Optimization and Upgrading of Key Technologies for Lead Electrolysis Equipment Sets

In response to the current key issues in the production of existing lead electrolysis equipment, such as poor verticality of lead cathode, inferior welding quality, short service life of conductive rod and high contact resistance, we have developed several technologies based on the principles of elastoplastic mechanics, bending and shaping, resistance spot welding and neural network control.These technologies include the stiffness-enhanced plate surface embossing, the dynamic multi-eigenvalue fusion of BP neural network resistance welding control, and the efficient rod-plate separation for cathode lead plates and conductive rod. They have been applied in a newly constructed 480kt/a project at a large domestic lead smelting enterprise. The results have demonstrated significant improvements in overall equipment production capacity, reducing the production time from 15s/plate to 11.5s/plate, decreasing the short-circuit rate caused by cathode verticality to 0.43%, and extending the service life of conductive rod by 12.8%.


Introduction
Lead is an essential foundational metal in the industry and is generally purified through electrolytic refining.There are two production processes for lead electrolysis: small plate and large plate.Due to the fact that the comprehensive energy consumption of lead electrolytic refining with large plates is only 70 kgce/t, which is 50 kgce/t less than the 120 kgce/t of small plate electrolysis [1] , in the past decade, most newly established lead electrolytic refining projects have adopted the large plate process along with corresponding equipment.Since the introduction of complete sets of large plate production equipment from Japan by a Yunnan-based company in China in 2005 [2] , domestic enterprises have been continuously assimilating, absorbing and improving based on production experience.They have made remarkable contributions to the localization of equipment, essentially meeting the equipment requirements of the electrolytic process.The equipment is with high automation, low investment, and convenient maintenance.
With the continuous expansion of single lead electrolysis workshop production capacity, ranging from 100kt/a to gradually reach 150kt/a (even 180kt/a), the demands for performance and stability of a single set of lead electrolysis equipment have become increasingly stringent.Currently, domestically produced equipment struggles to meet the production requirements.Simultaneously, in order to reduce energy consumption during electrolysis and decrease maintenance and operation costs of the equipment, smelting enterprises have diverse and complex demands for equipment performance.This paper focuses on the

Verticality of starting sheet
The verticality of lead starting sheet is a crucial parameter for the product and also one of the significant reasons for potential short-circuit during later-stage electrolysis.In general, the verticality of lead starting sheet before it is placed into the tank needs to be controlled within 10mm to 15mm. Figure 1 illustrates the situation of improper verticality between anode and cathode.Currently, key factors affect the verticality of lead starting sheet during its manufacturing process include: 1) Unreasonable design of the embossed pattern on the anode, inadequate embossing depth, and poor board rigidity.
2) After embossing and leveling the anode, secondary deformation occurs before it enters the electrolytic cell.

Figure 1. Unqualified verticality of distance between cathode and anode
Through production practice, while ensuring that the welding seams of the embossing machine frame do not detach and the design of the embossing die pattern is reasonable, the embossing depth can be maintained stably.However, the main factors affect the deterioration of the verticality are primarily concentrated on factor 2. After completion of lead starting sheet , it needs to be alternately arranged with the anode, facilitating the lifting of both anodes and cathodes into the electrolytic cell.Currently, the technical solutions for spacing between anodes and cathodes involve inserting the cathode from the top down into the space between the anodes.The thickness of starting sheet of cathode itself is about 0.9mm, with a thin profile and large dimensions, making it prone to deformation.Before descending, the motion of starting sheet is with vibration, combined with the effects of air resistance or guide baffle, leading to potential deformation.In severe cases, during the descent, the bottom of starting sheet may collide with the anode plate, resulting in bending and damage to the cathode and necessitating equipment shutdown.Figure 2 and figure 3 show the on-site use and spatial situation of the downward insertion spacing scheme.

Connection Quality of Lug of Starting Sheet
During the processing of starting sheet, it's necessary to wrap the conductive rod with lead foil and assemble it, followed by securing it to the folded side.The originally introduced technique involved uniformly pressing three circular holes in the length direction of conductive rod on the first floor of the workshop, and then using an inclined lifting mechanism to elevate the starting sheet to the second floor.At a welding station, 14 to 16 points are welded in the length direction of the folded area of the lead foil, enhancing the starting sheet conductivity and verticality strength [3] . Figure 4 depicts the locations of the pressed hole and the spot welding on the cathode plate.The quality of the connection at starting sheet lug is closely tied to the quality of pressed holes and spot welding.During hole pressing, if the punch-matrix or assembly dimensions are poor, there is a tendency for loosening or opening of the connection.During the inclined lifting process, the lead foil might strip or shift, as shown in Figure 5.In the case of spot welding, if there is incomplete welding, during the transport or electrolysis after spacing the anode and cathode, the lead foil may strip and slide off the conductive rod, resulting in scrap pieces.

Lifespan of Conductive Rod
The conductive rod functions as conducting electricity and bearing the cathode lead plate during the electrolysis process.It is supplied in the cathode machine and withdrawn in the cathode lead cleaning machine.The conductive rod is circulated internally between the two machines in the workshop and the electrolytic cell.Made from copper-clad steel with layer thickness of less than 3mm, the conductive rod carries a high replacement cost.Its service life is primarily connected to damages, which predominantly occur during the rod withdrawal and cleaning processes.
Currently, there are two methods to withdraw the conductive rod, both aimed at removing it from the cathode lead .One method is manually pulling the conductive rod from one side of the deposited lead and then collecting it in a storage frame.The other adopts a mechanical arm to grip the conductive rod from the side of the deposited lead, and during this process, due to the dry friction between the lead foil and the rod, the copper layer on the surface of the rod gets damaged.Meanwhile, the scraped-off copper results in secondary contamination of the lead product.
There are currently two cleaning methods in use: one is used with introduced equipment, employing steel brushes to polish and remove impurities and oxide layers from the conductive rod.This method causes significant wear on the conductive rod and generates a considerable amount of dust, leading to poor working conditions.The other method involves using an acid cleaning tank equipped with an ultrasonic generator, where high-frequency ultrasonic vibrations are employed to remove oxides during the cleaning process.This solution has low efficiency, high operating costs, and the acid fumes negatively impact the workshop environment.

Enhanced Rigidity Plate Embossing Technology
In order to enhance the flatness and rigidity of the starting sheet surface, optimization of the embossing technology is required.In this project, measures have been taken including: 1) Embossing Die Design: Based on the trace direction of subsequent transfer of lead starting sheet, the embossing die pattern has been rationally optimized.This results in the formation of a regular embossed pattern on the surface of the lead starting sheet, ensuring even pressure distribution during the embossing process to increases rigidity while preventing excessive stress concentration.The traditional pattern from the imported technology is shown in Figure 6, while the pattern used in this project is as depicted in Figure 7.This adapted pattern can accommodate the deformation effects caused by rapid lateral transfer of starting sheet by the robot.2) Embossing Parameter Control: the control of parameters during the embossing process, such as pressure, speed and time, is essential to achieve the optimal rigidity-enhancing effect.In this project, the oil cylinder pressure used is within the range of 170 to 182 kN, the embossing speed ranges from 20 to 30mm/s, and the time varies from 3.0 to 3.5s.
3) Die Material Selection: materials that possess good rigidity and wear resistance for the embossing die.Implement suitable heat treatment or surface treatment methods to enhance their mechanical properties.Additionally, ensure that the flatness of the embossing die surface is controlled within 2mm, which contributes to the uniform and consistent alignment of embossed patterns.This reduces the risks of bending and deformation after embossing.

Neural Network-Controlled Resistance Welding Technology
Resistance welding technology is widely applied, particularly suited for connecting metals with low melting points and poor thermal conductivity.It has been widely adopted in the production of lead staring sheet.Due to the thin thickness of lead foil, multiple spot welding is necessary to ensure loadbearing capacity.Typically, 14 to 16 spots are used in the industry.To simplify the welding equipment, each set of welding equipment is shared by every 7 to 8 welding heads, two sets in total.Several parameters, such as welding current, welding time, electrode spacing and pressure influence the quality of resistance welding during the process.In order to enhance the uniformity of connection quality across multiple welding points, control technology for resistance welding is crucial.In this project, a BP neural network-controlled resistance welding technology with dynamic multifeature fusion has been adopted.It makes use of multiple characteristic values in combination with the back propagation (BP) neural network to control the resistance welding process.The topology structure of this technology is depicted in Figure 8.The BP neural network establishes a model between the characteristic values and welding quality by learning from historical data and utilizing the back propagation algorithm.The transformation from the input layer to the hidden layer is nonlinear, while from the hidden layer to the output layer is linear.In this project, the network structure and parameters are obtained by extracting a subset of experimental data for network training, and the remaining data is used as samples for network prediction [5] .The network structure used consists of an input layer with 5 nodes, a hidden layer with 15 nodes and an output layer with 1 node.The iteration count is set at 120, the learning rate is 0.1.The sigmoid function is used as the activation function.The performance of the neural network is evaluated using the root mean square error of shear strength for welding points [6] , as follows:

Robot Material Transfer Technology by Using S-Curve Algorithm
After the initial production of starting sheet, it is flipped to a vertical position using a flipper machine.The sheet then undergo embossing and straightening processes before being transferred to a spacing machine for arrangement.To minimize space occupation and equipment investment, an industrial robot is used in this project to transfer the starting sheet between the flipper , embossing, straightening and spacing machines.Managing the overall transfer time and preventing deformation of starting sheet during the process is critical.
In this project, the S-curve algorithm is adopted to plan the robot's motion trajectory.The S-curve exhibits smooth and continuous characteristics.By considering the physical properties of the robot and materials, adjusting acceleration and deceleration at turning points, optimizing the trajectory path and controlling the radius of curvature, the robot can achieve smooth acceleration and deceleration during material transfer process.It avoids sudden acceleration changes that could impact or damage the material, thereby enhancing transfer stability and precision.Ultimately, it achieves optimal material transfer quality and efficiency.Figure 8 illustrates the optimized path during robot material transfer.

Figure 9. Optimized path for material transfer by robot
In the project implementation, the robot follows steps ① to ⑦ after retrieving the plates from the flipper machine.This sequence facilitates the movement of the cathode between the flipper, straightening and spacing workstations.The S-curve algorithm is applied at each turning point, ensuring smooth material transfer.It has achieved a favorable outcome with a single cycle transfer time controlled within 11.5s.

High-Efficiency Rod-Plate Separation Technology
In order to reduce the wear on the conductive rod caused by traditional withdrawal, a high-efficiency rod-to-plate separation technology is adopted.Before the separation, the conductive rod and lead plate are securely fastened.A cutting tool is then used to start cutting from the V-shaped notch on one side of the cathode lead plate, moving towards the other side.After separation, the lead plate slides into a shredder from below.Figure 10 illustrates the mechanism of rod-to-plate separation.To achieve efficient and reliable rod-plate separation, the following measures have been implemented in the project: 1) Optimization of Separation Mechanism Design: this includes the design of the cutting tool holder, positioning device and auxiliary force control device.The shape, size and material of the cutting tool, especially the optimization of its design, have been addressed to enhance separation efficiency and life service.Various cutting tool shapes like double-edged, internal and external figure-eight, sickle-like tools have been designed.Materials with sufficient hardness and wear resistance have been chosen to ensure efficient cutting of the lead foil.
2) Detection and Feedback Control: by using sensors and feedback control systems, the force, position and effectiveness during the separation process are monitored in real time.Adjustments and optimizations are made based on the feedback information to ensure an efficient separation operation.
3) Cutting Speed Control: the speed of cutting tool is controlled.Excessive cutting speed might lead to wear and damage of the tool, while a too low speed could decrease separation efficiency and impact the cutting process.Generally, the cutting speed is controlled within 1 to 1.2 m/s.

Efficiency Improvement of Machine
By implementing the enhanced rigidity plate embossing technology, the robot material transfer technology by S-curve algorithm and the high-efficiency rod-plate separation technology, the production yield of starting sheet has been improved.The quality of cathode-anode spacing has also been enhanced, and the efficiency of material transfer and separation has been ensured.As a result, the overall operational efficiency of workshop machine has been elevated to 11.5s/sheet, and operation rate of machine has reached over 96% (excluding the impact of manual bending).Figure 11 shows the cathode-anode spacing status.

Electrolytic Energy Consumption Reduction
By adopting the enhanced rigidity plate embossing technology and the robot material transfer technology by S-curve algorithm, the surface tolerance of the starting sheet has been effectively controlled.
Additionally, the material transfer process has been executed smoothly and steadily, while the subsequent deformation of the material has been maintained within an acceptable range.The verticality of starting sheet after spacing is depicted in Figure 12, showing that the verticality is less than 8mm.
Based on the BP neural network-controlled resistance welding technology, its strong nonlinear fitting ability comprehensively considers the mutual influence among multiple characteristic values.This enhances the precision and stability of the welding process control.The technology ensures a more uniform distribution of strength and resistance values between multiple welding points, leading to a reduction in the contact resistance of the starting sheet itself.
The combination and optimization of these measures effectively reinforce the surface rigidity, improve the starting sheet verticality, and reduce the short-circuit rate during electrolysis.The average short-circuit rate has decreased from 1.12% (in the direct smelting plant electrolytic workshop) to 0.43%.Additionally, the technology lowers the material's intrinsic contact resistance.This multifaceted improvement in the lead electrolysis workshop enhances energy efficiency, resulting in a 15kWh reduction in lead electrolytic energy consumption per ton.The short-circuit rate data over a three-month period is presented in Table 1.

Extension of Conductive Rod Life Service
The high-efficiency rod-plate separation technology enhances the efficiency, accuracy and stability of separation, safeguarding the conductive rod from dry wear caused by the lead foil during the withdrawal process, which effectively protects the copper layer.Additionally, during the cleaning of the conductive rod, high-pressure water cleaning technology can be used.By using high-pressure cleaning liquid without abrasive particles, the surface of conductive rod is impacted from a specific angle with a certain pressure and flow rate.This process removes the adhered oxides and impurities from the surface, avoiding damage to the copper layer matrix caused by other physical cleaning methods.Consequently, the life service of the conductive rod is further prolonged.After experimental verification, when compared to the conventional steel brush cleaning, the copper layer's thickness change is slow, resulting in a 12.8% extension in the life service of the conductive rod copper layer.

Conclusion
Addressing various critical challenges in traditional lead electrolytic equipment and recognizing the increasing demands on equipment processing capacity, stability and user-friendliness with capacity enhancements, this study has delved into technologies such as enhanced rigidity plate embossing, dynamic multi-feature BP neural network resistance welding control, high-efficiency rod-plate separation for cathode lead plate and conductive rod.These technologies are then implemented in practical production equipment, confirming their feasibility and reliability.The practical application of these technologies resulted in an overall production capacity of 11.5s/plate, a reduced electrolytic short-circuit rate of 0.43%, and an extended conductive rod life service of 12.8%.These positive outcomes have brought considerable economic benefits to enterprises and played a proactive role in advancing the upgrade of equipment in the lead electrolysis industry.
For a certain large domestic smelting enterprise constrained by project investment limits, manual cleaning for conductive rod is temporarily adopted.In the future, considering the physical and chemical properties of metallic copper and its surface oxide layer, options such as high-pressure water cleaning or laser cleaning could be tried.These approaches aim to achieve effective cleaning while ensuring the copper matrix remains undamaged, which could contribute positively to enhancing the life service and electrolytic efficiency of the conductive rod.It is anticipated that practical implementation will be realized in future production.

Figure 4 .Figure 5 .
Figure 4. Pressed hole and spot welding on the cathode plate

Figure 6 .
Figure 6.Cathode pattern in the traditional imported technology

Figure 8 .
Figure 8. Topology structure of BP neural network

Figure 11 .
Figure 11.Cathode-anode spacing status Figure 12.Verticality condition of starting sheet The 10th International Conference onLead and Zinc Processing (Lead-Zinc 2023)research of key issues in traditional equipment such as poor verticality of lead starting sheet, low welding quality, short service life of conductive rod and large contact resistance.Corresponding solutions and technologies have been proposed and developed through this research.The equipment has been applied in real projects, yielding certain positive outcomes.

Table 1 .
Short-circuit rate statistics in the electrolytic workshop.
[1] QIAN Q and ZHANG Y P 2008 Introduction to the New Process of Lead Electrolytic Refining Large Plate Equipment Proceedings of the 10th National Symposium on Lead & Zinc Metallurgy Technology and Production Equipment p 252-255.[2] LI T and ZHANG J L 2015 Technology and Application of Large Plate Lead Electrolysis Process World Nonferrous Metals.02 p 31-32.[3] CHEN Z H and ZHANG G S 2007 cheed Lead Smelting Process and Equipment Engineering Design and Research 122 p 7-8. [4] LU X K and GAO Z L 2021 Intelligent Control of Electric Resistance Welding Based on BP Neural Network and PID Self-adaptive Tuning Hot Machining Process 50 p 141-142.[5] LIN W Y, REN X Y and ZHOU T T 2018 A novel robust algorithm for position and orientation detection based on cascaded deep neural network Neurocomputing 308 p 138 − 146.[6] LUO L X and XING Y F 2019 CMT Spot Welding Deformation of Sheet Metal Based on BP Neural Network and Genetic Algorithm Transactions of the China Welding Institution 40 p 80-81.