A control system for feeding materials in high-level silo of converter

Abstract

The present invention relates to the technical field of feeding program control, and provides a feeding control system for a high-level silo of a converter, wherein a material detection module detects the real-time changes of materials in the high-level silo of the converter in real time, and transmits the real-time changes in material demand to an intelligent adjustment module, the intelligent adjustment module obtains historical feeding information, and calculates an adjustment strategy according to the information on changes in material demand, the equipment driving module adjusts the speed or amount of the feeding equipment in real time according to the adjustment strategy, the safety monitoring module monitors the feeding equipment in real time, and sends an alarm when a fault or abnormal situation is monitored, the data storage module records the feeding data information of the above modules, and transmits it to an interactive module; the present invention realizes real-time automatic adjustment of feeding parameters through the material detection module and the intelligent adjustment module, reduces frequent manual intervention, and improves the adaptability of the system, and the safety monitoring module ensures the safe and stable operation of the feeding process.

Description

Technical Field

[0001] The invention relates to the technical field of material feeding program control, in particular to a material feeding control system for a converter high-level material bin. Background Art

[0002] Converter steelmaking uses molten iron, scrap steel and ferroalloy as the main raw materials. It does not rely on external energy sources, but relies on the physical heat of the molten iron itself and the heat generated by the chemical reaction between the components of the molten iron to complete the steelmaking process in the converter. The converter is the most commonly used steelmaking equipment in steelmaking, mainly used for the production of carbon steel, alloy steel and the smelting of copper and nickel.

[0003] The converter high-level silo is a storage container installed at a higher position above the converter. It is used to store various raw materials required in the steelmaking process, such as ferroalloys, slag-forming agents, lime, etc.; it is usually made of metal materials and has a certain strength and sealing to ensure that the stored materials will not leak or be contaminated by the outside world. It is generally cylindrical or square in shape and has a certain volume, which can meet the material needs of the converter for a period of time.

[0004] The loading of the converter high-level silo refers to the transportation of various raw materials required for steelmaking, such as ferroalloys, slag-making agents, lime, etc., from the ground or lower position to the converter high-level silo located at a high place through specific conveying equipment, so that they can be added to the converter in time according to the process requirements during the steelmaking process. The existing loading methods include belt conveyor loading, vibration feeding loading, bucket lifting loading and pneumatic transport loading. These loading methods have their own advantages and disadvantages. When choosing, it is necessary to comprehensively consider many factors, including production needs, site conditions, material characteristics, equipment reliability and cost, etc.; by reasonably selecting and optimizing the loading method, the production efficiency and quality of converter steelmaking can be improved, providing strong support for the sustainable development of the steel industry.

[0005] With the needs of the steelmaking industry, the existing patent (Announcement No.: CN108153232B) discloses an automatic control system and control method for feeding materials to the high-level silo of the converter, which includes a control system including a comprehensive control module; a feeding clock control module, a high-level silo feeding amount calculation module, a feeding priority determination module, a material stacking density correction module, a feeding trolley control module, an underground material yard vibration feeding control module, a feeding amount real-time metering module, and a feeding process comprehensive control module. The comprehensive control module is respectively connected to the feeding clock control module, the high-level silo feeding amount calculation module, the feeding priority determination module, the material stacking density correction module, and the feeding process comprehensive control module. The feeding process comprehensive control module is simultaneously connected to the feeding trolley control module, the underground material yard vibration feeding control module, and the feeding amount real-time metering module. This invention has a simple structure and is easy to use. Through this method, the automatic control of the high-level silo vibration feeding method greatly reduces the operating intensity, improves the work efficiency and control accuracy, and realizes the transformation of the original traditional equipment control to modern digital control, which is convenient for large-scale promotion and application. However, there are still shortcomings in the following aspects;

[0006] 1. The current system has obvious deficiencies in security. In the actual steelmaking feeding process, when there are abnormal situations such as sudden motor failure, conveyor belt tearing, etc. that cause the feeding tool to malfunction, the system has no corresponding sensing mechanism to detect these failures. In addition, in some potentially dangerous situations, such as loose parts in key areas that may cause unstable equipment operation, the system cannot monitor them. This poses a great threat to the safe and stable operation of the entire steelmaking feeding process.

[0007] 2. Insufficient adaptability. It is limited to one feeding method and cannot flexibly switch the feeding method according to different steelmaking process requirements. In addition, the real-time changes in materials used in the steelmaking process are complex and diverse. The types of raw materials, feeding frequency and speed are constantly changing, but the system has limited adaptability to these changes and cannot automatically adjust the feeding parameters according to demand in real time. As a result, frequent manual intervention is often required to adjust relevant parameters during the production process. This not only increases the complexity and labor intensity of the operation, but may also affect production efficiency and product quality due to untimely manual adjustments. Summary of the invention

[0008] In order to solve the above technical problems, the present invention provides a converter high-level silo feeding control system, which automatically adjusts the speed of the feeding equipment by detecting the real-time changes in material demand, and the safety monitoring module uses the ant colony algorithm to calculate the equipment operation data path, and monitors whether the equipment fails by judging the path changes, so as to solve the problems in the prior art.

[0009] The technical solution of the present invention is as follows:

[0010] A converter high-level silo feeding control system, applicable to any one of belt conveyor feeding equipment, vibrating feeder feeding equipment, bucket lifting feeding equipment or pneumatic transport feeding equipment, including:

[0011] The material detection module includes but is not limited to a weight sensor, a volume sensor, a density sensor and an optical sensor, and is used to detect changes in material demand in the silo in real time.

[0012] The equipment driving module includes belt conveying feeding equipment, vibrating feeding feeding equipment, bucket lifting feeding equipment and pneumatic transport feeding equipment, and provides corresponding driving control for different feeding equipment.

[0013] An intelligent adjustment module, which first acquires historical feeding information for learning, and then automatically adjusts the feeding speed and feeding amount according to the changes of the materials detected by the material detection module;

[0014] A safety monitoring module, which is used to monitor the operating status of the feeding equipment in real time and send an alarm message when a fault or abnormal situation is monitored;

[0015] The data storage module is used to record various data in the feeding process, including material quantity, feeding time and equipment operating parameters.

[0016] The interactive module provides an intuitive user interface for operators to monitor the operating status of the feeding system, set parameters and handle alarm information.

[0017] The interaction relationship between the modules in the converter high-level silo feeding control system is as follows:

[0018] The material detection module detects the real-time changes of materials in the high-level silo of the converter in real time, and transmits the real-time changes in material demand to the intelligent adjustment module. The intelligent adjustment module obtains historical loading information and calculates the adjustment strategy based on the material demand change information. The equipment driving module adjusts the loading equipment speed or loading amount in real time according to the adjustment strategy. The safety monitoring module monitors the loading equipment in real time and sends an alarm when a fault or abnormal situation is detected. The data storage module records the loading data information of the above modules and transmits it to the interactive module.

[0019] Preferably, the material detection module can accurately measure the amount of material entering the feeding system and the situation of the remaining material in the silo based on the weight sensor, and use the optical sensor to detect the composition and quality of the material to ensure that the raw materials fed meet the steelmaking requirements.

[0020] Preferably, the device driving module can accurately control the running speed, power and start-stop operation of the device according to the intelligent adjustment module, and stop the operation of the device in time according to the alarm information of the safety monitoring module.

[0021] Preferably, the intelligent adjustment module includes a calculation unit, an adjustment strategy unit and an adjustment unit;

[0022] The calculation unit is used to obtain the correct speed and frequency of feeding corresponding to the changes in various requirements in the historical feeding data;

[0023] The adjustment strategy unit generates an adjustment strategy according to the detection information fed back by the material detection module and the calculation result of the calculation unit;

[0024] The adjustment unit sends the adjustment strategy generated by the adjustment strategy unit to the device driver module for device adjustment.

[0025] Preferably, the calculation unit uses the changes in the demand for various materials in the historical feeding data as input features, outputs the corresponding output value through double calculation, compares the output value with the acquired historical correct feeding speed or feeding amount, and continuously adjusts the weight coefficient in the double calculation to achieve a one-to-one correspondence between the calculated output value and the historical correct feeding speed or feeding amount;

[0026] Suppose the input feature set is , n is the total number of features, where Represents the i-th feature, and calculates the features of each input and its corresponding output value respectively;

[0027] The first calculation, That is The corresponding output of the first calculation;

[0028]

[0029] The second calculation, That is The corresponding output of the second calculation;

[0030]

[0031] Among them, w represents the weight, Represents the weight of the first calculation, represents the weight of the second calculation, b represents the bias term, represents the bias term for the first calculation, Represents the bias term of the second calculation. Because there is a nonlinear transformation between the input features and the output values, an activation function is added to the calculation to calculate these nonlinear features. The hyperbolic tangent activation function f is added in the first calculation, and e is a natural constant.

[0032]

[0033] The first calculation is: ;

[0034] Add a linear activation function in the second calculation , , that is, the calculation process is consistent with the second calculation process;

[0035] set up In order to collect the actual material speed or frequency in the corresponding situation, the weight coefficient and bias term are continuously adjusted until L is 0:

[0036]

[0037] When L is 0, the current weight coefficient and bias term are confirmed as this input feature Corresponding output value Calculate the weight coefficients and bias terms of the formula;

[0038] In real-time calculation, because the real-time material demand changes of the converter high-level silo are various, the feature set is , when calculating, the feature set needs to be Substitute:

[0039]

[0040]

[0041] in, It means the input feature set The average result of the first calculation is Meaning of input The average result of the second calculation is the average of the output values, which is the feeding speed or feeding amount that needs to be adjusted.

[0042] Preferably, the security monitoring module includes a monitoring unit, an analysis unit and an alarm unit;

[0043] The monitoring unit includes a variety of sensors, including but not limited to temperature sensors, pressure sensors and vibration sensors, which are installed in the feeding equipment to monitor the real-time data of various parts of the feeding equipment in real time;

[0044] The analysis unit calculates the ant colony paths between various parts of the feeding equipment using the ant colony algorithm according to the real-time data monitored by the monitoring unit, and determines whether the equipment is faulty or abnormal according to the changes in the ant colony paths;

[0045] The alarm unit is used to issue an alarm and send early warning information according to the judgment result of the analysis unit;

[0046] Preferably, the analysis unit regards each monitoring point of the feeding system as a node in the ant colony algorithm. The transfer probability of ants, i.e., data monitoring information, between nodes is set according to the change law of sensor data. When the equipment operates normally, the transfer path of the ants is relatively stable. When a fault occurs, the change of sensor data will cause the transfer probability to change, and the path of the ants will also change. The fault is determined by monitoring this path change.

[0047] The ant path calculation process is as follows:

[0048] Initialization: Ants are randomly placed at each monitoring point, and each ant has a list of possible movement directions, that is, other monitoring points that can be reached from the current monitoring point;

[0049] Transition probability selection: Let r be the ant number, representing the rth ant, t be the time, i and j represent different nodes i and j; at each step, calculate the transition probability of ant k transferring from node i to node j , the formula is:

[0050]

[0051] in, is the pheromone concentration on the path from node i to node j at time t, is the weight parameter of pheromone concentration, which determines the degree of influence of pheromone concentration on ants' path selection; the larger it is, the more the ants will rely on pheromone concentration to choose the path. It is a heuristic factor, which is a kind of prompt information used to characterize the quality of the ant's choice from node i to node j. It is determined according to the correlation of the data of the two nodes or the physical distance factor. If the correlation or physical distance changes, the heuristic factor will change. is the weight parameter of the heuristic factor, which determines the influence of the heuristic information on the ant's path selection;

[0052] s is an index variable, d represents the set of next nodes that ant k can choose. When the ant is at a node, it has multiple possible destinations, which constitute the set d. s traverses these possible destination nodes in turn and calculates the distance from the current node to each s node. The value of

[0053] According to the calculated transfer probability, the ants make random selections through a random number generator. After each selection, the pheromone concentration of the path chosen by the ants is updated and calculated.

[0054]

[0055] in, Indicates the volatilization of pheromones. is the pheromone volatilization coefficient, which controls the speed of pheromone volatilization. As time goes by, the pheromone on the path will gradually decrease due to this coefficient. The value of determines the amount of calculation. The larger the The larger the amount of calculation, the smaller the The smaller the amount of calculation;

[0056] Pheromone Increase: Indicates the path that the ants pass through Newly added pheromones, It is the increase in pheromones left by the ants that changes the attractiveness of the path to other ants;

[0057] As the ants make their move choices step by step based on the transfer probability, the path of the ants from the starting monitoring point to each subsequent monitoring point is gradually constructed;

[0058] The calculation formula of pheromone change rate A is:

[0059]

[0060] In a number of consecutive iterations, if the rate of change of pheromone concentration on all paths is less than a preset 0.1, the pheromone concentration is considered to be stable, the algorithm stops, and the normal operation path of the system is clear;

[0061] When the monitoring unit monitors the real-time data of each part of the feeding equipment and substitutes it into real-time calculation, if the new path deviates from the original normal operating path, it is considered that the equipment has a fault or abnormal situation.

[0062] Preferably, the data storage module stores data with precise timestamps, ensuring that the timestamps of relevant real-time data are accurate to the second level, so as to more accurately record the time-related information of the loading process and provide support for subsequent analysis.

[0063] Preferably, the user interface provided by the interactive module can display real-time data including the operation status of the feeding equipment and the real-time change information of the material demand, allowing the operator to set relevant calculation parameters for the system.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] 1. The present invention realizes the monitoring of real-time data of various parts of the feeding equipment by setting a monitoring unit of multiple sensors in the safety monitoring module, which can timely discover equipment failure or abnormal conditions, thereby improving the safety of the entire steelmaking feeding process; the analysis unit uses the ant colony algorithm to calculate the ant colony paths between various parts of the feeding equipment, and determines whether the equipment is faulty according to the path changes, and can monitor potential dangerous situations, thereby enhancing the system's perception of potential risks and ensuring the safe and stable operation of the feeding process; the alarm unit sends early warning information according to the judgment result of the analysis unit, so that the operator can take measures quickly to avoid the expansion of the failure and reduce the occurrence of safety accidents.

[0066] 2. The present invention detects changes in material demand in real time in the material detection module, realizes real-time perception of complex and diverse changes in steelmaking materials, and provides a data basis for the system to adjust feeding parameters; the intelligent adjustment module automatically adjusts the feeding speed and feeding amount according to the information of the material detection module and historical feeding information, realizes real-time automatic adjustment of feeding parameters, reduces frequent manual intervention, reduces operation complexity and labor intensity, and can timely adapt to material changes in the production process, improve production efficiency and product quality; the equipment drive module provides corresponding drive control for different feeding equipment, realizes the coordinated work of multiple feeding equipment, and enables the system to flexibly switch feeding methods according to different steelmaking process requirements, thereby improving the adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a work flow chart of the present invention;

[0068] Figure 2 It is a working flow diagram of the intelligent adjustment module of the present invention;

[0069] Figure 3 It is a work flow chart of the safety monitoring module of the present invention. DETAILED DESCRIPTION

[0070] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0071] The present invention provides a converter high-level silo feeding control system, which is applicable to any one of belt conveying feeding equipment, vibrating feeding feeding equipment, bucket lifting feeding equipment or pneumatic transport feeding equipment, and includes:

[0072] Material detection module, which is based on weight sensors, volume sensors, density sensors and optical sensors, and is used to detect the changes in material demand in the silo in real time. The weight and volume sensors intuitively present the material inventory to determine demand, the density sensor determines the change in material stacking state by density changes, and the optical sensor analyzes the material quality by detecting the material's reaction to light;

[0073] An equipment driving module, which performs loading of the converter high-level silo based on belt conveyor loading equipment, vibrating feeder loading equipment, bucket lifting loading equipment and pneumatic transport loading equipment, and provides corresponding drive control for different loading equipment;

[0074] An intelligent adjustment module, which first obtains historical feeding information for learning, and then automatically adjusts the feeding speed and feeding amount according to the changes in material demand in the silo detected by the material detection module;

[0075] A safety monitoring module, which is used to monitor the operating status of the feeding equipment in real time and send an alarm message when a fault or abnormal situation is monitored;

[0076] The data storage module is used to record various data in the feeding process, including material quantity, feeding time and equipment operating parameters.

[0077] The interactive module provides an intuitive user interface for operators to monitor the operating status of the feeding system, set parameters and handle alarm information.

[0078] The interaction relationship between the modules in the converter high-level silo feeding control system is as follows:

[0079] The material detection module detects the real-time changes of materials in the high-level silo of the converter in real time, and transmits the real-time changes in material demand to the intelligent adjustment module. The intelligent adjustment module obtains historical loading information and calculates the adjustment strategy based on the material demand change information. The equipment driving module adjusts the loading equipment speed or loading amount in real time according to the adjustment strategy. The safety monitoring module monitors the loading equipment in real time and sends an alarm when a fault or abnormal situation is detected. The data storage module records the loading data information of the above modules and transmits it to the interactive module.

[0080] Embodiment 1: Figure 1-3 As shown, in this embodiment, during the steelmaking process of a steel plant, due to the diverse steelmaking processes and complex changes in raw materials, the original feeding system cannot be flexibly adjusted, which seriously affects the production efficiency and product quality. Therefore, this system is introduced to control the feeding of the converter high-level silo;

[0081] The material detection module uses a weight sensor to accurately measure the amount of material entering the feeding system and the remaining material in the silo. It uses an optical sensor to detect information such as the intensity of reflected light or transmitted light to detect the composition and quality of the material, ensuring that the raw materials fed meet steelmaking requirements.

[0082] The device driver module can accurately control the operating speed, power and start and stop operations of the equipment according to the adjustment strategy generated by the intelligent adjustment module, and stop the equipment operation in time according to the alarm information of the safety monitoring module.

[0083] The intelligent adjustment module includes a calculation unit, an adjustment strategy unit and an adjustment unit;

[0084] The calculation unit is used to obtain the feeding speed and feeding amount corresponding to the changes in various demand situations in the historical feeding data;

[0085] The adjustment strategy unit generates an adjustment strategy based on the detection information fed back by the material detection module and the calculation result of the calculation unit. The adjustment strategy is used to adjust the feeding speed or feeding amount;

[0086] The calculation unit takes the changes in various material requirements in the historical feeding data as input features, including the inventory of materials in the silo and the speed of material consumption, etc., and outputs the corresponding output value through double calculation, compares the output value with the historical correct feeding speed or feeding amount, and continuously adjusts the weight coefficient in the double calculation to achieve a one-to-one correspondence between the calculated output value and the historical correct feeding speed or feeding amount;

[0087] Suppose the input feature set is , n is the total number of features, where Represents the i-th feature, such as the material consumption rate in the silo, the material density in the material yard, that is, the material distribution and the physical properties of the material, and calculates each input feature and its corresponding output value;

[0088] The first calculation, That is The corresponding output of the first calculation;

[0089]

[0090] The second calculation, That is The corresponding output of the second calculation;

[0091]

[0092] Among them, w represents the weight, Represents the weight of the first calculation, represents the weight of the second calculation, b represents the bias term, represents the bias term for the first calculation, Represents the bias term of the second calculation. Because there is a nonlinear transformation between the input features and the output values, an activation function is added to the calculation to calculate these nonlinear features. The hyperbolic tangent activation function f is added in the first calculation, and e is a natural constant.

[0093]

[0094] The first calculation is: ;

[0095] Add a linear activation function in the second calculation , , that is, the calculation process is consistent with the second calculation process;

[0096] set up In order to collect the actual material speed or frequency in the corresponding situation, the weight coefficient and bias term are continuously adjusted until L is 0:

[0097]

[0098] When L is 0, the current weight coefficient and bias term are confirmed as this input feature Corresponding output value Calculate the weight coefficients and bias terms of the formula;

[0099] In real-time calculation, because the real-time material demand changes of the converter high-level silo are various, the feature set is , when calculating, the feature set needs to be Substitute:

[0100]

[0101]

[0102] in, It means the input feature set The average result of the first calculation is Meaning of input The average result of the second calculation is the average of the output values, which is the feeding speed or feeding amount that needs to be adjusted.

[0103] An adjustment strategy is generated according to the current feeding speed or feeding amount and the feeding speed or feeding amount that needs to be adjusted. The adjustment unit issues the adjustment strategy generated by the adjustment strategy unit to the device driving module for device adjustment.

[0104] Through a unique dual calculation method, the weight coefficient is continuously adjusted according to historical feeding data, so that the output value corresponds to the historical correct feeding speed or feeding amount. In the continuous production process, this precise adjustment method can make the feeding process more stable and efficient. For example, in a certain stage of the steelmaking process, the demand for raw materials fluctuated slightly due to changes in steelmaking, but the intelligent adjustment module was able to respond quickly to maintain the stability of feeding and ensure that the quality of steel is not affected.

[0105] Embodiment 2: Figure 1-3 As shown, in this embodiment, during the steelmaking process of a certain steel plant, the loading link has always had problems of insufficient safety and poor adaptability. The traditional loading method cannot detect motor failure, conveyor belt abnormality and other situations in time, resulting in production stagnation and even safety accidents. Therefore, this system is introduced to control the loading of the converter high-level silo;

[0106] The security monitoring module includes a monitoring unit, an analysis unit and an alarm unit;

[0107] The monitoring unit includes a variety of sensors, including but not limited to temperature sensors, pressure sensors and vibration sensors, which are installed in the feeding equipment to monitor the real-time data of various parts of the feeding equipment in real time;

[0108] The analysis unit calculates the ant colony paths between various parts of the feeding equipment using the ant colony algorithm based on the real-time data monitored by the monitoring unit, and determines whether the equipment is faulty or abnormal based on the changes in the ant colony paths;

[0109] The analysis unit regards each monitoring point of the feeding system as a node in the ant colony algorithm. The transfer probability of ants, i.e. data monitoring information, between nodes is set according to the change law of sensor data. When the equipment is operating normally, the transfer path of the ants is relatively stable. When a fault occurs, the change of sensor data will cause the transfer probability to change, and the path of the ants will also change. The fault is judged by monitoring this path change.

[0110] The ant path calculation process is as follows:

[0111] Initialization: Ants are randomly placed at each monitoring point, and each ant has a list of possible movement directions, that is, other monitoring points that can be reached from the current monitoring point;

[0112] Transition probability selection: Let r be the ant number, representing the rth ant, t be the time, i and j represent different nodes i and j; at each step, calculate the transition probability of ant k transferring from node i to node j , the formula is:

[0113]

[0114] in, is the pheromone concentration on the path from node i to node j at time t, is the weight parameter of pheromone concentration, which determines the degree of influence of pheromone concentration on ants' path selection; the larger it is, the more the ants will rely on pheromone concentration to choose the path. It is a heuristic factor, which is a kind of prompt information used to characterize the quality of the ant's choice from node i to node j. It is determined according to the correlation of the data of the two nodes or the physical distance factor. If the correlation or physical distance changes, the heuristic factor will change. is the weight parameter of the heuristic factor, which determines the influence of the heuristic information on the ant's path selection;

[0115] s is an index variable, d represents the set of next nodes that ant k can choose. When the ant is at a node, it has multiple possible destinations, which constitute the set d. s traverses these possible destination nodes in turn and calculates the distance from the current node to each s node. The value of

[0116] According to the calculated transfer probability, the ants make random selections through a random number generator. After each selection, the pheromone concentration of the path chosen by the ants is updated and calculated.

[0117]

[0118] in, Indicates the volatilization of pheromones. is the pheromone volatilization coefficient, which controls the speed of pheromone volatilization. As time goes by, the pheromone on the path will gradually decrease due to this coefficient. The value of determines the amount of calculation. The larger the The larger the amount of calculation, the smaller the The smaller the amount of calculation;

[0119] Pheromone Increase: Indicates the path that the ants pass through Newly added pheromones, It is the increase in pheromones left by the ants that changes the attractiveness of the path to other ants;

[0120] As the ants make their move choices step by step based on the transfer probability, the path of the ants from the starting monitoring point to each subsequent monitoring point is gradually constructed;

[0121] The calculation formula of pheromone change rate A is:

[0122]

[0123] In a number of consecutive iterations, if the rate of change of pheromone concentration on all paths is less than a preset 0.1, the pheromone concentration is considered to be stable, the algorithm stops, and the normal operation path of the system is clear;

[0124] In this system, the sensor data is stable when the equipment is operating normally, and the heuristic factor between the sensor data of nodes i and j is Maintain a certain synchronous and unchanging relationship, so the pheromone concentration is stable and the ant transfer path is also stable.

[0125] For example, in a belt conveyor feeding device, vibration sensors are set at two locations of the belt. These two vibration sensors are set as nodes i and j. During normal operation of the device, the vibration sensors at nodes i and j maintain the same vibration frequency, so the heuristic factor Maintain a synchronous equal relationship, so the equipment operates normally at this time, the pheromone concentration including the update of pheromone concentration is stable, and the ant transfer path remains stable.

[0126] When the monitoring unit monitors the real-time data of each part of the feeding equipment and substitutes it into the heuristic factor for real-time calculation, if the vibration sensor data of one of the nodes changes or stops vibrating directly due to a fault or abnormal situation, it will cause the heuristic factor The change in the pheromone concentration makes it impossible to maintain a synchronous equality relationship, which leads to a change in the pheromone concentration and ultimately a change in the ant's path. If the new path deviates from the original normal operating path, it is considered that the equipment has a fault or abnormality; the alarm unit issues an alarm based on the judgment result of the analysis unit and sends a warning message.

[0127] The ant colony algorithm is used to calculate the ant colony paths between various parts of the feeding equipment, and the changes in the paths are used to determine whether the equipment is faulty or abnormal. This unique algorithm can detect potential faults in advance. During another production process, a key part of the feeding equipment was slightly loose, causing the ant path to change. The safety monitoring module issued an alarm in time, allowing maintenance personnel to handle it before the equipment failed, avoiding the occurrence of an accident.

[0128] The data storage module stores data with accurate timestamps, ensuring that the timestamps of relevant real-time data are accurate to the second level, so as to more accurately record the time-related information of the feeding process and provide support for subsequent analysis. The user interface provided by the interactive module can display real-time data including the operation status of the feeding equipment and the real-time changes in material demand, allowing operators to set relevant calculation parameters for the system.

[0129] The embodiments of the present invention are provided for the purpose of illustration and description. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present invention. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A converter high-level silo feeding control system, suitable for any one of belt conveyor feeding equipment, vibrating feeding equipment, bucket lifting feeding equipment or pneumatic transport feeding equipment, characterized in that: include: A material detection module, which is based on a weight sensor, a volume sensor, a density sensor and an optical sensor, and is used to detect changes in material demand in the silo in real time; An equipment driving module, which performs loading of the converter high-level silo based on belt conveyor loading equipment, vibrating feeder loading equipment, bucket lifting loading equipment and pneumatic transport loading equipment, and provides corresponding drive control for different loading equipment; An intelligent adjustment module, which first obtains historical feeding information for learning, and then automatically adjusts the feeding speed and feeding amount according to the changes in material demand in the silo detected by the material detection module; A safety monitoring module, which is used to monitor the operating status of the feeding equipment in real time and send an alarm message when a fault or abnormal situation is monitored; A data storage module, which is used to record various data during the loading process, including material quantity, loading time and equipment operating parameters; An interactive module, which provides an intuitive user interface for operators to monitor the operating status of the feeding system, set parameters and handle alarm information; The interaction relationship between the modules in the converter high-level silo feeding control system is as follows: The material detection module detects the real-time changes of the materials in the high-level silo of the converter in real time, and transmits the real-time changes of the material demand to the intelligent adjustment module. The intelligent adjustment module obtains the historical feeding information, calculates the adjustment strategy according to the information of the material demand change, and the equipment driving module adjusts the feeding equipment speed or feeding amount in real time according to the adjustment strategy. The safety monitoring module monitors the feeding equipment in real time and sends an alarm when a fault or abnormal situation is monitored. The data storage module records the feeding data information of the above modules and transmits it to the interactive module; The intelligent adjustment module includes a calculation unit, an adjustment strategy unit and an adjustment unit; The calculation unit is used to obtain the feeding speed and feeding amount corresponding to the changes in various demand situations in the historical feeding data; The adjustment strategy unit generates an adjustment strategy according to the detection information fed back by the material detection module and the calculation result of the calculation unit; The regulating unit sends the regulating strategy generated by the regulating strategy unit to the device driver module for device regulation; The calculation unit uses the changes in the demand for various materials in the historical feeding data as input features, outputs the corresponding output value through double calculation, compares the output value with the acquired historical correct feeding speed or feeding amount, and continuously adjusts the weight coefficient in the double calculation to achieve a one-to-one correspondence between the calculated output value and the historical correct feeding speed or feeding amount; Suppose the input feature set is , n is the total number of features, where Represents the i-th feature, and calculates the features of each input and its corresponding output value respectively; The first calculation, That is The corresponding output of the first calculation; The second calculation, That is The corresponding output of the second calculation; Among them, w represents the weight, Represents the weight of the first calculation, represents the weight of the second calculation, b represents the bias term, represents the bias term for the first calculation, Represents the bias term of the second calculation. Because there is a nonlinear transformation between the input features and the output values, an activation function is added to the calculation to calculate these nonlinear features. The hyperbolic tangent activation function f is added in the first calculation, and e is a natural constant. The first calculation is: ; Add a linear activation function in the second calculation , , that is, the calculation process is consistent with the second calculation process; set up In order to collect the actual material speed or frequency in the corresponding situation, the weight coefficient and bias term are continuously adjusted until L is 0: When L is 0, the current weight coefficient and bias term are confirmed as this input feature Corresponding output value Calculate the weight coefficients and bias terms of the formula; In real-time calculation, because the real-time material demand changes of the converter high-level silo are various, the feature set is , when calculating, the feature set needs to be Substitute: in, It means the input feature set The average result of the first calculation is Meaning of input The average result of the second calculation is the average of the output values, which is the feeding speed or feeding amount that needs to be adjusted.

2. A converter high-level silo feeding control system as claimed in claim 1, characterized in that: The material detection module is based on a weight sensor that can accurately measure the amount of material entering the feeding system and the situation of the remaining material in the silo, and uses an optical sensor to detect the composition and quality of the material to ensure that the raw materials fed meet the steelmaking requirements.

3. A converter high-level silo feeding control system as claimed in claim 1, characterized in that: The device driving module can accurately control the running speed, power and start-stop operation of the device according to the intelligent adjustment module, and stop the device operation in time according to the alarm information of the safety monitoring module.

4. A converter high-level silo feeding control system as claimed in claim 1, characterized in that: The security monitoring module includes a monitoring unit, an analysis unit and an alarm unit; The monitoring unit includes a variety of sensors, including but not limited to temperature sensors, pressure sensors and vibration sensors, which are installed in the feeding equipment to monitor the real-time data of various parts of the feeding equipment in real time; The analysis unit calculates the ant colony paths between various parts of the feeding equipment using the ant colony algorithm according to the real-time data monitored by the monitoring unit, and determines whether the equipment is faulty or abnormal according to the changes in the ant colony paths; The alarm unit is used to issue an alarm and send early warning information according to the judgment result of the analysis unit.

5. A converter high-level silo feeding control system as claimed in claim 4, characterized in that: The analysis unit regards each monitoring point of the feeding system as a node in the ant colony algorithm. The transfer probability of ants, i.e., data monitoring information, between nodes is set according to the change law of sensor data. When the equipment is operating normally, the transfer path of the ants is relatively stable. When a fault occurs, the change of sensor data will cause the transfer probability to change, and the path of the ants will also change. The fault is determined by monitoring this path change. The ant path calculation process is as follows: Initialization: Ants are randomly placed at each monitoring point, and each ant has a list of possible movement directions, that is, other monitoring points that can be reached from the current monitoring point; Transition probability selection: Let r be the number of the ant, representing the rth ant, t be the time, i and j represent different nodes, i and j; At each step, calculate the transition probability of ant k moving from node i to node j , the formula is: in, is the pheromone concentration on the path from node i to node j at time t, is the weight parameter of pheromone concentration, which determines the degree of influence of pheromone concentration on ants' path selection; the larger it is, the more the ants will rely on pheromone concentration to choose the path. It is a heuristic factor, which is a kind of prompt information used to characterize the quality of the ant's choice from node i to node j. It is determined according to the correlation of the data of the two nodes or the physical distance factor. If the correlation or physical distance changes, the heuristic factor will change. is the weight parameter of the heuristic factor, which determines the influence of the heuristic information on the ant's path selection; s is an index variable, d represents the set of next nodes that ant k can choose. When the ant is at a node, it has multiple possible destinations, which constitute the set d. s traverses these possible destination nodes in turn and calculates the distance from the current node to each s node. The value of According to the calculated transfer probability, the ants make random selections through a random number generator. After each selection, the pheromone concentration of the path chosen by the ants is updated and calculated. in, Indicates the volatilization of pheromones. is the pheromone volatilization coefficient, which controls the speed of pheromone volatilization. As time goes by, the pheromone on the path will gradually decrease due to this coefficient. The value of determines the amount of calculation. The larger the The larger the amount of calculation, the smaller the The smaller the amount of calculation; Pheromone Increase: Indicates the path that the ants pass through Newly added pheromones, It is the increase in pheromones left by the ants that changes the attractiveness of the path to other ants; As the ants make their move choices step by step based on the transfer probability, the path of the ants from the starting monitoring point to each subsequent monitoring point is gradually constructed; The calculation formula of pheromone change rate A is: If the rate of change of pheromone concentration on all paths is less than a preset value of 0.1 in a number of consecutive iterations, the pheromone concentration is considered to be stable, the algorithm stops, and the normal operation path of the system is clear; When the monitoring unit monitors the real-time data of each part of the feeding equipment and substitutes it into real-time calculation, if the new path deviates from the original normal operating path, it is considered that the equipment has a fault or abnormal situation.

6. A converter high-level silo feeding control system as claimed in claim 1, characterized in that: The data storage module stores data with precise timestamps, ensuring that the timestamps of relevant real-time data are accurate to the second level, so as to more accurately record the time-related information of the loading process and provide support for subsequent analysis.

7. A converter high-level silo loading control system as claimed in claim 1, characterized in that: The user interface provided by the interactive module can display real-time data including the operation status of the feeding equipment and the real-time change information of the material demand, allowing the operator to set relevant calculation parameters for the system.

Images