Laser cladding welding parameter optimization control method based on particle swarm optimization

Abstract

The invention discloses a laser cladding welding parameter optimization control method based on particle swarm optimization, which relates to the technical field of laser cladding welding control, and comprises the following steps: establishing a molten pool kinetic model and a heat conduction model according to the physical phenomenon of laser cladding welding; determining optimization target parameters and constraint condition parameters; the optimization target parameters and the constraint condition parameters are coded into particle positions, a group of particles are randomly initialized in a parameter space, and each particle corresponds to a group of welding parameters; evaluating the fitness of each particle using molten pool kinetics and a thermal conduction model; and updating an individual optimal solution and a global optimal solution according to the fitness of each particle. According to the method, intelligent selection and control of the welding parameters are achieved, the shape and size of the molten pool can be more accurately controlled by optimizing the welding parameters, and therefore welding defects are reduced, the optimized parameters can effectively reduce the width of a heat affected zone, the adverse effect of the heat affected zone on the performance of base metal, and the original performance of the base metal is kept.

Description

Technical Field

[0001] The invention relates to the technical field of laser cladding welding control, and in particular to a laser cladding welding parameter optimization control method based on particle swarm optimization. Background Art

[0002] Laser cladding welding control technology is an advanced material processing technology, which is widely used in aerospace, automobile manufacturing, mold repair and other fields. Laser cladding welding has high energy density and can achieve precision welding. The heat affected zone of laser cladding welding is small, which is conducive to reducing welding deformation. During the laser cladding welding process, parameters such as laser parameters and welding speed are easy to control. It is suitable for welding a variety of metal and non-metal materials.

[0003] However, due to the large number of parameters involved in laser cladding welding, parameter fluctuations have a greater impact on welding quality. Traditional optimization methods rely on experience and are difficult to achieve global optimization. The quality control of the cladding layer is difficult: the thickness, composition and properties of the cladding layer are difficult to accurately control. Summary of the invention

[0004] In order to solve the above-mentioned technical problems of laser cladding welding control, the present invention provides a laser cladding welding parameter optimization control method based on particle swarm optimization. The following technical solutions are adopted: The laser cladding welding parameter optimization control method based on particle swarm optimization includes the following steps: Step 1, establishing a molten pool dynamics model and a heat conduction model based on the physical phenomena of laser cladding welding; Step 2, determine the optimization target parameters and constraint condition parameters; Step 3, encode the optimization target parameters and constraint condition parameters into particle positions, and randomly initialize a group of particles in the parameter space, where each particle corresponds to a set of welding parameters; Step 4, using the melt pool dynamics and heat conduction model to evaluate the fitness of each particle; Step 5: Update the individual optimal solution and the global optimal solution according to the fitness of each particle; Step 6, adjusting the position and speed of the particles according to the speed update formula and the position update formula of the particle swarm optimization algorithm, and obtaining corresponding welding parameters based on the updated position and speed of the particles; Step 7, converting the optimized welding parameters into welding control instructions, and exchanging the welding control instructions with the laser cladding welding controller.

[0005] By adopting the above technical solutions, the combination of particle swarm optimization algorithm realizes the intelligent selection and control of welding parameters, laying the foundation for the automation and intelligence of the welding process. By optimizing the welding parameters, the shape and size of the molten pool can be controlled more accurately, thereby reducing welding defects such as cracks, pores and lack of fusion, and improving the mechanical properties and appearance quality of the weld. The optimized parameters can effectively reduce the width of the heat-affected zone, reduce the adverse effects of the heat-affected zone on the properties of the parent material, and maintain the original properties of the parent material. The automated parameter optimization process reduces the time for manual parameter adjustment, improves the stability and repeatability of the welding process, and thus improves production efficiency.

[0006] Production costs are reduced by reducing welding defects and rework rates, while more efficient use of heat also reduces energy consumption.

[0007] It can quickly adapt to different welding tasks and materials, provide optimized parameter settings for a variety of welding applications, and enhance the flexibility of the welding process.

[0008] By precisely controlling welding parameters, material waste can be minimized and resource utilization optimized.

[0009] Automated parameter optimization reduces the possibility of human error and improves the safety of the welding process.

[0010] The laser cladding welding parameter optimization control method based on particle swarm optimization not only improves the welding quality and efficiency, but also reduces the cost and environmental impact, providing strong technical support for the advancement and development of welding technology.

[0011] Optionally, in step 5, the molten pool dynamics and heat conduction models are run using the updated welding parameters. If the optimization target is not achieved, return to step 1 to continue iterating.

[0012] By adopting the above technical solution, the molten pool dynamics and heat conduction models can be run with updated welding parameters in the iterative process, which can more accurately evaluate the impact of parameter changes on welding results, thereby further improving the accuracy of optimization results. Iteration until the optimization goal is reached ensures that the final welding parameters can meet the predetermined quality standards and improve the reliability of the process.

[0013] Optionally, the optimization target parameters in step 2 are uniformity of the cladding layer and minimization of thermal stress, and the constraint condition parameters are the laser power range, welding speed range, and powder feeding rate range.

[0014] By adopting the above technical solution and optimizing the uniformity of the cladding layer, it is possible to ensure that the macro and micro structures of the weld are more uniform, thereby improving the mechanical properties and service life of the weld.

[0015] Minimizing thermal stress: Minimizing thermal stress helps reduce cracks and deformation during welding, thereby improving the integrity and stability of the welded structure.

[0016] Laser power range constraints: Limiting the laser power range helps prevent overheating or insufficient heat input, ensuring that the welding process is carried out at an appropriate power and improving welding quality.

[0017] Welding speed range constraints: Constraining the welding speed can ensure that the molten pool has enough time to form and cool, while avoiding incomplete fusion caused by too fast a speed or overheating caused by too slow a speed.

[0018] Powder feed rate range constraints: Optimizing the powder feed rate can ensure that the cladding material is deposited at an appropriate rate to obtain uniform cladding layer thickness and composition.

[0019] Optionally, in step 3, a set of welding parameters includes laser power P, welding speed V and powder feeding rate d.

[0020] Optionally, in step 3, the method of randomly initializing a group of particles in the parameter space is: Step 31, determining the number of particles N in the particle group; Step 32, for each parameter laser power P, welding speed V and powder feeding rate d, a value is randomly generated within the set range using the following formula: ; ; ; in , and is A random number uniformly distributed in the range; Step 33, repeat step 32 until all N particles are initialized.

[0021] By adopting the above technical solutions, the uniform distribution of particles in the parameter space is ensured, which helps the algorithm to explore diverse solutions in the early stages. By adjusting the value of N, the complexity and computational cost of the algorithm can be flexibly controlled to adapt to different application requirements. Repeating the random generation process until all particles are initialized ensures that the algorithm can cover multiple areas of the parameter space at the beginning of the search, laying a good foundation for subsequent iterative optimization. This random initialization method provides a good starting point for the particle swarm optimization algorithm, which helps to improve the effect and efficiency of laser cladding welding parameter optimization.

[0022] Optionally, the formula for evaluating the fitness of each particle in step 4 is: ; in is the fitness value of the particle, is the desired melt pool diameter, is the calculated melt pool diameter, is the desired heat affected zone width, is the calculated heat affected zone width, and is the weight coefficient, which is used to adjust the importance of the molten pool diameter and the heat affected zone width in the fitness evaluation; The calculation formula for the molten pool diameter is: ; in is the calculated value of the melt pool diameter, is the laser power, k is the thermal conductivity of the welding object material, is the thermal enthalpy of the welding object material; The calculation formula for the width of the heat affected zone is: ; in is the calculated value of the width of the heat affected zone, is the heat input per unit length.

[0023] By adopting the above technical solutions, an effective tool is provided for the optimization of laser cladding welding parameters through accurate fitness evaluation, which helps to achieve high-quality welding results while improving production efficiency and reducing costs.

[0024] Optionally, step 5 includes the following sub-steps: Step 51, before the particle swarm optimization algorithm starts, the initial position of each particle is set to the individual optimal solution; Step 52, based on the fitness value of the particle in step 4, if the fitness value of the current particle is higher than the fitness value of the corresponding individual optimal solution, then update the current particle to the individual optimal solution, otherwise keep the individual optimal solution unchanged; Step 53: After the individual optimal solutions of all particles are determined, the individual optimal solution with the maximum fitness value is selected as the global optimal solution.

[0025] By adopting the above technical solutions, each particle is ensured to have a reference point at the beginning of optimization, which helps the algorithm converge quickly. Dynamically updating the individual optimal solution can ensure that the particle evolves towards a better solution by comparing the fitness value of the current particle with the fitness value of the individual optimal solution, thereby improving the algorithm's search efficiency. The individual optimal solution with the maximum fitness value is selected as the global optimal solution, which provides a reference for the optimal solution for the entire particle swarm and helps guide other particles to search the solution space more effectively.

[0026] Optionally, in step 6, the particle velocity update formula of the particle swarm optimization algorithm is as follows: ; in yes At time , the velocity of the i-th particle in the j-th dimension; is the velocity of the ith particle in the jth dimension at time t; w is the inertia weight; The weight of particles learning from the individual optimal solution; is the weight of the particle learning towards the global optimal solution; is a random number in the interval [0,1], used to simulate random behavior of particles; is a random number in the interval [0,1], used to simulate random behavior of particles; is the individual optimal position of the ith particle in the jth dimension; is the position of the i-th particle in the j-th dimension at time t; is the value of the global optimal position in the jth dimension; The particle position update formula of the particle swarm optimization algorithm is: ; in It's time The position of the i-th particle at time ; It's time The position of the i-th particle at time ; It's time is the time of the ith particle in the jth dimension.

[0027] By adopting the above technical solution and introducing inertia weight, individual learning weight and global learning weight, particles can balance the ability to explore (search new areas) and utilize (utilize known optimal positions), thereby improving the global and local search capabilities of the algorithm.

[0028] The position update formula ensures that particles move efficiently in the solution space while staying within the defined search range.

[0029] Optionally, a method for obtaining corresponding welding parameters based on the updated particle position and velocity is: Suppose there are three welding parameters: laser power P, welding speed V and powder feeding rate d, the particle position vector Each element in is mapped to the actual range of these three welding parameters. The following is the mapping formula: ; ; ; The mapping function map is defined as: ; in is the position value of the particle, and are the minimum and maximum values ​​of the particle position vector, and are the minimum and maximum values ​​of the corresponding welding parameters.

[0030] By adopting the above technical solution, the position vector of the particle is mapped to the actual welding parameter range through the mapping formula, ensuring that the parameters generated by the particle swarm optimization algorithm are operational and meet the actual process requirements.

[0031] Improve parameter accuracy. The mapping function map can convert the continuous position values ​​of particles into the actual values ​​of welding parameters, which improves the accuracy of parameter optimization and the value of practical application.

[0032] To enhance algorithm flexibility, the mapping method allows the algorithm to work within different welding parameter ranges, making the optimization process adaptable to different materials and welding conditions.

[0033] In summary, the present invention includes at least the following beneficial technical effects: The present invention can provide a laser cladding welding parameter optimization control method based on particle swarm optimization. The combination of particle swarm optimization algorithm realizes the intelligent selection and control of welding parameters, laying a foundation for the automation and intelligence of the welding process. By optimizing the welding parameters, the shape and size of the molten pool can be more accurately controlled, thereby reducing welding defects. The optimized parameters can effectively reduce the width of the heat-affected zone, reduce the adverse effects of the heat-affected zone on the performance of the parent material, and maintain the original performance of the parent material. The automated parameter optimization process reduces the time for manual parameter adjustment, improves the stability and repeatability of the welding process, and thus improves production efficiency.

[0034] The laser cladding welding parameter optimization control method based on particle swarm optimization not only improves the welding quality and efficiency, but also reduces the cost and environmental impact, providing strong technical support for the advancement and development of welding technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of the laser cladding welding parameter optimization control method based on particle swarm optimization of the present invention. DETAILED DESCRIPTION

[0036] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0037] The embodiment of the present invention discloses a laser cladding welding parameter optimization control method based on particle swarm optimization.

[0038] Reference Figure 1 , Embodiment 1, a laser cladding welding parameter optimization control method based on particle swarm optimization, comprising the following steps: Step 1, establishing a molten pool dynamics model and a heat conduction model based on the physical phenomena of laser cladding welding; Step 2, determine the optimization target parameters and constraint condition parameters; Step 3, encode the optimization target parameters and constraint condition parameters into particle positions, and randomly initialize a group of particles in the parameter space, where each particle corresponds to a set of welding parameters; Step 4, using the melt pool dynamics and heat conduction model to evaluate the fitness of each particle; Step 5: Update the individual optimal solution and the global optimal solution according to the fitness of each particle; Step 6, adjusting the position and speed of the particles according to the speed update formula and the position update formula of the particle swarm optimization algorithm, and obtaining corresponding welding parameters based on the updated position and speed of the particles; Step 7, converting the optimized welding parameters into welding control instructions, and exchanging the welding control instructions with the laser cladding welding controller.

[0039] The combination of particle swarm optimization algorithm realizes the intelligent selection and control of welding parameters, laying the foundation for the automation and intelligence of welding process. By optimizing welding parameters, the shape and size of molten pool can be controlled more accurately, thereby reducing welding defects such as cracks, pores and lack of fusion, and improving the mechanical properties and appearance quality of welds. The optimized parameters can effectively reduce the width of heat-affected zone, reduce the adverse effect of heat-affected zone on the properties of parent material, and maintain the original properties of parent material. The automated parameter optimization process reduces the time of manual parameter adjustment, improves the stability and repeatability of welding process, and thus improves production efficiency.

[0040] Production costs are reduced by reducing welding defects and rework rates, while more efficient use of heat also reduces energy consumption.

[0041] It can quickly adapt to different welding tasks and materials, provide optimized parameter settings for a variety of welding applications, and enhance the flexibility of the welding process.

[0042] By precisely controlling welding parameters, material waste can be minimized and resource utilization optimized.

[0043] Automated parameter optimization reduces the possibility of human error and improves the safety of the welding process.

[0044] The laser cladding welding parameter optimization control method based on particle swarm optimization not only improves the welding quality and efficiency, but also reduces the cost and environmental impact, providing strong technical support for the advancement and development of welding technology.

[0045] In Example 2, in step 5, the molten pool dynamics and heat conduction models are run using the updated welding parameters. If the optimization target is not achieved, return to step 1 to continue iterating.

[0046] By continuously running the molten pool dynamics and heat conduction models with updated welding parameters during the iteration process, the impact of parameter changes on the welding results can be more accurately evaluated, thereby further improving the accuracy of the optimization results. The iteration cycle is repeated until the optimization goal is reached, ensuring that the final welding parameters can meet the predetermined quality standards and improve the reliability of the process.

[0047] In Example 3, the optimization target parameters in step 2 are the uniformity of the cladding layer and the minimization of thermal stress, and the constraint condition parameters are the laser power range, the welding speed range and the powder feeding rate range.

[0048] By optimizing the uniformity of the cladding layer, it is possible to ensure a more uniform macro- and micro-structure of the weld, improving the mechanical properties and service life of the weld.

[0049] Minimizing thermal stress: Minimizing thermal stress helps reduce cracks and deformation during welding, thereby improving the integrity and stability of the welded structure.

[0050] Laser power range constraints: Limiting the laser power range helps prevent overheating or insufficient heat input, ensuring that the welding process is carried out at an appropriate power and improving welding quality.

[0051] Welding speed range constraints: Constraining the welding speed can ensure that the molten pool has enough time to form and cool, while avoiding incomplete fusion caused by too fast a speed or overheating caused by too slow a speed.

[0052] Powder feed rate range constraints: Optimizing the powder feed rate can ensure that the cladding material is deposited at an appropriate rate to obtain uniform cladding layer thickness and composition.

[0053] In Example 4, in step 3, a set of welding parameters includes laser power P, welding speed V and powder feeding rate d.

[0054] In Example 5, in step 3, the method of randomly initializing a group of particles in the parameter space is: Step 31, determining the number of particles N in the particle group; Step 32, for each parameter laser power P, welding speed V and powder feeding rate d, a value is randomly generated within the set range using the following formula: ; ; ; in , and is A random number uniformly distributed in the range; Step 33, repeat step 32 until all N particles are initialized.

[0055] It ensures the uniform distribution of particles in the parameter space, which helps the algorithm to explore diverse solutions in the early stages. By adjusting the value of N, the complexity and computational cost of the algorithm can be flexibly controlled to adapt to different application requirements. Repeating the random generation process until all particles are initialized ensures that the algorithm can cover multiple areas of the parameter space in the early stages of the search, laying a good foundation for subsequent iterative optimization. This random initialization method provides a good starting point for the particle swarm optimization algorithm, which helps to improve the effect and efficiency of laser cladding welding parameter optimization.

[0056] Example 6, the formula for evaluating the fitness of each particle in step 4 is: ; in is the fitness value of the particle, is the desired melt pool diameter, is the calculated melt pool diameter, is the desired heat affected zone width, is the calculated heat affected zone width, and is the weight coefficient, which is used to adjust the importance of the molten pool diameter and the heat affected zone width in the fitness evaluation; The calculation formula for the molten pool diameter is: ; in is the calculated value of the melt pool diameter, is the laser power, k is the thermal conductivity of the welding object material, is the thermal enthalpy of the welding object material; The calculation formula for the width of the heat affected zone is: ; in is the calculated value of the width of the heat affected zone, is the heat input per unit length.

[0057] Through accurate fitness evaluation, an effective tool is provided for the optimization of laser cladding welding parameters, which helps to achieve high-quality welding results while improving production efficiency and reducing costs.

[0058] In embodiment 7, step 5 comprises the following sub-steps: Step 51, before the particle swarm optimization algorithm starts, the initial position of each particle is set to the individual optimal solution; Step 52, based on the fitness value of the particle in step 4, if the fitness value of the current particle is higher than the fitness value of the corresponding individual optimal solution, then update the current particle to the individual optimal solution, otherwise keep the individual optimal solution unchanged; Step 53: After the individual optimal solutions of all particles are determined, the individual optimal solution with the maximum fitness value is selected as the global optimal solution.

[0059] Ensuring that each particle has a reference point at the beginning of optimization helps the algorithm converge quickly. Dynamically updating the individual optimal solution can ensure that the particle evolves towards a better solution by comparing the fitness value of the current particle with the fitness value of the individual optimal solution, thereby improving the algorithm's search efficiency. Selecting the individual optimal solution with the maximum fitness value as the global optimal solution provides a reference for the optimal solution for the entire particle swarm, helping to guide other particles to search the solution space more effectively.

[0060] In Example 8, in step 6, the particle velocity update formula of the particle swarm optimization algorithm is as follows: ; in yes At time , the velocity of the i-th particle in the j-th dimension; is the velocity of the ith particle in the jth dimension at time t; w is the inertia weight; The weight of particles learning from the individual optimal solution; is the weight of the particle learning towards the global optimal solution; is a random number in the interval [0,1], used to simulate random behavior of particles; is a random number in the interval [0,1], used to simulate random behavior of particles; is the individual optimal position of the ith particle in the jth dimension; is the position of the i-th particle in the j-th dimension at time t; is the value of the global optimal position in the jth dimension; The particle position update formula of the particle swarm optimization algorithm is: ; in It's time The position of the i-th particle at time ; It's time The position of the i-th particle at time ; It's time is the time of the ith particle in the jth dimension.

[0061] By introducing inertia weights, individual learning weights, and global learning weights, particles can balance the ability to explore (search new areas) and exploit (utilize known optimal positions), improving the global and local search capabilities of the algorithm.

[0062] The position update formula ensures that particles move efficiently in the solution space while staying within the defined search range.

[0063] Embodiment 9, the method of obtaining corresponding welding parameters based on the updated particle position and velocity is: Suppose there are three welding parameters: laser power P, welding speed V and powder feeding rate d, the particle position vector Each element in is mapped to the actual range of these three welding parameters. The following is the mapping formula: ; ; ; The mapping function map is defined as: ; in is the position value of the particle, and are the minimum and maximum values ​​of the particle position vector, and are the minimum and maximum values ​​of the corresponding welding parameters.

[0064] Through the mapping formula, the position vector of the particle is mapped to the actual welding parameter range, ensuring that the parameters generated by the particle swarm optimization algorithm are operational and meet the actual process requirements.

[0065] Improve parameter accuracy. The mapping function map can convert the continuous position values ​​of particles into the actual values ​​of welding parameters, which improves the accuracy of parameter optimization and the value of practical application.

[0066] To enhance algorithm flexibility, the mapping method allows the algorithm to work within different welding parameter ranges, making the optimization process adaptable to different materials and welding conditions.

[0067] The following specific examples are used to illustrate the implementation principle of the laser cladding welding parameter optimization control method based on particle swarm optimization: According to the physical phenomena of laser cladding welding, the molten pool dynamics model and heat conduction model were established. These models can simulate the formation of the molten pool, heat distribution and material behavior during welding.

[0068] Determine the optimization target parameters and constraint parameters: The optimization target parameters were determined to be the uniformity of the cladding layer and the minimization of thermal stress, and the constraint parameters were the laser power range (2000-3000 W), welding speed range (5-10 mm / s) and powder feeding rate range (2-5 g / min).

[0069] Randomly initialize the particle swarm: The number of particles N in the particle swarm is determined to be 30.

[0070] For each parameter, a value is randomly generated within the set range using the following formula: Laser power P: P = 2000 + (3000 - 2000) * rand(); Welding speed V: V = 5 + (10 - 5) * rand(); Powder feeding rate d: d = 2 + (5 - 2) * rand(); Where rand() is a random number uniformly distributed in the interval [0, 1].

[0071] Here is a specific example: Particle 1: P1 = 2000 + (3000 - 2000) * 0.3 = 2300 W, V1 = 5 + (10 - 5) *0.5 = 7.5 mm / s, d1 = 2 + (5 - 2) * 0.4 = 2.8 g / min Particle 2: P2 = 2000 + (3000 - 2000) * 0.7 = 2700 W, V2 = 5 + (10 - 5) *0.2 = 6 mm / s, d2 = 2 + (5 - 2) * 0.9 = 4.8 g / min Particle 3: P3 = 2000 + (3000 - 2000) * 0.1 = 2100 W, V3 = 5 + (10 - 5) *0.8 = 9 mm / s, d3 = 2 + (5 - 2) * 0.3 = 2.6 g / min And so on, until the parameters of all 30 particles are initialized.

[0072] Evaluate the fitness of each particle: The fitness of each particle is evaluated using melt pool dynamics and heat conduction models.

[0073] Fitness evaluation formula: ; Update individual optimal solution and global optimal solution: Before the particle swarm optimization algorithm starts, the initial position of each particle is set to the individual optimal solution.

[0074] Based on the fitness value of the particle, if the current particle fitness value is higher than the fitness value of the corresponding individual optimal solution, the current particle is updated to the individual optimal solution, otherwise the individual optimal solution remains unchanged.

[0075] After the individual optimal solutions of all particles are determined, the individual optimal solution with the maximum fitness value is selected as the global optimal solution.

[0076] Adjust particle position and speed; map welding parameters: Mapping formula: P = map(x1, min_x, max_x, 2000, 3000); V = map(x2, min_x, max_x, 5, 10); d = map(x3, min_x, max_x, 2, 5); Assume that the position vector of particle 1 during the iteration is [x1, x2, x3], where x1 = 0.3, x2 = 0.5, and x3 = 0.4. Then the parameter values ​​after mapping are: P1 = map(0.3, 0, 1, 2000, 3000) = 2000 + (3000 - 2000) * 0.3 = 2300W; V1 = map(0.5, 0, 1, 5, 10) = 5 + (10 - 5) * 0.5 = 7.5 mm / s; d1 = map(0.4, 0, 1, 2, 5) = 2 + (5 - 2) * 0.4 = 2.8 g / min; In this way, we get specific P, V and d values, which will be used for actual welding parameter control.

[0077] The laser cladding welding parameter optimization control method based on particle swarm optimization can be specifically applied to achieve high-quality welding results.

[0078] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A laser cladding welding parameter optimization control method based on particle swarm optimization, characterized in that: The following steps are involved: Step 1, establishing a molten pool dynamics model and a heat conduction model based on the physical phenomena of laser cladding welding; Step 2, determine the optimization target parameters and constraint condition parameters; Step 3, encode the optimization target parameters and constraint condition parameters into particle positions, and randomly initialize a group of particles in the parameter space, where each particle corresponds to a set of welding parameters; Step 4, using the melt pool dynamics and heat conduction model to evaluate the fitness of each particle; Step 5: Update the individual optimal solution and the global optimal solution according to the fitness of each particle; Step 6, adjusting the position and speed of the particles according to the speed update formula and the position update formula of the particle swarm optimization algorithm, and obtaining corresponding welding parameters based on the updated position and speed of the particles; Step 7, converting the optimized welding parameters into welding control instructions, and exchanging the welding control instructions with the laser cladding welding controller.

2. The laser cladding welding parameter optimization control method based on particle swarm optimization according to claim 1 is characterized in that: In step 5, the molten pool dynamics and heat conduction models are run using the updated welding parameters. If the optimization target is not achieved, return to step 1 to continue iteration.

3. The laser cladding welding parameter optimization control method based on particle swarm optimization according to claim 2 is characterized in that: The optimization target parameters in step 2 are the uniformity of the cladding layer and the minimization of thermal stress, and the constraint parameters are the laser power range, welding speed range, and powder feeding rate range.

4. The laser cladding welding parameter optimization control method based on particle swarm optimization according to claim 3 is characterized in that: In step 3, a set of welding parameters includes laser power P, welding speed V and powder feeding rate d.

5. The laser cladding welding parameter optimization control method based on particle swarm optimization according to claim 4 is characterized in that: In step 3, the method to randomly initialize a group of particles in the parameter space is: Step 31, determining the number of particles N in the particle group; Step 32, for each parameter laser power P, welding speed V and powder feeding rate d, a value is randomly generated within the set range using the following formula: ； ； ； in , and is A random number uniformly distributed within a range; Step 33, repeat step 32 until all N particles are initialized.

6. The laser cladding welding parameter optimization control method based on particle swarm optimization according to claim 5 is characterized in that: The formula for evaluating the fitness of each particle in step 4 is: ； in is the fitness value of the particle, is the desired melt pool diameter, is the calculated melt pool diameter, is the desired heat affected zone width, is the calculated heat affected zone width, and is the weight coefficient, which is used to adjust the importance of the molten pool diameter and the heat affected zone width in the fitness evaluation; The calculation formula for the molten pool diameter is: ； in is the calculated value of the melt pool diameter, is the laser power, k is the thermal conductivity of the welding object material, is the thermal enthalpy of the welding object material; The calculation formula for the width of the heat affected zone is: ； in is the calculated value of the width of the heat affected zone, is the heat input per unit length.

7. The laser cladding welding parameter optimization control method based on particle swarm optimization according to claim 6 is characterized in that: Step 5 includes the following sub-steps: Step 51, before the particle swarm optimization algorithm starts, the initial position of each particle is set to the individual optimal solution; Step 52, based on the fitness value of the particle in step 4, if the fitness value of the current particle is higher than the fitness value of the corresponding individual optimal solution, then update the current particle to the individual optimal solution, otherwise keep the individual optimal solution unchanged; Step 53: After the individual optimal solutions of all particles are determined, the individual optimal solution with the maximum fitness value is selected as the global optimal solution.

8. The particle swarm optimization-based laser cladding welding parameter optimization control method according to claim 7, characterized in that: In step 6, the particle velocity update formula of the particle swarm optimization algorithm is as follows: ； in yes At time , the velocity of the i-th particle in the j-th dimension; is the velocity of the ith particle in the jth dimension at time t; w is the inertia weight; The weight of particles learning from the individual optimal solution; is the weight of the particle learning towards the global optimal solution; is a random number in the interval [0,1], used to simulate random behavior of particles; is a random number in the interval [0,1], used to simulate random behavior of particles; is the individual optimal position of the ith particle in the jth dimension; is the position of the i-th particle at time t in the j-th dimension; is the value of the global optimal position in the jth dimension; The particle position update formula of the particle swarm optimization algorithm is: ; in It's time The position of the i-th particle at time ; It's time The position of the i-th particle at time ; It's time is the time of the i-th particle in the j-th dimension.

9. The laser cladding welding parameter optimization control method based on particle swarm optimization according to claim 8, characterized in that: The method to obtain the corresponding welding parameters based on the updated particle position and velocity is: There are three welding parameters: laser power P, welding speed V and powder feeding rate d, particle position vector, Each element in is mapped to the actual range of these three welding parameters. The following is the mapping formula: ； ； ； The mapping function map is defined as: ； in is the position value of the particle, and are the minimum and maximum values ​​of the particle position vector, and are the minimum and maximum values ​​of the corresponding welding parameters.

Images