, the deviation correction rate corresponding to each melting rate deviation value is set according to the preset gradient, wherein the melting rate deviation values of the deviation test raw material steels in the same control group are the same, the deviation test raw material steels in each control group are corrected according to the deviation correction rate, after the correction is completed, the index information of the liquid steel sample corresponding to each deviation correction rate is detected, and the mean value of the deviation correction rate corresponding to the liquid steel sample most consistent with the standard index information in each control group is obtained, which is recorded as the best deviation correction rate of the target raw material steel.

8. The intelligent electrode melting rate control system based on start-stop control according to claim 7, characterized in that, The electrode melting rate data at the historical several continuous time points is obtained, and then the electrode melting rate at the next time point is predicted, and the specific process is as follows: The electrode melting rate monitored at the historical several continuous time points is obtained from the data center, the change characteristics of the electrode melting rate are captured through the time series prediction algorithm, and then the melting rate of the electrode at the next time is predicted.

Images