Blast furnace pressure difference control method and device, electronic equipment and storage medium

Abstract

According to the blast furnace pressure difference control method and device, the electronic equipment and the storage medium, the current blast furnace pressure difference value can be predicted more accurately through the blast furnace pressure difference prediction model trained in advance or in real time, and compared with a traditional mode depending on artificial experience, the neural network model is trained through a large amount of data, so that the blast furnace pressure difference value is more accurately predicted. Compared with the prior art, more influence factors can be comprehensively considered, the prediction precision and reliability can be improved, when the predicted pressure difference value exceeds the preset safety threshold value, the system can automatically give an alarm prompt to remind an operator to take measures in time, such as gradually reducing the air volume so as to prevent possible abnormal furnace conditions or safety accidents, so that the safety of blast furnace operation is enhanced, and the safety of blast furnace operation is improved. And for the condition that the predicted pressure difference value is within the safety range, the optimal air volume adjustment amplitude suggestion can be automatically given according to the air volume adjustment data table, stable operation of the blast furnace is guaranteed, fine adjustment of the air volume is achieved, the stability of the pressure difference of the blast furnace is maintained, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management, and in particular to a blast furnace pressure difference control method, device, electronic equipment and storage medium. Background Art

[0002] As the core link in the steel production process, the blast furnace not only bears the heavy responsibility of reducing iron ore to molten iron, but also is a complex system that highly integrates multiple physical and chemical reactions. In this closed high-temperature environment, raw materials (such as iron ore, coke and flux) undergo a series of pyrolysis, melting, reduction and other processes during the descent process, accompanied by the generation and flow of gas, forming a unique state of coexistence of gas, liquid and solid phases. This series of reactions requires not only precise temperature control and raw material ratio, but also depends on the reasonable distribution of airflow in the furnace to ensure the effective transmission of heat and materials. The blast furnace pressure difference, as one of the important indicators for monitoring furnace conditions, is directly related to the smoothness of the descent of the charge, the uniformity of the distribution of the gas flow and the quality of the air permeability in the furnace. The stability of the pressure difference directly affects the production efficiency, fuel utilization and life of the blast furnace. Therefore, maintaining an appropriate pressure difference range is of vital significance to ensure the stable operation of the blast furnace and improve production efficiency.

[0003] At present, the adjustment of blast furnace pressure difference mainly depends on the operator's experience and judgment and the feedback of real-time monitoring data. The operator observes the changing trend of the pressure difference, combines other parameters such as furnace temperature, furnace top gas composition, air volume, etc., and adjusts the air volume, tuyere layout or air inlet temperature of the blower in a timely manner based on the long-term accumulated experience, in order to optimize the air flow distribution in the furnace and maintain the pressure difference stable. However, although this experience-based adjustment method can effectively cope with conventional fluctuations to a certain extent, it may be difficult to make the optimal adjustment quickly when faced with emergencies or complex changes in furnace conditions. Summary of the invention

[0004] The present application provides a blast furnace pressure difference control method, device, electronic equipment and storage medium to solve the above-mentioned technical problems.

[0005] A blast furnace pressure difference control method provided in an embodiment of the present application comprises: obtaining blast furnace parameters to be detected; inputting the blast furnace parameters to be detected into a blast furnace pressure difference prediction model trained in advance or in real time to obtain a current predicted pressure difference value; if the current predicted pressure difference value is greater than the pressure difference upper limit threshold, issuing an alarm prompt message, and gradually reducing the initial air volume in response to the alarm prompt message until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold; if the current predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, performing a search operation in an air volume adjustment data table according to the predicted pressure difference value to obtain an optimal air volume adjustment range, and adjusting the air volume based on the optimal air volume adjustment range to control the blast furnace pressure difference; wherein the air volume adjustment data table stores a correspondence between the predicted pressure difference value and the optimal air volume adjustment range, and the air volume adjustment data table is stored in an operation suggestion module.

[0006] In one embodiment of the present application, the training process of the blast furnace pressure difference prediction model includes: obtaining key parameters of the blast furnace, wherein the key parameters of the blast furnace include airflow state parameters, airflow operation parameters, blast furnace pressure difference parameters and furnace condition marking parameters; performing time series feature extraction on the key parameters of the blast furnace to obtain a training data set; and training the initial blast furnace pressure difference prediction model through the training data set to obtain a blast furnace pressure difference prediction model.

[0007] In one embodiment of the present application, key parameters of a blast furnace are obtained, including: collecting blast furnace airflow state parameters, the blast furnace airflow state parameters include furnace body static pressure, wall temperature, throat cross temperature measurement and water temperature difference; collecting blast furnace airflow operation parameters, the blast furnace airflow operation parameters include air volume and oxygen content; collecting blast furnace pressure difference parameters, the blast furnace pressure difference parameters include wind pressure, top pressure and pressure difference parameters, the pressure difference parameters are the difference between wind pressure and top pressure; collecting furnace condition marking parameters, the furnace condition marking parameters include furnace change signal and wind stop signal, and assigning marking values ​​to the furnace condition marking parameters, the furnace condition marking parameter marking values ​​are used to indicate special situations encountered by the blast furnace during operation, and the furnace condition marking parameter marking values ​​include unavailable thresholds; the blast furnace airflow state parameters, the blast furnace airflow operation parameters, the blast furnace pressure difference parameters and the furnace condition marking parameters are stored in a blast furnace parameter database.

[0008] In one embodiment of the present application, time series feature extraction is performed on the blast furnace key parameters to obtain a training data set, including: collecting prediction points according to a preset time window; extracting the blast furnace key parameters within a preset time period before the collection prediction points, recording them as the blast furnace key parameters to be processed, and preprocessing the data of the blast furnace key parameters to be processed to obtain a feature vector; setting a corresponding blast furnace pressure difference parameter for the feature vector as the blast furnace pressure difference parameter of the prediction point, and determining the feature vector and the corresponding blast furnace pressure difference parameter as training samples; traversing the blast furnace key parameters to generate multiple training samples, and combining all training samples into the training data set, which will be used to train a blast furnace pressure difference prediction model.

[0009] In one embodiment of the present application, data of the to-be-processed blast furnace key parameters are preprocessed to obtain a feature vector, including: extracting the timestamp data of the furnace condition marking parameter, recorded as the furnace condition marking timestamp data, extracting the timestamp data of the blast furnace airflow state parameter, recorded as the airflow state timestamp data, extracting the timestamp data of the blast furnace airflow operation parameter, recorded as the airflow operation timestamp data; screening out the same timestamp data from the furnace condition marking timestamp data, the airflow state timestamp data and the airflow operation timestamp data, and recording them as the same timestamp data; determining the furnace condition marking parameters, blast furnace airflow state parameters and blast furnace airflow operation parameters corresponding to the same timestamp data as an associated blast furnace key parameter group; extracting the furnace condition marking parameter mark value in the associated blast furnace key parameter group, if the furnace condition marking parameter mark value is an unavailable threshold, marking other parameters in the associated blast furnace key parameter group as null values; using linear interpolation to complete the associated blast furnace key parameter group corresponding to the null value to obtain a feature vector.

[0010] In one embodiment of the present application, the air volume adjustment data table includes: collecting historical data, the historical data including multiple sample pressure difference values, setting multiple sample air volume adjustment values ​​for each sample pressure difference value, and obtaining multiple groups of adjustment data; verifying the adjustment result of each group of adjustment parameters, and performing efficiency scoring on each group of adjustment results; extracting the adjustment parameter with the highest efficiency score of the adjustment result corresponding to each sample pressure difference value, and recording it as the optimal adjustment parameter; integrating the optimal adjustment parameters corresponding to each sample pressure difference value to obtain the air volume adjustment data table.

[0011] In one embodiment of the present application, the initial air volume is gradually reduced in response to the alarm prompt information until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, including: obtaining the initial air volume adjustment value corresponding to the current predicted pressure difference value, subtracting the preset air volume threshold from the initial air volume adjustment value to obtain the next air volume value; obtaining new blast furnace parameters to be detected based on the next air volume value, and inputting the new blast furnace parameters to be detected into a pre- or real-time trained blast furnace pressure difference prediction model until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold.

[0012] The blast furnace pressure difference control device provided in the embodiment of the present application comprises: an acquisition module for acquiring the blast furnace parameters to be detected; a pressure difference prediction module for inputting the blast furnace parameters to be detected into a blast furnace pressure difference prediction model trained in advance or in real time to obtain a current predicted pressure difference value; an alarm module for issuing an alarm prompt message if the current predicted pressure difference value is greater than the pressure difference upper limit threshold, and gradually reducing the initial air volume in response to the alarm prompt message until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold; an air volume adjustment module for performing a search operation in an air volume adjustment data table according to the predicted pressure difference value if the current predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, and obtaining an optimal air volume adjustment range, and adjusting the air volume based on the optimal air volume adjustment range to control the blast furnace pressure difference; wherein the air volume adjustment data table stores the corresponding relationship between the predicted pressure difference value and the optimal air volume adjustment range, and the air volume adjustment data table is stored in the operation suggestion module.

[0013] An embodiment of the present application provides an electronic device, including a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the above-mentioned blast furnace pressure difference control method.

[0014] The beneficial effects of the present application are as follows: the present invention can more accurately predict the current blast furnace pressure difference value through a blast furnace pressure difference prediction model trained in advance or in real time. Compared with the traditional method relying on manual experience, the neural network model is trained with a large amount of data and can comprehensively consider more influencing factors to improve the accuracy and reliability of the prediction. When the predicted pressure difference value exceeds the preset safety threshold, the system can automatically issue an alarm prompt to promptly remind the operator to take measures, such as gradually reducing the air volume, to prevent possible abnormal furnace conditions or safety accidents, thereby enhancing the safety of blast furnace operation. For the situation where the predicted pressure difference value is within the safety range, the optimal air volume adjustment range recommendation can be automatically given according to the air volume adjustment data table. By integrating advanced prediction models and intelligent decision support, the accuracy and efficiency of blast furnace pressure difference control can be significantly improved, the stable operation of the blast furnace can be ensured, the fine adjustment of the air volume can be achieved, the stability of the blast furnace pressure difference can be maintained, and production efficiency can be improved.

[0015] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0017] Figure 1 is a blast furnace pressure difference control system architecture diagram shown in an exemplary embodiment of the present application;

[0018] Figure 2 is a flow chart of a blast furnace pressure difference control method shown in an exemplary embodiment of the present application;

[0019] Figure 3 is a block diagram of a blast furnace pressure difference control device shown in an exemplary embodiment of the present application;

[0020] Figure 4 It is a structural schematic diagram of an electronic device shown in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0021] The following will describe the implementation methods of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, not for limiting the scope of protection of the present application.

[0022] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0023] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0024] The embodiments of the present application respectively propose a blast furnace pressure difference control method, a blast furnace pressure difference control device, and an electronic device, and these embodiments will be described in detail below.

[0025] See also Figure 1 , Figure 1 The blast furnace pressure difference control system architecture diagram according to an embodiment of the present application is shown, including a data acquisition module 110, a data processing module 120, an offline modeling module 130, a real-time prediction module 140 and an operation suggestion module 150. The key parameters of the blast furnace are collected and stored through the data acquisition module, and the training samples of the key parameters of the blast furnace are extracted through the data processing module to generate multiple training samples. All the training samples are combined into a training data set. The training data set will be used to train the blast furnace pressure difference prediction model. The blast furnace pressure difference prediction model is established based on the key parameters of the blast furnace through the offline modeling module, and the model is stored. The blast furnace pressure difference prediction model is deployed in the real-time prediction module. In the real-time prediction module, the blast furnace parameters to be detected are input into the blast furnace pressure difference prediction model trained in advance or in real time to obtain the current predicted pressure difference value. If the current predicted pressure difference value is greater than the pressure difference upper limit threshold, an alarm prompt message is issued. In response to the alarm prompt message, the initial air volume is gradually reduced until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold. If the current predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, then in the operation suggestion module, a search operation is performed in the air volume adjustment data table according to the predicted pressure difference value to obtain the optimal air volume adjustment range, and the air volume is adjusted based on the optimal air volume adjustment range to control the blast furnace pressure difference. The blast furnace pressure difference prediction model trained in advance or in real time can more accurately predict the current blast furnace pressure difference value. Compared with the traditional method that relies on manual experience, the neural network model has been trained with a large amount of data and can comprehensively consider more influencing factors to improve the accuracy and reliability of the prediction. When the predicted pressure difference value exceeds the preset safety threshold, the system can automatically issue an alarm prompt to promptly remind the operator to take measures, such as gradually reducing the air volume, to prevent possible abnormal furnace conditions or safety accidents, thereby enhancing the safety of blast furnace operation. For situations where the predicted pressure difference value is within a safe range, the optimal air volume adjustment range can be automatically given based on the air volume adjustment data table. By integrating advanced prediction models and intelligent decision support, the accuracy and efficiency of blast furnace pressure difference control can be significantly improved, ensuring the stable operation of the blast furnace, achieving refined adjustment of the air volume, maintaining the stability of the blast furnace pressure difference, and improving production efficiency.

[0026] See also Figure 2 , Figure 2 FIG. 1 is a flow chart showing a method for controlling a blast furnace pressure difference according to an embodiment of the present application. Figure 2 As shown, the method at least includes steps S210 to S240, which are described in detail as follows:

[0027] Step S210, obtaining the blast furnace parameters to be detected.

[0028] In one embodiment of the present application, the blast furnace parameters to be detected include furnace condition marking parameters to be detected, blast furnace airflow state parameters to be detected, and blast furnace airflow operation parameters to be detected corresponding to the same timestamp data.

[0029] Step S220, inputting the blast furnace parameters to be detected into a blast furnace pressure difference prediction model trained in advance or in real time to obtain a current predicted pressure difference value.

[0030] Step S230: If the current predicted pressure difference value is greater than the pressure difference upper limit threshold, an alarm prompt message is issued, and in response to the alarm prompt message, the initial air volume is gradually reduced until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold.

[0031] In one embodiment of the present application, every 10 minutes, the minute-level data value of the pressure difference in the next hour is predicted in real time and compared with the upper limit threshold of the pressure difference. If it exceeds the upper limit, an alarm is prompted.

[0032] Step S240, if the current predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, a search operation is performed in the air volume adjustment data table according to the predicted pressure difference value to obtain the optimal air volume adjustment range, and the air volume is adjusted based on the optimal air volume adjustment range to control the blast furnace pressure difference, wherein the air volume adjustment data table stores the corresponding relationship between the predicted pressure difference value and the optimal air volume adjustment range, and the air volume adjustment data table is stored in the operation suggestion module.

[0033] exist Figure 2 In the technical scheme shown, the blast furnace pressure difference prediction model trained in advance or in real time can more accurately predict the current blast furnace pressure difference value. Compared with the traditional method that relies on manual experience, the neural network model is trained with a large amount of data and can comprehensively consider more influencing factors to improve the accuracy and reliability of the prediction. When the predicted pressure difference value exceeds the preset safety threshold, the system can automatically issue an alarm prompt to promptly remind the operator to take measures, such as gradually reducing the air volume, to prevent possible abnormal furnace conditions or safety accidents, thereby enhancing the safety of blast furnace operation. For the situation where the predicted pressure difference value is within the safety range, the optimal air volume adjustment range recommendation can be automatically given based on the air volume adjustment data table. By integrating advanced prediction models and intelligent decision support, the accuracy and efficiency of blast furnace pressure difference control can be significantly improved, the stable operation of the blast furnace can be guaranteed, the fine adjustment of the air volume can be achieved, the stability of the blast furnace pressure difference can be maintained, and production efficiency can be improved.

[0034] In one embodiment of the present application, the training process of the blast furnace pressure difference prediction model includes: obtaining the key parameters of the blast furnace, which include airflow state parameters, airflow operation parameters, blast furnace pressure difference parameters and furnace condition marking parameters; extracting time series features of the key parameters of the blast furnace to obtain a training data set; training the initial blast furnace pressure difference prediction model through the training data set to obtain the blast furnace pressure difference prediction model.

[0035] As an example, the scenario is multi-dimensional time series prediction, that is, using historical multi-dimensional parameters to predict the target parameters for a short period of time in the future, and using time series prediction related neural networks for prediction. Time series data is at certain regular time intervals, so a series of corresponding data is generated in each time period. The corresponding problem of time series (here mainly time series problem prediction) is to predict the data of the next moment based on the data of the previous n moments, n is a natural number; the backbone network of the initial blast furnace pressure difference prediction model can be RNN (Recurrent Neural Networks), LSTM (Long Short-Term Memory), GRU (Gated Recurrent Unit), Transformer (deep learning model based on self-attention mechanism); usually RNN refers to recurrent neural network. Recurrent neural network is a type of neural network used to process sequence data. The difference between RNN and other neural networks is that RNN can better process sequence information, that is, it recognizes the relationship between the previous and next inputs; LSTM is a type of neural network containing LSTM blocks or other blocks. LSTM blocks may be described as intelligent network units because LSTM blocks can remember values ​​of indefinite time lengths; GRU is a variant of recurrent neural network that overcomes the gradient vanishing problem in traditional RNN by using a gating mechanism. The computational complexity of GRU is relatively small because GRU has fewer parameters and is a model that processes input sequences step by step. At each time step, GRU only needs to calculate some simple linear transformations and nonlinear activation functions; Transformer is a neural network architecture based on the attention mechanism for processing sequence data. Transformer introduces a self-attention mechanism that allows the model to pay weighted attention to each element of the input sequence at different positions. Transformer needs to calculate the fully connected attention matrix, which requires a large amount of computation.

[0036] In one embodiment of the present application, key parameters of a blast furnace are obtained, including: collecting blast furnace airflow state parameters, which include furnace body static pressure, wall temperature, throat cross temperature measurement and water temperature difference; collecting blast furnace airflow operation parameters, which include air volume and oxygen content; collecting blast furnace pressure difference parameters, which include wind pressure, top pressure and pressure difference parameters, and the pressure difference parameters are the difference between wind pressure and top pressure; collecting furnace condition marking parameters, which include furnace change signal and wind stop signal, and assigning marking values ​​to furnace condition marking parameters, which are used to indicate special situations encountered by the blast furnace during operation, and which include unavailable thresholds; storing blast furnace airflow state parameters, blast furnace airflow operation parameters, blast furnace pressure difference parameters and furnace condition marking parameters in a blast furnace parameter database.

[0037] In one embodiment of the present application, the key parameters of the blast furnace are subjected to time series feature extraction to obtain a training data set, including: collecting prediction points according to a preset time window; extracting the key parameters of the blast furnace in the preset time period before the collection prediction point, recording them as the key parameters of the blast furnace to be processed, preprocessing the data of the key parameters of the blast furnace to be processed, and obtaining a feature vector; setting the corresponding blast furnace pressure difference parameter to the feature vector, as the blast furnace pressure difference parameter of the prediction point, and determining the feature vector and the corresponding blast furnace pressure difference parameter as a training sample; traversing the blast furnace key parameters, generating multiple training samples, and combining all training samples into a training data set, and the training data set will be used to train the blast furnace pressure difference prediction model. The blast furnace pressure difference prediction model trained in advance or in real time can more accurately predict the pressure difference value of the current blast furnace. Compared with the traditional method of relying on artificial experience, the neural network model can be trained with a large amount of data, and can comprehensively consider more influencing factors and improve the accuracy and reliability of the prediction.

[0038] As an example, the preset time window is 1 hour, and the preset time period is 8 hours. According to the preset time window "1 hour", the prediction point is collected, and the key parameters of the blast furnace within 8 hours before the collection prediction point are extracted and recorded as the key parameters of the blast furnace to be processed. The data of the key parameters of the blast furnace to be processed are preprocessed to obtain a feature vector, and the corresponding blast furnace pressure difference parameter is set for the feature vector as the blast furnace pressure difference parameter of the prediction point, and the feature vector and the corresponding blast furnace pressure difference parameter are determined as training samples; the blast furnace key parameters are traversed to generate multiple training samples, and all training samples are combined into a training data set. The training data set will be used to train the blast furnace pressure difference prediction model, wherein the training data set includes independent variables, i.e., the key parameters of the blast furnace in the previous 8 hours, and dependent variables, i.e., the blast furnace pressure difference in the next 1 hour. This scenario is a multi-dimensional time series prediction, i.e., using historical multi-dimensional parameters to predict the target parameters for a short period of time in the future, and a time series prediction related neural network is selected for prediction. The blast furnace pressure difference prediction model trained in advance or in real time can more accurately predict the current blast furnace pressure difference value. Compared with the traditional method that relies on manual experience, the neural network model has been trained with a large amount of data and can comprehensively consider more influencing factors to improve the accuracy and reliability of the prediction.

[0039] In one embodiment of the present application, data of the key parameters of the blast furnace to be processed are preprocessed to obtain a feature vector, including: extracting timestamp data of furnace condition marking parameters, recorded as furnace condition marking timestamp data, extracting timestamp data of blast furnace airflow state parameters, recorded as airflow state timestamp data, extracting timestamp data of blast furnace airflow operation parameters, recorded as airflow operation timestamp data; screening out identical timestamp data from the furnace condition marking timestamp data, the airflow state timestamp data and the airflow operation timestamp data, and recording them as identical timestamp data; determining the furnace condition marking parameters, the blast furnace airflow state parameters and the blast furnace airflow operation parameters corresponding to the identical timestamp data as an associated blast furnace key parameter group; extracting the furnace condition marking parameter mark value in the associated blast furnace key parameter group, if the furnace condition marking parameter mark value is an unavailable threshold, marking other parameters in the associated blast furnace key parameter group as null values; and completing the associated blast furnace key parameter group corresponding to the null value using a linear interpolation method to obtain a feature vector.

[0040] In one embodiment of the present application, an air volume adjustment data table includes: collecting historical data, the historical data including multiple sample pressure difference values, setting multiple sample air volume adjustment values ​​for each sample pressure difference value, and obtaining multiple groups of adjustment data; verifying the adjustment result of each group of adjustment parameters, and performing efficiency scoring on each group of adjustment results; extracting the adjustment parameter with the highest efficiency score of the adjustment result corresponding to each sample pressure difference value, and recording it as the optimal adjustment parameter; integrating the optimal adjustment parameters corresponding to each sample pressure difference value to obtain an air volume adjustment data table.

[0041] In one embodiment of the present application, the initial air volume is gradually reduced in response to the alarm prompt information until the predicted pressure difference value of the output is less than or equal to the pressure difference upper limit threshold, including: obtaining the initial air volume adjustment value corresponding to the current predicted pressure difference value, subtracting the preset air volume threshold from the initial air volume adjustment value, and obtaining the next air volume value; based on the next air volume value, a new blast furnace parameter to be detected is obtained, and the new blast furnace parameter to be detected is input into the blast furnace pressure difference prediction model trained in advance or in real time, until the predicted pressure difference value of the output is less than or equal to the pressure difference upper limit threshold. Excessive pressure difference may cause instability inside the blast furnace, and even cause safety accidents. By gradually reducing the air volume, the pressure difference can be effectively controlled to prevent it from exceeding the safety upper limit, thereby ensuring the safe operation of the blast furnace.

[0042] As an example, when the current predicted pressure difference value is greater than the pressure difference upper limit threshold, the initial air volume adjustment value is deducted from the preset air volume threshold, for example, the preset air volume threshold is a multiple of 100, that is, the single air volume reduction is a multiple of 100, so the scheme for giving operation suggestions in the present invention is: assuming that the initial air volume is f, the prediction result of the blast furnace pressure difference prediction model is y; assuming that the air volume after decreasing is f-100, the new model prediction result is calculated as y1; comparing the new model prediction result y1 with the pressure difference upper limit threshold, if the limit is exceeded, the air volume is repeatedly decreased, and then 100 is subtracted and entered into the calculation, until the cycle is exited after not exceeding the limit, the air volume that needs to be adjusted is output, then according to the predicted pressure difference value, a search operation is performed in the air volume adjustment data table, and the optimal air volume adjustment range is obtained, and the air volume is adjusted based on the optimal air volume adjustment range to control the blast furnace pressure difference. Excessive pressure difference may cause instability inside the blast furnace, and even cause safety accidents. By gradually reducing the air volume, the pressure difference can be effectively controlled to avoid it exceeding the safety upper limit, thereby ensuring the safe operation of the blast furnace.

[0043] Figure 3 FIG. 1 is a block diagram of a blast furnace pressure difference control device according to an exemplary embodiment of the present application. Figure 3As shown, according to an embodiment of the present application, a blast furnace pressure difference control device 300 includes: an acquisition module 310, a pressure difference prediction module 320, an alarm module 330 and an air volume adjustment module 340, the acquisition module is used to acquire the blast furnace parameters to be detected; the pressure difference prediction module is used to input the blast furnace parameters to be detected into a blast furnace pressure difference prediction model trained in advance or in real time to obtain the current predicted pressure difference value; the alarm module is used to issue an alarm prompt message if the current predicted pressure difference value is greater than the pressure difference upper limit threshold, and gradually reduce the initial air volume in response to the alarm prompt message until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold; the air volume adjustment module is used to perform a search operation in the air volume adjustment data table according to the predicted pressure difference value if the current predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, obtain the optimal air volume adjustment range, and adjust the air volume based on the optimal air volume adjustment range to control the blast furnace pressure difference; wherein, the air volume adjustment data table stores the corresponding relationship between the predicted pressure difference value and the optimal air volume adjustment range, and the air volume adjustment data table is stored in the operation suggestion module. The blast furnace pressure difference prediction model trained in advance or in real time can more accurately predict the current blast furnace pressure difference value. Compared with the traditional method that relies on manual experience, the neural network model has been trained with a large amount of data and can comprehensively consider more influencing factors to improve the accuracy and reliability of the prediction. When the predicted pressure difference value exceeds the preset safety threshold, the system can automatically issue an alarm prompt to promptly remind the operator to take measures, such as gradually reducing the air volume, to prevent possible abnormal furnace conditions or safety accidents, thereby enhancing the safety of blast furnace operation. For situations where the predicted pressure difference value is within a safe range, the optimal air volume adjustment range can be automatically given based on the air volume adjustment data table. By integrating advanced prediction models and intelligent decision support, the accuracy and efficiency of blast furnace pressure difference control can be significantly improved, ensuring the stable operation of the blast furnace, achieving refined adjustment of the air volume, maintaining the stability of the blast furnace pressure difference, and improving production efficiency.

[0044] In one embodiment of the present application, the blast furnace pressure difference control device also includes a model training module 350, which is used to obtain key parameters of the blast furnace, including airflow state parameters, airflow operation parameters, blast furnace pressure difference parameters and furnace condition marking parameters; perform time series feature extraction on the key parameters of the blast furnace to obtain a training data set; train the initial blast furnace pressure difference prediction model through the training data set to obtain a blast furnace pressure difference prediction model.

[0045] In one embodiment of the present application, the model training module is used to collect blast furnace airflow state parameters, which include furnace body static pressure, wall temperature, throat cross temperature measurement and water temperature difference; collect blast furnace airflow operation parameters, which include air volume and oxygen content; collect blast furnace pressure difference parameters, which include wind pressure, top pressure and pressure difference parameters, and the pressure difference parameter is the difference between wind pressure and top pressure; collect furnace condition marking parameters, which include furnace change signal and wind stop signal, and assign mark values ​​to the furnace condition marking parameters, and the furnace condition marking parameter mark values ​​are used to indicate special situations encountered by the blast furnace during operation, and the furnace condition marking parameter mark values ​​include unavailable thresholds; store the blast furnace airflow state parameters, blast furnace airflow operation parameters, blast furnace pressure difference parameters and furnace condition marking parameters in the blast furnace parameter database.

[0046] In one embodiment of the present application, the model training module is used to collect prediction points according to a preset time window; extract the key parameters of the blast furnace within a preset time period before the prediction points are collected, record them as the key parameters of the blast furnace to be processed, preprocess the data of the key parameters of the blast furnace to be processed, and obtain a feature vector; set the corresponding blast furnace pressure difference parameter for the feature vector as the blast furnace pressure difference parameter of the prediction point, and determine the feature vector and the corresponding blast furnace pressure difference parameter as training samples; traverse the blast furnace key parameters to generate multiple training samples, and combine all training samples into a training data set, which will be used to train the blast furnace pressure difference prediction model.

[0047] In one embodiment of the present application, the model training module is used to extract the timestamp data of the furnace condition marking parameter, recorded as the furnace condition marking timestamp data, extract the timestamp data of the blast furnace airflow state parameter, recorded as the airflow state timestamp data, extract the timestamp data of the blast furnace airflow operation parameter, recorded as the airflow operation timestamp data; the same timestamp data are screened out from the furnace condition marking timestamp data, the airflow state timestamp data and the airflow operation timestamp data, and recorded as the same timestamp data; the furnace condition marking parameters, the blast furnace airflow state parameters and the blast furnace airflow operation parameters corresponding to the same timestamp data are determined as the associated blast furnace key parameter group; the furnace condition marking parameter mark value in the associated blast furnace key parameter group is extracted, and if the furnace condition marking parameter mark value is an unavailable threshold value, the other parameters in the associated blast furnace key parameter group are marked as null values; the associated blast furnace key parameter group corresponding to the null value is completed using the linear interpolation method to obtain a feature vector

[0048] In one embodiment of the present application, the air volume adjustment module is used to collect historical data, which includes multiple sample pressure difference values. Multiple sample air volume adjustment values ​​are set for each sample pressure difference value to obtain multiple groups of adjustment data; the adjustment results of each group of adjustment parameters are verified, and each group of adjustment results is scored for efficiency; the adjustment parameter with the highest efficiency score of the adjustment result corresponding to each sample pressure difference value is extracted and recorded as the optimal adjustment parameter; the optimal adjustment parameters corresponding to each sample pressure difference value are integrated to obtain an air volume adjustment data table.

[0049] In one embodiment of the present application, an alarm module is used to obtain an initial air volume adjustment value corresponding to a current predicted pressure difference value, subtract a preset air volume threshold from the initial air volume adjustment value, and obtain a next air volume value; based on the next air volume value, a new blast furnace parameter to be detected is obtained, and the new blast furnace parameter to be detected is input into a pre- or real-time trained blast furnace pressure difference prediction model until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold.

[0050] It should be noted that the device provided in the above embodiment and the method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, and will not be repeated here. In practical applications, the device provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0051] See also Figure 3 , Figure 3 is a schematic diagram of a structure of an electronic device shown in an exemplary embodiment of the present application. It should be noted that: Figure 3 The electronic device 300 shown is only an example and should not bring any limitation to the functions and usage area of ​​the embodiments of the present application.

[0052] like Figure 3 As shown, the electronic device 300 includes a processor 301, a memory 302 and a communication bus 303; the communication bus 303 is used to connect the processor 301 and the memory 302; the processor 301 is used to execute the computer program stored in the memory 302 to implement one or more methods in the above embodiments.

[0053] The electronic device provided in the present application includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used to communicate, and the processor and the transceiver are used to run the computer program so that the electronic device executes each step of the above method.

[0054] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0055] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0056] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A blast furnace pressure difference control method, characterized in that: The method comprises: Obtain the parameters of the blast furnace to be tested; Inputting the blast furnace parameter to be detected into a blast furnace pressure difference prediction model trained in advance or in real time to obtain a current predicted pressure difference value; If the current predicted pressure difference value is greater than the pressure difference upper limit threshold, an alarm prompt message is issued, and in response to the alarm prompt message, the initial air volume is gradually reduced until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold; If the current predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, a search operation is performed in the air volume adjustment data table according to the predicted pressure difference value to obtain the optimal air volume adjustment range, and the air volume is adjusted based on the optimal air volume adjustment range to control the blast furnace pressure difference; The air volume adjustment data table stores the corresponding relationship between the predicted pressure difference value and the optimal air volume adjustment range, and the air volume adjustment data table is stored in the operation suggestion module.

2. The blast furnace pressure difference control method according to claim 1, characterized in that: The training process of the blast furnace pressure difference prediction model includes: Acquiring key parameters of the blast furnace, wherein the key parameters of the blast furnace include airflow state parameters, airflow operation parameters, blast furnace pressure difference parameters and furnace condition mark parameters; Extracting time series features of the key parameters of the blast furnace to obtain a training data set; The initial blast furnace pressure difference prediction model is trained using the training data set to obtain a blast furnace pressure difference prediction model.

3. The blast furnace pressure difference control method according to claim 2, characterized in that: Obtain key parameters of blast furnace, including: Collecting blast furnace air flow state parameters, wherein the blast furnace air flow state parameters include furnace body static pressure, wall temperature, throat cross temperature measurement and water temperature difference; Collecting blast furnace airflow operating parameters, wherein the blast furnace airflow operating parameters include air volume and oxygen content; Collecting blast furnace pressure difference parameters, wherein the blast furnace pressure difference parameters include wind pressure, top pressure and pressure difference parameters, wherein the pressure difference parameter is the difference between the wind pressure and the top pressure; Collecting furnace condition marking parameters, the furnace condition marking parameters include furnace change signal and wind stop signal, and assigning marking values ​​to the furnace condition marking parameters, the furnace condition marking parameter marking values ​​are used to indicate special conditions encountered during the operation of the blast furnace, and the furnace condition marking parameter marking values ​​include unavailable thresholds; The blast furnace air flow state parameters, the blast furnace air flow operation parameters, the blast furnace pressure difference parameters and the furnace condition mark parameters are stored in a blast furnace parameter database.

4. The blast furnace pressure difference control method according to claim 2, characterized in that: The time series features of the key parameters of the blast furnace are extracted to obtain a training data set, including: Collect forecast points according to the preset time window; Extracting key parameters of the blast furnace within a preset time period before the acquisition prediction point, recording them as key parameters of the blast furnace to be processed, and preprocessing data of the key parameters of the blast furnace to be processed to obtain a feature vector; Setting a corresponding blast furnace pressure difference parameter for the feature vector as the blast furnace pressure difference parameter of the prediction point, and determining the feature vector and the corresponding blast furnace pressure difference parameter as training samples; The key parameters of the blast furnace are traversed to generate a plurality of training samples, and all the training samples are combined into the training data set, which will be used to train the blast furnace pressure difference prediction model.

5. The blast furnace pressure difference control method according to claim 4, characterized in that: Preprocess the data of the key parameters of the blast furnace to be processed to obtain a feature vector, including: Extracting the timestamp data of the furnace condition marking parameter, recorded as the furnace condition marking timestamp data, extracting the timestamp data of the blast furnace airflow state parameter, recorded as the airflow state timestamp data, extracting the timestamp data of the blast furnace airflow operation parameter, recorded as the airflow operation timestamp data; Screening out identical timestamp data from the furnace condition mark timestamp data, the airflow state timestamp data, and the airflow operation timestamp data, and recording them as identical timestamp data; Determine the furnace condition marking parameters, blast furnace air flow state parameters and blast furnace air flow operation parameters corresponding to the same timestamp data as an associated blast furnace key parameter group; Extracting the furnace condition marking parameter mark value in the associated blast furnace key parameter group, if the furnace condition marking parameter mark value is an unavailable threshold, marking other parameters in the associated blast furnace key parameter group as null values; The associated blast furnace key parameter group corresponding to the null value is completed by using a linear interpolation method to obtain a feature vector.

6. The blast furnace pressure difference control method according to claim 1, characterized in that: Air volume adjustment data sheet, including: Collecting historical data, wherein the historical data includes a plurality of sample pressure difference values, and setting a plurality of sample air volume adjustment values ​​for each sample pressure difference value to obtain a plurality of groups of adjustment data; Verify the adjustment results of each set of adjustment parameters and give efficiency scores to each set of adjustment results; Extract the adjustment parameter with the highest adjustment result efficiency score corresponding to each sample pressure difference value and record it as the optimal adjustment parameter; The optimal adjustment parameters corresponding to each sample pressure difference value are integrated to obtain the air volume adjustment data table.

7. The blast furnace pressure difference control method according to claim 6, characterized in that: In response to the alarm prompt information, the initial air volume is gradually reduced until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold, including: Obtaining an initial air volume adjustment value corresponding to the current predicted pressure difference value, and subtracting a preset air volume threshold from the initial air volume adjustment value to obtain a next air volume value; Based on the next air volume value, a new blast furnace parameter to be detected is obtained, and the new blast furnace parameter to be detected is input into a blast furnace pressure difference prediction model trained in advance or in real time until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold.

8. A blast furnace pressure difference control device, characterized in that: The device comprises: An acquisition module is used to acquire the blast furnace parameters to be detected; A pressure difference prediction module is used to input the blast furnace parameters to be detected into a blast furnace pressure difference prediction model trained in advance or in real time to obtain a current predicted pressure difference value; an alarm module, configured to issue an alarm prompt message if the current predicted pressure difference value is greater than the pressure difference upper limit threshold, and gradually reduce the initial air volume in response to the alarm prompt message until the output predicted pressure difference value is less than or equal to the pressure difference upper limit threshold; An air volume adjustment module is used for, if the current predicted pressure difference value is less than or equal to the pressure difference upper limit threshold value, searching in the air volume adjustment data table according to the predicted pressure difference value to obtain the optimal air volume adjustment range, and adjusting the air volume based on the optimal air volume adjustment range to control the blast furnace pressure difference; The air volume adjustment data table stores the corresponding relationship between the predicted pressure difference value and the optimal air volume adjustment range, and the air volume adjustment data table is stored in the operation suggestion module.

9. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enables the electronic device to implement the blast furnace pressure difference control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is used to make a computer execute the method according to any one of claims 1 to 7.

Images