Remote control method and system for electrolytic aluminum crown block

Abstract

The invention relates to the technical field of industrial automation. The electrolytic aluminum crown block remote control method comprises the steps that before the electrolytic aluminum crown block works, initial position detection is conducted through sensors arranged on all executing mechanisms of the electrolytic aluminum crown block, and coordinate systems, corresponding to electrolytic bath positions, of all the executing mechanisms are established; transmitting detection data to a remote control terminal in real time; an operation instruction is input on the terminal interface, and the remote control terminal generates a corresponding control instruction through a preset algorithm based on the input instruction and the received detection data; and sending the control instruction to a control system of the electrolytic aluminum crown block, and driving each execution mechanism of the electrolytic aluminum crown block to perform corresponding actions according to the control instruction so as to complete production operation. The problems that in the existing electrolytic aluminum industry, the labor intensity is large, the working efficiency and quality are affected due to an electrolytic multifunctional unit depending on manual operation, and the safety risk is high due to the severe production environment are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation, and more specifically, to a method and system for remotely controlling an electrolytic aluminum crane. Background Art

[0002] In modern industrial production, the electrolytic aluminum industry, as an important basic industry, has always attracted much attention in terms of production efficiency and safety. During the electrolytic aluminum production process, the electrolytic multifunctional unit undertakes key production operations such as crust breaking, electrode changing, slag skimming, feeding, tapping, and busbar lifting. Currently, most of these operations rely on the operation of machine operators and the cooperation of electrolysis workers, which has many drawbacks. On the one hand, the labor intensity is extremely high. Workers need to perform complicated operations on-site for a long time, which is prone to fatigue, thus affecting work efficiency and quality. On the other hand, the safety risk is high. The production environment of electrolytic aluminum is harsh, with harmful factors such as strong magnetism, dust, and HF gas. The physical health of workers is easily threatened, and the on-site equipment operates complexly. A slight mistake in manual operation may trigger a safety accident.

[0003] With the continuous progress of industrial automation and intelligent technologies, remote control technology has been widely applied in many fields, providing a new direction for the upgrading and transformation of the electrolytic aluminum industry. However, when applying remote control technology to electrolytic aluminum cranes, a series of technical problems are faced. Each operation process has high requirements for the positioning accuracy and operation control of equipment. For example, the positioning accuracy of the trolley and the tapping trolley needs to be better than ±100 mm, and the slewing accuracy of the tapping trolley needs to be better than ±10°. It is difficult to achieve such high-precision remote positioning and control. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for remotely controlling an electrolytic aluminum crane, aiming to solve the problems of high labor intensity, affected work efficiency and quality caused by relying on manual operation of the electrolytic multifunctional unit in the existing electrolytic aluminum industry, and high safety risks due to the harsh production environment.

[0005] The present invention is achieved through the following technical solutions:

[0006] A method for remotely controlling an electrolytic aluminum crane includes the steps of:

[0007] Before the electrolytic aluminum crane operates, perform initial position detection through sensors arranged on each actuator of the electrolytic aluminum crane, establish a coordinate system corresponding to each actuator and the electrolytic cell position, and transmit the detection data to the remote control terminal in real time;

[0008] According to the operation requirements, input operation instructions on the terminal interface. The remote control terminal generates corresponding control instructions based on the input instructions and the received detection data;

[0009] Send control instructions to the control system of the aluminum electrolysis crane. According to the control instructions, drive each actuator of the aluminum electrolysis crane to perform corresponding actions to complete the production operation.

[0010] Optionally, before the aluminum electrolysis crane operates, obtain the absolute position information of each actuator through sensors installed on each actuator of the aluminum electrolysis crane.

[0011] Optionally, the specific process of establishing the coordinate system corresponding to each actuator and the electrolytic cell positions is as follows:

[0012] Install an absolute encoder on the driven wheel shaft of the trolley, and evenly and spacedly set several RFID reset tags beside the workshop track. Obtain the absolute position information of the trolley through the absolute encoder and the RFID reset tags;

[0013] Install a laser ranging device at the running position of the tapping carriage, and obtain the absolute position information of the tapping carriage through the laser ranging device;

[0014] Install corresponding absolute position measurement units at the cylinder positions of the crust breaking mechanism, anode mechanism, slag skimming mechanism, feeding mechanism, tapping lifting mechanism, and frame lifting mechanism respectively to obtain the absolute position information of each mechanism;

[0015] According to the obtained absolute position information of each actuator, combined with the actual layout and dimension information of the electrolytic cell, establish the coordinate system corresponding to each actuator and the electrolytic cell positions through mathematical calculation and spatial solution.

[0016] Optionally, the specific process of the remote control terminal generating corresponding control instructions based on the input instructions and the received detection data is as follows:

[0017] Deeply analyze the input instructions, identify the operation type, extract the specific operation parameters included in the input instructions, and preprocess the received detection data;

[0018] According to the identified operation type, call the corresponding preset model from the algorithm model library built in the remote control terminal, input the extracted specific operation parameters into the corresponding preset model, and perform iterative calculation and optimization solution through the preset model to obtain the target operation parameters;

[0019] Generate control instructions for driving each actuator of the aluminum electrolysis crane based on the target operation parameters.

[0020] Optionally, the specific process of preprocessing the received detection data is as follows:

[0021] Detect missing values in the detection data. If there are missing values, use the data interpolation method to fill in the missing values according to the change trend of adjacent valid data.

[0022] Detect outliers. Based on the 3σ principle of statistics, identify data outside the range of the mean ± 3 times the standard deviation as outliers, and correct or eliminate the outliers by comparing with historical normal data and combining on-site actual working conditions for judgment;

[0023] Normalize the data, mapping detection data with different dimensions to a specified interval.

[0024] Optionally, the specific process of inputting the extracted specific operation parameters into the corresponding preset model, performing iterative calculations and optimization solutions through the preset model, and obtaining the target operating parameters is as follows:

[0025] Input the extracted specific operation parameters as the initial values into the preset model. The preset model constructs an objective function based on the dynamic equation, kinematic constraints, and operation process requirements;

[0026] During the iterative calculation process, the preset model calculates the performance index under the current parameters according to the objective function and adjusts the operation parameters using an intelligent optimization algorithm;

[0027] After each iterative calculation, determine whether the preset convergence condition is met. If it is met, output the operation parameters at this time as the target operating parameters; if not, continue the next round of iterative calculation until the convergence condition is met.

[0028] Optionally, the construction process of the preset model is as follows:

[0029] Construct a corresponding initial mathematical model according to the dynamic characteristics, kinematic laws, and specific requirements of the operation process of each actuator of the aluminum electrolysis crane;

[0030] Obtain historical operation data and simulation experiment data of each actuator under different operation conditions, screen out characteristic parameters closely related to the operation state of each actuator, and obtain a characteristic parameter set;

[0031] Use the characteristic parameter set as the input variable, train with machine learning algorithms, and determine the parameters in the initial mathematical model by iteratively adjusting the weights and biases of the parameters in the initial mathematical model to fit the non-linear relationship between the input variable and the operation parameters of the actuator, thereby obtaining the preset model.

[0032] Optionally, during the construction process of the preset model, incorporate the interference factors in the aluminum electrolysis production environment and the performance changes caused by long-term operation into the preset model in the form of correction coefficients or additional constraint conditions, and optimize the preset model in combination with the physical characteristic parameters of each actuator.

[0033] Optionally, the operating states of each actuator and the job execution data during the execution process are fed back to the remote control terminal;

[0034] Compare the actual operating states of each actuator and the job execution data fed back with the theoretical data corresponding to the control instructions;

[0035] If the actual positioning accuracy or actual rotation accuracy of the key parts of the overhead crane does not meet the preset accuracy requirements, an error correction instruction is generated and sent to the control system of the electrolytic aluminum overhead crane, and the actuator is fine-tuned according to the error correction instruction.

[0036] Based on the same inventive concept, the present invention also provides a remote control system for an electrolytic aluminum overhead crane for implementing the remote control method of the electrolytic aluminum overhead crane, including:

[0037] A detection module, which is arranged on each actuator of the electrolytic aluminum overhead crane and is used for performing initial position detection before the electrolytic aluminum overhead crane operates to obtain the absolute position information of each actuator;

[0038] A remote control terminal, which is used for receiving the detection data transmitted by the detection module, receiving the operation instructions input by the operator, generating corresponding control instructions through a preset algorithm based on the input instructions and the received detection data, and receiving the feedback of the operating states of each actuator and the job execution data during the execution process, comparing the feedback data with the theoretical data corresponding to the control instructions, and generating an error correction instruction if the actual positioning accuracy or actual rotation accuracy of the key parts of the overhead crane does not meet the preset accuracy requirements;

[0039] A communication module, which is used to realize data transmission between the detection module and the remote control terminal and between the remote control terminal and the control system of the electrolytic aluminum overhead crane;

[0040] The control system of the electrolytic aluminum overhead crane is used for receiving the control instructions and error correction instructions sent by the remote control terminal, driving each actuator of the electrolytic aluminum overhead crane to perform corresponding actions according to the instructions to complete the production operation, and feeding back the operating states of each actuator and the job execution data to the remote control terminal.

[0041] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0042] Significantly reduce the labor intensity: By adopting the remote control method, the operator can input the operation instructions at the remote control terminal without being on-site, which greatly reduces the labor intensity, enables the staff to get rid of the harsh working environment, and can devote more energy to work, thereby ensuring the stability of work efficiency and quality.

[0043] Significantly improve production safety: The electrolytic aluminum production environment is full of harmful factors such as strong magnetism, dust, HF gas, etc., which pose a great threat to the health of operators. In addition, the operation of on-site equipment is complex, and manual operation errors can easily lead to safety accidents. Through remote control, operators stay away from dangerous areas, effectively avoiding the harm of harmful factors to the body. At the same time, it reduces the possibility of safety accidents caused by manual on-site operation errors, providing a strong guarantee for the safe production of enterprises.

[0044] Achieve high-precision operation control: By setting sensors on each actuator for initial position detection, establishing a coordinate system corresponding to the actuator and the electrolytic cell position, and combining the preset algorithm of the remote control terminal to generate control instructions, it is possible to accurately drive the various actuators of the electrolytic aluminum overhead crane, achieve high-precision remote positioning and control, meet the strict requirements of production operations for precision, and help improve product quality and production efficiency.

[0045] Improve the level of intelligent production: Introducing remote control technology into electrolytic aluminum overhead crane operations makes the input of operating instructions more convenient and efficient, and can receive detection data in real time. Precise control instructions are generated through preset algorithms, realizing intelligent management of the production process. This not only improves the overall intelligent level of electrolytic aluminum production, but also helps enterprises adapt to the needs of future industrial development and enhance market competitiveness.

[0046] Facilitates production management and scheduling: The remote control terminal can obtain real-time detection data from each actuator of the electrolytic aluminum overhead crane. Managers can intuitively understand the equipment's operating status and production progress. Based on this information, they can conduct more reasonable production management and scheduling, optimize production processes, and improve equipment utilization, thereby maximizing production benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 Schematic diagram of the flow of a remote control method for an electrolytic aluminum overhead crane according to an embodiment of the present invention;

[0048] Figure 2 This is a schematic structural diagram of a remote control system for an electrolytic aluminum overhead crane according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following is a specific implementation method with reference to the accompanying drawings.

[0050] Reference Figure 1 A remote control method for an electrolytic aluminum overhead crane comprises the following steps:

[0051] Step 1: Before the electrolytic aluminum overhead crane operates, perform initial position detection through sensors installed on each actuator of the electrolytic aluminum overhead crane, establish a coordinate system corresponding to each actuator and the electrolytic cell position, and transmit the detection data to the remote control terminal in real time.

[0052] In some embodiments, before the operation of the electrolytic aluminum crane, absolute position information of each actuator is obtained through sensors provided on each actuator of the electrolytic aluminum crane.

[0053] An absolute encoder can be installed on the driven wheel shaft of the trolley. The absolute encoder can convert the rotation angle of the trolley wheel into a digital signal. By precisely measuring the number of wheel rotations and the angle, the displacement of the trolley on the track can be calculated. Since the circumference of the wheel is known, according to the number of pulses fed back by the encoder, the distance the trolley moves can be accurately obtained, thereby acquiring the absolute position information of the trolley. A number of RFID reset tags are evenly spaced beside the workshop track. When the trolley moves past the RFID reset tag, the RFID reader installed on the trolley can read the tag information. The positions of these tags are precisely set in advance. Through interaction with the tags, the trolley can perform position calibration to eliminate position deviations caused by factors such as wheel slippage and encoder cumulative error, ensuring the accuracy and reliability of the acquired absolute position information.

[0054] A laser ranging device can be installed at the operating position of the tapping carriage. The laser ranging device emits a laser beam. After the laser beam hits a reflecting object (such as the workshop wall or a specially set reflector), it is reflected back and received by the laser ranging device. According to the propagation speed and round-trip time of the laser, the distance between the tapping carriage and the reflecting object can be accurately calculated. By setting multiple reflecting objects in different directions and combining mathematical methods such as trigonometric functions, the absolute position information of the tapping carriage in space can be determined.

[0055] Absolute position measurement units can be installed respectively at the cylinder positions of the crust breaking mechanism, anode mechanism, slag fishing mechanism, feeding mechanism, aluminum tapping lifting mechanism and frame lifting mechanism. These absolute position measurement units can adopt magnetostrictive displacement sensors, etc. Magnetostrictive displacement sensors utilize the magnetostrictive principle. When an electric current pulse is generated by the electronic module inside the sensor, a magnetic field that propagates along the waveguide wire will be generated in the waveguide wire. This magnetic field interacts with the magnetic field of the permanent magnet in the float, generating a strain pulse. By measuring the time difference between the strain pulse and the electric current pulse, the position of the float, that is, the length of the piston rod of the cylinder extending or retracting, can be accurately calculated, thereby obtaining the absolute position information of each mechanism. Through the combined application of multiple high-precision sensors, the strict requirements for positioning accuracy of each actuator of the electrolytic aluminum crane can be met. For example, the positioning accuracy of the trolley and the aluminum tapping trolley can reach better than ±100 mm, and the slewing accuracy of the aluminum tapping trolley is better than ±10°. This provides a basis for subsequent precise production operations, ensuring that operations such as crust breaking and pole changing can be carried out accurately and improving production quality. Different types of sensors cooperate with each other. For example, the combination of an absolute encoder and an RFID reset tag effectively reduces the error accumulation and failure risk that may occur in a single sensor. Even if a certain sensor has a short-term abnormality, the information of other sensors can still ensure the accurate judgment of the actuator position by the system, enhancing the reliability of the entire remote control system and reducing problems such as production interruption and equipment damage caused by inaccurate position detection.

[0056] In some embodiments, the specific process of establishing a coordinate system corresponding to each actuator and the electrolytic cell position is as follows:

[0057] An absolute encoder is installed on the driven wheel shaft of the large vehicle, and several RFID reset tags are evenly spaced beside the workshop track. The absolute position information of the large vehicle is obtained through the absolute encoder and the RFID reset tags. A laser ranging device is installed at the running position of the tapping trolley, and the absolute position information of the tapping trolley is obtained through the laser ranging device. At the oil cylinder positions of the crust breaking mechanism, anode mechanism, slag skimming mechanism, feeding mechanism, tapping lifting mechanism and frame lifting mechanism, corresponding absolute position measuring units are installed respectively to obtain the absolute position information of each mechanism. According to the obtained absolute position information of each actuator, combined with the actual layout and dimension information of the electrolytic cell, including the row and column distribution, spacing, etc. of the electrolytic cell, and the dimension information such as the length, width and height of a single electrolytic cell, a coordinate system corresponding to each actuator and the electrolytic cell position is established through mathematical calculation and spatial solution. Taking a fixed point in the workshop as the coordinate origin, such as a corner of the workshop. For the position of the large vehicle, according to the number of pulses of the absolute encoder and the position of the RFID reset tag, combined with parameters such as the diameter of the large vehicle wheel, the displacement of the large vehicle along the track direction is calculated through the relationship between the circumference and the number of pulses, and the X-axis coordinate of the large vehicle in the workshop coordinate system is determined. For the tapping trolley, the laser ranging device measures the distance between the trolley and a specific reference plane. Combined with the geometric relationship between its running track and the large vehicle track, the coordinates of the tapping trolley in the local coordinate system with the large vehicle as the reference are calculated through trigonometric functions, etc., and then converted to the workshop global coordinate system. The data of the absolute position measuring unit of each mechanism determines the position coordinates of each mechanism in its own local coordinate system according to the geometric relationship between its installation position and the moving parts of the corresponding mechanism. Considering the actual layout of the electrolytic cells, the electrolytic cells are numbered according to rows and columns. According to the row and column spacing of the electrolytic cells and their relative position to the coordinate origin, the position range of each electrolytic cell in the workshop global coordinate system is determined. For each actuator, through coordinate transformation and spatial geometric relationship, its coordinates in its own local coordinate system are converted to the coordinate system corresponding to the electrolytic cell. For example, when the large vehicle moves above a certain row of electrolytic cells, the center line of this row of electrolytic cells is taken as the new X-axis, and the direction perpendicular to this center line and pointing to one side of the workshop is taken as the Y-axis to establish the local coordinate system of this row of electrolytic cells. Then, the coordinates of the tapping trolley and each mechanism in the workshop global coordinate system are converted to the local coordinate system of this row of electrolytic cells through coordinate transformation operations such as translation and rotation, so as to establish a coordinate system corresponding to each actuator and the electrolytic cell position. The coordinate system established through precise mathematical calculation and spatial solution enables each actuator to accurately know its relative position to the electrolytic cell position, greatly improving the operation positioning accuracy. For example, during the crust breaking operation, the crust breaking mechanism can accurately position to the target position of the electrolytic cell, avoiding poor crust breaking effect caused by positioning deviation and ensuring the quality of the production operation. The positioning accuracy of the large vehicle and the tapping trolley can be stably better than ±100mm, and the swing accuracy of the tapping trolley is better than

[0058] ±10°, meeting the high-precision requirements of each operation process for the equipment.

[0059] Step 2: According to the operation requirements, input operation instructions on the terminal interface. The remote control terminal generates corresponding control instructions based on the input instructions and the received detection data.

[0060] In some embodiments, the specific process of the remote control terminal generating corresponding control instructions based on the input instructions and the received detection data is as follows:

[0061] Deeply analyze the input instructions, identify the operation type, extract the specific operation parameters included in the input instructions, and preprocess the received detection data; the remote control terminal uses natural language processing (NLP) technology and a specific syntax analyzer to deeply analyze the input instructions. For example, when an operator inputs an instruction like "Perform aluminum tapping operation, the aluminum tapping amount is 500 kg, and the moving speed of the aluminum tapping trolley is set to 0.5 m / s", the system first determines the operation type as "aluminum tapping operation" through semantic recognition, and then uses regular expressions or a dedicated parameter extraction algorithm to accurately extract specific operation parameters such as "aluminum tapping amount 500 kg" and "moving speed of the aluminum tapping trolley 0.5 m / s".

[0062] According to the identified operation type, call the corresponding preset model from the algorithm model library built in the remote control terminal, input the extracted specific operation parameters into the corresponding preset model, and perform iterative calculation and optimization solution through the preset model to obtain the target operation parameters;

[0063] Based on the target operation parameters, generate control instructions for driving each actuator of the electrolytic aluminum crane.

[0064] In some embodiments, the specific process of preprocessing the received detection data is as follows:

[0065] Detect missing values in the detection data. If there are missing values, use the data interpolation method to fill the missing values according to the change trend of adjacent valid data; linear interpolation can be used. Suppose the position data of the aluminum tapping trolley detected by a certain sensor is missing at a certain moment. The system will calculate the missing value position according to the positions and time intervals of the two adjacent valid data points of this sensor according to the linear relationship.

[0066] Detect outliers. Based on the 3σ principle of statistics, identify data outside the range of the mean ± 3 times the standard deviation as outliers, and correct or eliminate the outliers by comparing with historical normal data and combining with the actual on-site working conditions. For example, when detecting the cylinder pressure data of the crust breaking mechanism, if a certain pressure value significantly deviates from the normal range, the system will first compare it with the historical normal pressure data to determine whether the pressure value conforms to the pressure change law of the mechanism under similar working conditions. If not, it will combine the actual on-site working conditions, such as the current state of the electrolytic cell, etc., to correct, eliminate or make a fault judgment on the outlier.

[0067] Normalize the data, and map the detection data with different dimensions to a specified interval. For example, voltage values, current values, position coordinate values, etc., are mapped to the [0, 1] interval through the maximum-minimum normalization method.

[0068] In some embodiments, the specific process of inputting the extracted specific operation parameters into the corresponding preset model and performing iterative calculation and optimization solution through the preset model to obtain the target operation parameters is as follows:

[0069] Input the extracted specific operation parameters as the initial values into the preset model, and the preset model constructs an objective function based on the dynamic equation, kinematic constraints, and operation process requirements.

[0070] In the electrolytic aluminum crane system, different actuators have their own unique dynamic and kinematic characteristics. For example, the movement of the trolley is mainly a linear movement along the track, and its dynamic characteristics can be described by Newton's second law; in addition to the linear movement, the tapping trolley may also have a rotary movement, and factors such as the moment of inertia need to be considered; while the crust breaking mechanism, anode mechanism and other hydraulically driven actuators are closely related to parameters such as the pressure and flow rate of the cylinder. Different operation types (such as tapping operation, crust breaking operation, etc.) have different process requirements. Taking the tapping operation as an example, it is required that the tapping amount is accurate and the moving speed of the tapping trolley is appropriate; the crust breaking operation requires that the crust breaking mechanism can accurately locate the target position of the electrolytic cell and has sufficient impact force. According to the above dynamic, kinematic characteristics and operation process requirements, construct an objective function that can comprehensively reflect the system performance. The objective function is usually a function of the operation parameters, and by optimizing this function, the optimal operation parameters that meet the operation requirements can be obtained. In actual operation, the operation of the system is restricted by various constraint conditions, such as the power limit of the motor, the stroke limit of the cylinder, etc. These constraint conditions need to be considered in the construction of the objective function. Specifically, the objective function for the trolley movement is constructed as follows:

[0071] Let the mass of the trolley be m, the driving force be F, the resistance be f, the initial position be x0, the initial velocity be v0, and the target position be x d, the target time is t1. According to Newton's second law, we can obtain where v is the speed of the large vehicle. The kinematic equation of the large vehicle is The operation process requires the large vehicle to reach the target position x d within the target time t d , and at the same time, the energy consumption should be minimized. Therefore, the objective function in Equation (1) can be constructed as follows:

[0072]

[0073] where w1 and w2 represent the weight coefficients, which are used to adjust the importance of the position error and the energy consumption in the objective function; the constraint conditions of Equation (1) are shown in Equation (2) as follows:

[0074]

[0075] where F max represents the maximum driving force of the large vehicle; v max represents the maximum speed of the large vehicle.

[0076] The objective function for the movement of the tapping trolley is constructed as follows:

[0077] Let the moment of inertia of the tapping trolley be I a , the turning moment be M, the initial angle be θ0, the initial angular velocity be ω0, and the target angle be θ d , and the target time be t2. According to the law of rotation where ω a is the angular velocity of the tapping trolley. The kinematic equation of the tapping trolley is The operation process requires the tapping trolley to reach the target angle θ d within the target time t2, and at the same time, the smoothness of the turning process should be ensured. Therefore, the objective function in Equation (3) can be constructed as follows:

[0078]

[0079] where w3 and w4 represent the weight coefficients, which are used to adjust the importance of the angle error and the angular velocity smoothness in the objective function; the constraint conditions of Equation (3) are shown in Equation (4) as follows:

[0080]

[0081] where M max represents the maximum turning moment of the tapping trolley; ω max represents the maximum angular velocity of the tapping trolley.

[0082] The objective function for the operation of the crust breaking mechanism is constructed as follows:

[0083] Let the pressure of the cylinder of the crust breaking mechanism be P, the displacement of the piston rod be s, the initial displacement be s0, and the target displacement be s d , and the target time be t3. The dynamic equation of the cylinder can be expressed as where A represents the piston area of the cylinder, P0 represents the initial pressure of the cylinder, m shell represents the equivalent mass of the crust breaking mechanism, and f shell represents the resistance of the crust breaking mechanism.

[0084] The operation process requires the crust breaking mechanism to reach the target displacement s within the target time t3 d , and have sufficient impact force. Therefore, the objective function in Equation (5) can be constructed as follows:

[0085] J3 = w5(s(t3) - s d ) 2 + w6(P - P threshold )(5)

[0086] where w5 and w6 represent the weight coefficients, used to adjust the importance of displacement error and pressure satisfaction degree in the objective function, and P threshold represents the minimum pressure required for crust breaking. The constraint conditions of Equation (5) are shown in Equation (6) as follows:

[0087]

[0088] where P max represents the maximum pressure of the cylinder of the crust breaking mechanism; s max represents the maximum displacement of the piston rod of the crust breaking mechanism.

[0089] The objective function of the anode mechanism operation is constructed as follows:

[0090] The anode mechanism mainly drives the anode plate to move up and down and other operations through the cylinder. Let the pressure of the cylinder of the anode mechanism be P a , the displacement of the piston rod be s a , the equivalent mass of the anode mechanism be m a , the piston area of the cylinder be A a , and the resistance be f a . According to Newton's second law, the dynamic equation of the cylinder is where P a0 is the initial pressure of the cylinder. The kinematic equation is v a represents the moving speed of the anode mechanism. The anode mechanism needs to move accurately to the specified position to ensure the correct cooperation between the anode plate and the electrolytic cell, and at the same time avoid damaging the equipment due to too fast speed during the moving process.

[0091] Let the target displacement of the anode mechanism be sad If the target time is t4, the objective function for the anode mechanism operation is as shown in the following equation (7):

[0092]

[0093] Wherein, w7 and w8 represent weight coefficients, respectively adjusting the importance of displacement error and velocity smoothness in the objective function. The constraint conditions of equation (7) are as shown in the following equation (8):

[0094]

[0095] Wherein, P a_max represents the maximum pressure of the anode mechanism oil cylinder; s a_max represents the maximum displacement of the anode mechanism piston rod; v a_max represents the maximum movement speed of the anode mechanism.

[0096] The objective function for the slag scraping mechanism operation is constructed as follows:

[0097] Let the displacement of the slag scraping mechanism in the horizontal direction be x l , and the displacement in the vertical direction be y l , its driving force in the horizontal direction be F lx , and the driving force in the vertical direction be F ly , and the total mass of the slag scraping mechanism be m l . The dynamic equation in the horizontal direction is The dynamic equation in the vertical direction is Wherein, f lx and f ly are the resistances in the horizontal and vertical directions respectively, and g is the acceleration due to gravity. The kinematic equation is v lx represents the horizontal movement speed of the slag scraping mechanism, and v ly represents the vertical movement speed of the slag scraping mechanism.

[0098] Let the horizontal target displacement be x ld , and the vertical target displacement be y ld , and the target time be t5, then the objective function is as shown in the following equation (9):

[0099] T

[0100] Wherein, w9 and w 10 are weight coefficients, respectively adjusting the importance of position error and velocity smoothness. The constraint conditions of equation (9) are as shown in the following equation (10):

[0101]

[0102] Wherein, F lx_maxRepresents the maximum horizontal driving force of the slag scraping mechanism; F ly_max Represents the maximum vertical driving force of the slag scraping mechanism; x l_max Represents the maximum horizontal displacement of the slag scraping mechanism; y l_max Represents the maximum vertical displacement of the slag scraping mechanism; v lx_max Represents the maximum horizontal speed of the slag scraping mechanism; v ly_max Represents the maximum vertical speed of the slag scraping mechanism.

[0103] The objective function of the blanking mechanism operation is constructed as follows:

[0104] Let the blanking flow rate of the blanking mechanism be q, and the blanking time be t q , the torque of the motor driving the blanking mechanism be T, and the moment of inertia of the motor be I b . The dynamic equation of the motor is where, T f represents the load torque of the motor, ω b represents the angular velocity of the motor, and there is a functional relationship between q and ω b q = kω b , k is the proportionality coefficient.

[0105] Let the target blanking flow rate be q d , and the target blanking time be t6, then the objective function is shown in the following formula (11):

[0106] J6 = w 11 (q - q d ) 2 + w 12 (t q - t6) 2 (11)

[0107] where, w 11 and w 12 represent the weight coefficients, respectively adjusting the importance of the flow rate error and the time error. The constraint conditions of formula (11) are shown in the following formula (12):

[0108]

[0109] where, T max represents the maximum torque of the blanking mechanism motor; q max represents the maximum blanking volume of the blanking mechanism; t q_max represents the maximum blanking time of the blanking mechanism.

[0110] The objective function of the aluminum tapping and lifting mechanism is constructed as follows:

[0111] The aluminum tapping and lifting mechanism is mainly used to lift the aluminum tapping ladle. Let the pulling force of the lifting mechanism be F h , and the mass of the aluminum tapping ladle be m h, the lifted height is h. According to Newton's second law The kinematic equation is v h represents the lifting speed of the aluminum lifting mechanism. The aluminum lifting mechanism needs to accurately lift the aluminum ladle to the specified height, and the lifting process should be stable.

[0112] Let the target lifting height be h d , the target time be t7, then the objective function is shown in the following formula (13):

[0113]

[0114] where, w 13 and w 14 are weight coefficients, respectively adjusting the importance of height error and speed smoothness. The constraint conditions of formula (13) are shown in the following formula (14):

[0115]

[0116] where, F n_max represents the maximum pulling force of the aluminum lifting mechanism; h max represents the maximum lifting height of the aluminum lifting mechanism; v h_max represents the maximum lifting speed of the aluminum lifting mechanism.

[0117] The objective function of the frame lifting mechanism is constructed as follows:

[0118] The frame lifting mechanism is used to lift the entire frame structure. Let the total mass of the frame be m f , the pulling force of the lifting mechanism be F f , the height lifted by the frame be z. The dynamic equation is The kinematic equation is v z represents the lifting speed of the frame lifting mechanism. The frame lifting mechanism needs to accurately lift the frame to the specified height and ensure the stability of the lifting process at the same time.

[0119] Let the target lifting height be z d , the target time be t8, then the objective function is shown in the following formula (15):

[0120]

[0121] where, w 15 and w 16 represent weight coefficients, respectively adjusting the importance of height error and speed smoothness. The constraint conditions of formula (15) are shown in the following formula (16):

[0122]

[0123] Among them, F f_max represents the maximum pulling force of the frame lifting mechanism; z max represents the maximum lifting height of the frame lifting mechanism; v z_max represents the maximum lifting speed of the frame lifting mechanism.

[0124] Combining the objective functions of all actuators, the objective function of the entire aluminum electrolysis crane system is obtained, as shown in the following formula (17):

[0125]

[0126] Among them, J i represents the objective function corresponding to each mechanism. By inputting the extracted specific operation parameters as initial values into this comprehensive objective function and combining with the constraint conditions for optimization, the target operation parameters that meet the operation requirements are obtained.

[0127] During the iterative calculation process, the preset model calculates the performance index under the current parameters according to the objective function and adjusts the operation parameters using the intelligent optimization algorithm; taking the particle swarm optimization (PSO) algorithm as an example:

[0128] S101: Determine the scale N of the particle swarm, that is, the number of particles. Randomly initialize the position x i and velocity v i for each particle, where i = 1, 2,..., N. The position x i of the particle represents a set of operation parameters, and the velocity v i represents the moving speed of the particle in the parameter space.

[0129] S102: For each particle, substitute its position x i into the objective function of the entire aluminum electrolysis crane system, that is, formula (17), and calculate its fitness value f(x i ). The fitness value reflects the performance index of the system under this set of operation parameters.

[0130] S103: For each particle, compare its current fitness value f(x i ) with the historical optimal fitness value f(Pb i ). If f(x i ) < f(Pb i ), then update the individual optimal position Pb i = x i . Compare the individual optimal fitness values of all particles and find the minimum value among them. The corresponding particle position is the global optimal position Gb i .

[0131] S104: Update the velocity vi and position xi of each particle according to the following formulas (18) and (19) respectively:

[0132] v i (n + 1) = wv i (n) + c1r1(Pb i -x i (n)) + c2r2(Gb i -x i (n)) (18)

[0133] x i (n + 1) = x i (n) + v i (n + 1) (19)

[0134] Among them, w represents the inertia weight, which is used to control the degree of inheritance of the particle's own velocity; c1 and c2 represent acceleration constants, usually taking positive values; r1 and r2 represent random numbers in the range of [0, 1]; n represents the current iteration number.

[0135] S105: Determine whether the preset convergence condition is satisfied, such as reaching the maximum number of iterations or the change in the objective function value is less than a certain threshold. If the convergence condition is satisfied, output the global optimal position Gb i as the target operating parameter; otherwise, return to step 2 to continue the next round of iterative calculation.

[0136] In some embodiments, the construction process of the preset model is as follows:

[0137] According to the dynamic characteristics, kinematic laws of each actuator of the aluminum electrolysis crane and the specific requirements of the operation process, construct the corresponding initial mathematical model. Taking the trolley as an example, its dynamic equation is The kinematic equation is It can be written in state - space form as shown in the following formula (20):

[0138]

[0139] For the initial mathematical model of the entire aluminum electrolysis crane system, it can be expressed as a system of equations containing the state equations of multiple actuators, as shown in the following formula (21):

[0140]

[0141] where x i represents the state variables (such as position, velocity, etc.) of the i - th actuator, u i represents the input variables (such as driving force, torque, etc.) of the i - th actuator, f i represents the dynamic and kinematic equations of the i - th actuator.

[0142] Obtain the historical operation data and simulation experiment data of each actuator under different operating conditions, screen out the characteristic parameters closely related to the operating state of each actuator, and obtain a set of characteristic parameters. Assume that R characteristic parameters closely related to the operating state of the actuator are screened out from the historical operation data and simulation experiment data, denoted as p1, p2, …, p R , then the set of characteristic parameters can be expressed as {p1, p2, …, p R}.

[0143] Take the set of characteristic parameters as the input variable, and use machine learning algorithms for training. By iteratively adjusting the weights and biases of the parameters in the initial mathematical model, fit the non-linear relationship between the input variable and the operating parameters of the actuator, determine the parameters in the initial mathematical model, and obtain a preset model.

[0144] Taking a neural network as an example, assume that the input layer has R neurons (corresponding to R characteristic parameters), the hidden layer has h neurons, and the output layer has m neurons (corresponding to m operating parameters of the actuator). The weight matrix from the input layer to the hidden layer is W1, the weight matrix from the hidden layer to the output layer is W2, the bias vector of the hidden layer is b1, and the bias vector of the output layer is b2. For the input characteristic parameter vector P = [p1, p2, …, p R T , the expression of the output vector H of the hidden layer is shown in the following formula (22):

[0145] H = σ(W1P + b1) (22)

[0146] where σ is an activation function (such as Sigmoid function, ReLU function, etc.);

[0147] The loss function is shown in the following formula (23):

[0148]

[0149] where Q is the number of training samples, Y i is the predicted value of the model, is the actual value. Update the weights and biases through an optimization algorithm (such as stochastic gradient descent algorithm) to minimize the loss function, as shown in the following formula (24):

[0150]

[0151] where α is the learning rate. After multiple iterative trainings, when the loss function converges, the obtained weight and bias values determine the preset model.

[0152] ​In some embodiments, during the preset model construction process, the interference factors in the electrolytic aluminum production environment and the performance changes brought about by long-term operation are incorporated into the preset model in the form of correction factors or additional constraint conditions. Combining the physical characteristic parameters of each actuator, the preset model is optimized. The magnetic field intensity and direction in the electrolytic aluminum production environment are monitored in real time through a magnetic field sensor. Analyze the influence law of strong magnetic field on the sensor signal transmission and motor operation of each actuator of the crane, and quantify it as a magnetic field interference coefficient. For example, it is found through research that for every increase of X Tesla in the magnetic field intensity, the sensor signal will have a deviation of Y%. Based on this, the correction factor is determined.

[0153] Use a dust concentration sensor to monitor the dust concentration in the working area. Dust will affect the heat dissipation performance of the equipment and the wear degree of mechanical components. Through experiments and historical data statistics, determine the changes in the equipment performance parameters under different dust concentrations. For example, when the dust concentration reaches Z mg / m 3 ³, the resistance of the motor will increase by A Ω, thus affecting its output power, and this is converted into a corresponding correction factor.

[0154] Monitor the HF gas concentration and analyze its corrosion rate on metal components. Through material corrosion experiments, obtain the corrosion depth data of metal materials under different HF gas concentrations and times. For example, when the HF gas concentration is B ppm, after C hours, the thickness of a certain key component decreases by D mm, and this corrosion effect is converted into a correction factor.

[0155] Collect the historical data of the long-term operation of each actuator of the crane, including parameters such as the rotation speed and torque of the motor, and the pressure and displacement of the oil cylinder. Analyze the change trend of these parameters with the operation time. For example, after long-term operation of the motor, due to bearing wear, its efficiency decreases by E%, and this performance change is quantified as a time-related correction factor.

[0156] In the dynamic equation and kinematic constraints of the preset model, introduce the above-determined correction coefficients. For example, in the dynamic equation of the motor, multiply the magnetic field interference coefficient, the coefficient of change in the motor resistance caused by dust influence, and the coefficient of reduction in long-term operation efficiency by parameters such as the electromagnetic torque and moment of inertia of the motor to correct the output characteristics of the motor. In the kinematic equation of the oil cylinder, adjust the calculation of the stroke and speed of the oil cylinder according to the coefficient of change in component size caused by HF gas corrosion. Add additional constraint conditions according to the performance changes of the equipment after long-term operation. For example, due to wear of mechanical components, the movement range of some actuators may decrease, and corresponding position constraint conditions are set in the model. When the vibration of the motor intensifies after long-term operation, add constraint conditions for vibration limitation in the model to ensure that the operating parameters calculated by the model conform to the actual operating capacity of the equipment. Measure and record in detail the physical characteristic parameters of each actuator of the overhead crane, including the rated power, rated speed, and internal resistance of the motor, the cylinder diameter, piston rod diameter, and working pressure of the oil cylinder, and the transmission ratio and friction coefficient of mechanical transmission components. Substitute the physical characteristic parameters into the preset model, and iteratively adjust the weights and biases of various parameters in the model through machine learning algorithms. For example, according to the actual internal resistance and rated power of the motor, adjust the weights of the parameters related to the energy consumption and output power of the motor in the model to make the model more accurately reflect the operating conditions of the motor under different working conditions. Use the physical dimension parameters of the oil cylinder to optimize the deviation values of the calculation of the movement speed and pressure of the oil cylinder in the model to improve the accuracy of the model's control of the oil cylinder's actions. By integrating the interference factors in the aluminum electrolysis production environment and the performance changes during long-term operation into the preset model, various complex situations in actual production are considered, enabling the model to more accurately describe the operating states of each actuator of the overhead crane, thereby improving the accuracy of the target operating parameters calculated by the model and meeting the requirements of high-precision positioning and control in production operations.

[0157] Step 3: Send the control instruction to the control system of the aluminum electrolysis overhead crane, and drive each actuator of the aluminum electrolysis overhead crane to perform corresponding actions according to the control instruction to complete the production operation.

[0158] In some embodiments, feedback the operating states and operation execution data of each actuator during the execution process to the remote control terminal;

[0159] Compare the actual operating status of each actuator in the feedback and the operation execution data with the theoretical data corresponding to the control instruction; after receiving the feedback data, the remote control terminal first classifies and matches it with the theoretical data corresponding to the control instruction. For example, compare the actual running displacement data of the trolley with the expected displacement data in the control instruction, and compare the actual rotation angle data of the tapping trolley with the theoretical rotation angle data. For position data, use the difference calculation method to calculate the deviation value between the actual position and the theoretical position; for angle data, also calculate the difference between the actual angle and the theoretical angle. At the same time, set a reasonable error tolerance range, which is determined according to the positioning accuracy and operation control requirements of each operation process of the aluminum electrolysis crane for the equipment. For example, the error tolerance range for the positioning accuracy of the trolley and the tapping trolley is

[0160] ±100 mm, and the error tolerance range for the rotation accuracy of the tapping trolley is ±10° etc.

[0161] If the actual positioning accuracy or actual rotation accuracy of the key parts of the crane does not meet the preset accuracy requirements, that is, the deviation between the actual data and the theoretical data exceeds the error tolerance range, an error correction instruction will be generated. The remote control terminal will start the error correction instruction generation program, which calculates the control parameters that need to be fine-tuned for the actuator according to the magnitude and direction of the deviation value, combined with the kinematic and dynamic models of each actuator of the aluminum electrolysis crane. For example, if the actual position of the trolley is 50 mm to the left of the theoretical position, according to parameters such as the characteristics of the driving motor of the trolley and the friction between the wheels and the track, calculate that a certain amount of current needs to be increased to the driving motor to make the trolley move to the right to correct the deviation. Send the error correction instruction to the control system of the aluminum electrolysis crane, and fine-tune the actuator according to the error correction instruction. For the actuators driven by motors (such as the traveling motors of the trolley and the tapping trolley), the control system changes the speed and direction of the motor by adjusting parameters such as the input voltage, current or frequency of the motor, so as to realize the position fine-tuning of the actuator. For the actuators driven by hydraulic pressure (such as the oil cylinders of the crust breaking mechanism and the anode mechanism), the control system adjusts the opening of the hydraulic valve to control the flow and pressure of the hydraulic oil, and then adjusts the telescopic amount of the oil cylinder to realize the fine-tuning of the position or angle of the mechanism.

[0162] Based on the same inventive concept, corresponding to the positioning tooling in any of the above embodiments, referring to Figure 2 This invention provides an aluminum electrolysis crane remote control system for realizing the above-mentioned aluminum electrolysis crane remote control method, including:

[0163] A detection module, which is arranged on each actuator of the aluminum electrolysis crane and is used to perform initial position detection before the aluminum electrolysis crane operates to obtain the absolute position information of each actuator;

[0164] The remote control terminal is used to receive the detection data transmitted by the detection module, receive the operation instructions input by the operator, generate corresponding control instructions based on the input instructions and the received detection data through a preset algorithm, and receive the operation status and job execution data feedback of each actuator during the execution process. Compare the feedback data with the theoretical data corresponding to the control instructions. If the actual positioning accuracy or actual rotation accuracy of the key parts of the overhead crane does not meet the preset accuracy requirements, an error correction instruction is generated;

[0165] The communication module is used to realize data transmission between the detection module and the remote control terminal, and between the remote control terminal and the electrolytic aluminum overhead crane control system;

[0166] The electrolytic aluminum overhead crane control system is used to receive the control instructions and error correction instructions sent by the remote control terminal, drive each actuator of the electrolytic aluminum overhead crane to perform corresponding actions according to the instructions, complete the production operation, and feedback the operation status and job execution data of each actuator to the remote control terminal.

Claims

1. A remote control method for an electrolytic aluminum crane, characterized in that, Including the steps: Before the operation of the electrolytic aluminum crane, perform initial position detection through sensors installed on each actuator of the electrolytic aluminum crane, establish a coordinate system corresponding to each actuator and the electrolytic cell positions, and transmit the detection data to the remote control terminal in real time; According to the operation requirements, input operation instructions on the terminal interface. The remote control terminal generates corresponding control instructions based on the input instructions and the received detection data; Send the control instructions to the control system of the electrolytic aluminum crane. According to the control instructions, drive each actuator of the electrolytic aluminum crane to perform corresponding actions to complete the production operation.

2. The remote control method for an electrolytic aluminum crane according to claim 1, characterized in that Before the operation of the electrolytic aluminum crane, obtain the absolute position information of each actuator through sensors installed on each actuator of the electrolytic aluminum crane.

3. The remote control method of the electrolytic aluminum crane according to claim 2, characterized in that The specific process of establishing the coordinate system corresponding to each actuator and the electrolytic cell positions is as follows: Install an absolute value encoder on the driven wheel shaft of the trolley, and evenly and spacedly set several RFID reset tags beside the workshop track. Obtain the absolute position information of the trolley through the absolute value encoder and the RFID reset tags; Install a laser ranging device at the running position of the tapping carriage, and obtain the absolute position information of the tapping carriage through the laser ranging device; Install corresponding absolute position measurement units at the cylinder positions of the crust breaking mechanism, anode mechanism, slag skimming mechanism, feeding mechanism, tapping lifting mechanism, and frame lifting mechanism respectively to obtain the absolute position information of each mechanism; According to the obtained absolute position information of each actuator, combined with the actual layout and dimension information of the electrolytic cell, establish a coordinate system corresponding to each actuator and the electrolytic cell positions through mathematical calculation and spatial solution.

4. The remote control method for an electrolytic aluminum crane according to claim 1, characterized in that, The specific process of the remote control terminal generating corresponding control instructions based on the input instructions and the received detection data is as follows: Deeply analyze the input instructions, identify the operation type, extract the specific operation parameters included in the input instructions, and preprocess the received detection data; According to the identified operation type, call the corresponding preset model from the algorithm model library built in the remote control terminal, input the extracted specific operation parameters into the corresponding preset model, and perform iterative calculation and optimization solution through the preset model to obtain the target operation parameters; Based on the target operation parameters, generate control instructions for driving each actuator of the electrolytic aluminum crane.

5. The remote control method for an electrolytic aluminum crane according to claim 4, wherein The specific process of preprocessing the received detection data is as follows: Detect missing values in the detection data. If there are missing values, use the data interpolation method to fill in the missing values according to the change trend of adjacent valid data; Detect outliers. Based on the 3σ principle of statistics, identify the data outside the range of the mean ± 3 times the standard deviation as outliers, and correct or eliminate the outliers by comparing with historical normal data and combining on-site actual working conditions; Normalize the data to map the detection data with different dimensions to a specified interval.

6. The remote control method for the electrolytic aluminum crane according to claim 4, characterized in that The specific process of inputting the extracted specific operation parameters into the corresponding preset model and performing iterative calculation and optimization solution through the preset model to obtain the target operation parameters is as follows: Input the extracted specific operation parameters as initial values into a preset model, and the preset model constructs an objective function based on the dynamic equation, kinematic constraints, and operation process requirements; During the iterative calculation process, the preset model calculates the performance index under the current parameters according to the objective function, and uses an intelligent optimization algorithm to adjust the operation parameters; After each iterative calculation, determine whether the preset convergence condition is satisfied. If it is satisfied, output the current operation parameters as the target operation parameters; if not, continue the next round of iterative calculation until the convergence condition is satisfied.

7. The remote control method for an electrolytic aluminum crane according to claim 6, wherein The construction process of the preset model is as follows: According to the dynamic characteristics, kinematic laws, and specific requirements of the operation process of each actuator of the aluminum electrolysis crane, construct a corresponding initial mathematical model; Obtain the historical operation data and simulation experiment data of each actuator under different operation conditions, and screen out the characteristic parameters closely related to the operation state of each actuator to obtain a set of characteristic parameters; Use the set of characteristic parameters as input variables and train them using machine learning algorithms. By iteratively adjusting the weights and biases of the parameters in the initial mathematical model, fit the non-linear relationship between the input variables and the operation parameters of the actuator, determine the parameters in the initial mathematical model, and obtain the preset model.

8. The remote control method for an electrolytic aluminum crane according to claim 7, characterized in that, During the construction process of the preset model, incorporate the interference factors in the aluminum electrolysis production environment and the performance changes brought about by long-term operation into the preset model in the form of correction coefficients or additional constraint conditions, and optimize the preset model in combination with the physical characteristic parameters of each actuator.

9. The remote control method of the electrolytic aluminum crane according to claim 1, characterized in that, Feed back the operation states and operation execution data of each actuator during the execution process to the remote control terminal; Compare the actual operation states and operation execution data of each actuator fed back with the theoretical data corresponding to the control instructions; If the actual positioning accuracy or actual rotation accuracy of the key parts of the crane does not meet the preset accuracy requirements, generate an error correction instruction, send the error correction instruction to the control system of the aluminum electrolysis crane, and finely adjust the actuator according to the error correction instruction.

10. A remote control system for an electrolytic aluminum crane, which is used to implement the remote control method of an electrolytic aluminum crane described in any one of claims 1-9, characterized in that, Including: A detection module, which is arranged on each actuator of the aluminum electrolysis crane and is used to perform initial position detection before the aluminum electrolysis crane operates to obtain the absolute position information of each actuator; A remote control terminal, which is used to receive the detection data transmitted by the detection module, receive the operation instructions input by the operator, generate corresponding control instructions through a preset algorithm based on the input instructions and the received detection data, and receive the feedback of the operation states and operation execution data of each actuator during the execution process, compare the feedback data with the theoretical data corresponding to the control instructions, and generate an error correction instruction if the actual positioning accuracy or actual rotation accuracy of the key parts of the crane does not meet the preset accuracy requirements; A communication module, which is used to realize data transmission between the detection module and the remote control terminal, and between the remote control terminal and the control system of the aluminum electrolysis crane; The electrolytic aluminum crane control system is used to receive control instructions and error correction instructions sent by a remote control terminal, drive each actuator of the electrolytic aluminum crane to perform corresponding actions according to the instructions, complete production operations, and feedback the operating status of each actuator and operation execution data to the remote control terminal.

Images