Workpiece detection system and workpiece scheduling method based on multi-objective particle swarm optimization algorithm

Abstract

This invention provides a workpiece inspection system and a workpiece scheduling method based on a multi-objective particle swarm optimization algorithm. The system sets up workpiece stop stations on the production line corresponding to the testing machine, and applies the multi-objective particle swarm optimization algorithm to allocate carriers based on the relationship between the workpiece stop stations and the testing machine. The method achieves optimal conditions between workpieces, between workpieces and the testing machine, and between individual components and the overall system, based on actual production conditions. This significantly reduces material blockage on the feeding line, scientifically allocates carriers, and extends equipment lifespan.

Description

Technical Field

[0001] This invention relates to the field of detection technology, specifically to a production line workpiece detection system and a workpiece scheduling method based on a multi-objective particle swarm optimization algorithm. Background Technology

[0002] In display screen testing, with the development of industrial automation technology, the process is gradually transitioning from manual testing to intelligent and automated assembly line operations. However, in the wearable display testing industry, such as watches and VR glasses, high hourly production capacity is required. Factories often place numerous testing machines on both sides of the assembly line. Robotic arms or servo motion devices pick up the carriers containing screens at the loading point of the conveyor belt and place them into the testing machines for testing. After testing, the robotic arms or servo motion devices remove the carriers and place them on the corresponding unloading conveyor belt, allowing them to flow downstream for the next process. During this process, the number of workpieces on the loading conveyor belt cannot be determined in real time and is constantly changing, making it impossible to evenly distribute the carriers to the testing machines on one side within a unit of production time.

[0003] Typically, by comparing the idle waiting time of each testing machine, the one with the longer idle time is given priority for loading. If the times are the same, the load is prioritized for the next testing machine in the same section of the equipment. Under ideal conditions, this method can achieve a uniform distribution of carriers to each testing machine.

[0004] However, in actual production, if the amount of material arriving at upstream equipment is concentrated within a unit of production time, and some testing machines in this equipment occasionally experience hardware alarms requiring manual intervention, thus extending the testing time; or if network congestion occurs during a certain period, increasing the testing time; or if undesirable results occur during testing, preventing subsequent testing and necessitating immediate material unloading. All these situations will result in some testing machines within the equipment being partially busy and others idle. This phenomenon can lead to occasional blockages on some feeding conveyor belts, affecting the overall line's uptime; furthermore, the varying usage frequencies of testing machines within the same section affect the lifespan of the entire equipment section. Summary of the Invention

[0005] The purpose of this invention is to provide a workpiece inspection system and a workpiece scheduling method based on a multi-objective particle swarm optimization algorithm.

[0006] This invention provides a workpiece scheduling method based on a multi-objective particle swarm optimization algorithm, which controls the scheduling and allocation of workpieces traveling along a test section production line to different testing machines for testing. The workpieces are placed on a carrier and travel along the production line, and multiple testing machines are arranged along one or both sides of the production line. The method includes the following steps:

[0007] A certain number of workpieces are released into the testing section production line. Multiple workpiece stop stations are set along the production line, and the workpiece stop stations are respectively set at the entrance of the testing channel of the testing machine.

[0008] When there is an idle testing machine, after the workpiece travels to the workpiece stop station corresponding to the testing machine, it stops traveling and avoids the production line travel route, waiting to be transferred to the testing machine for testing.

[0009] The current number of available testing machines, the current number of available workpiece stop stations, the number of tests performed by the testing machines, the initial position of the workpiece, and the position of the testing machines are obtained. An objective function is then constructed, which represents the number of workpieces that can be carried at present.

[0010] Calculate the time required for the workpiece to reach different workpiece stopping positions and establish constraints.

[0011] Using a multi-objective particle swarm optimization algorithm, the optimal number of workpieces to be released and the optimal number of test machines to be allocated for the next time period are calculated based on the objective function and constraints. The number of workpieces to be released is then adjusted, and the workpieces are allocated to the optimal test machines for testing.

[0012] As a further improvement of the present invention, when there is an idle testing machine, after the workpiece travels to the workpiece stop station corresponding to the testing machine, it pauses its travel and avoids the production line path, specifically including:

[0013] When there is an idle testing machine, the sensor at the corresponding workpiece stop station detects the operation status of the production line;

[0014] When the sensor detects that a workpiece is passing by, the drive device located at the workpiece stop station descends and intercepts the workpiece through the interceptor on it. The drive device then rises to lift the workpiece, which is then transferred to the testing machine for testing.

[0015] As a further improvement of the present invention, the construction of the objective function specifically includes:

[0016] Construct the objective function based on the following formula:

[0017]

[0018] Where f represents the current number of workpieces that can be handled, n represents the total number of testing stations on all testing machines, x represents the current number of idle testing machines, y represents the current number of idle workpiece stopping stations, and T i pos(X) represents the current cumulative number of tests performed on the i-th test machine, M represents the total number of tests performed on the test machine, and pos(X) represents the total number of tests performed on the i-th test machine. i) represents the position of the test machine represented by the i-th task, pos(X0) represents the initial position of the workpiece, and K represents the proportion of the number of shielded test machines to the total number.

[0019] As a further improvement of the present invention, the establishment of the constraint conditions specifically includes:

[0020] Establish constraints based on the following formula:

[0021] pos(X o =os(test section pipeline start position)

[0022]

[0023] Wherein, the initial position of the workpiece is the starting position of the test section of the production line, t represents the time it takes for the workpiece to travel to the corresponding test machine, and v represents the speed of the production line.

[0024] As a further improvement of the present invention, the step of using a multi-objective particle swarm optimization algorithm to calculate the optimal number of workpieces released and the optimal number of testing machines allocated in the next time period based on the objective function and constraints specifically includes:

[0025] Population particle initialization is performed, where each particle represents a set of optimal workpiece release quantities n that need to be optimized. c+1 And the optimal number of test machines allocated to n t+1 ;

[0026] The fitness value of each particle is calculated based on the objective function, the dominance of each particle is compared pairwise, and the obtained non-dominated solutions are stored in the Pareto solution set.

[0027] Perform fitness value evaluation, calculate the objective function value of each particle, and store it in vector form;

[0028] The position and velocity of the particles are updated by updating the formula, and non-dominated solutions that satisfy the constraints are arbitrarily selected from the Pareto solution set as the historical best position and the global best position.

[0029] Check the Pareto solution set dominance of each particle in the particle swarm. If it is a non-dominant solution compared to the particles in the solution set, store the particle in the Pareto solution set as a new Pareto solution set obtained at the moment, and dynamically update the Pareto solution set.

[0030] Repeat the above steps until the iteration is complete, and obtain the final optimal number of workpieces released, n. c+1 And the optimal number of test machines allocated to n t+1 .

[0031] As a further improvement of the present invention, the population particle initialization specifically includes:

[0032] Within the initialization interval, the initial positions and initial velocities of the particle swarm are initialized, and it is checked whether they meet the constraints.

[0033] The initialization parameters are selected as follows: inertia factor w = 0.5 + rand, acceleration constant c1 = c2 = 1.5, particle swarm size popsize = 50, and the number of iterations is set to 50.

[0034] As a further improvement to the present invention, the update formula is as follows:

[0035] V i (t1+1)=w*V i (t1)+c1*rand()*(pbest(t1)-X i (t1))+c2*rand()*(gbest(t1)-X i (t1))

[0036] X i (t1+1)=X i (t1)+V i (t1)

[0037] Among them, V i (t1+1) represents the particle's velocity value at the next update time, X i (t1+1) represents the particle's position value at the next update time, V i (t1) represents the particle's velocity value at the current update moment, X i (t1) represents the position value of the particle at the current update time, w represents the inertial weight factor distributed in the interval [0,1], c1 and c2 represent acceleration constants, rand() represents a random number distributed in the interval [0,1], pbesti(t1) represents the position of the individual optimal solution at the update time, gbesti(t1) represents the position of the global optimal solution at the update time, and t1 represents the current update time.

[0038] As a further improvement to the present invention, it also includes:

[0039] When the production line is fully loaded, all workpiece stop stations have workpieces at their designated locations, and there are no available testing machines, the release of workpieces to the testing section production line is stopped.

[0040] The present invention also provides a workpiece inspection system for an assembly line, comprising a test section assembly line, workpieces carrying workpieces traveling along the assembly line, and a plurality of test machines arranged along one or both sides of the assembly line, characterized in that:

[0041] The production line includes at least one loading line, at least one unloading line, and a motor that drives the loading line and the unloading line;

[0042] Multiple testing machines are arranged sequentially along the extension direction of the feeding line;

[0043] Multiple workpiece stop stations are set on the feeding line. Each workpiece stop station corresponds to the entrance of the testing machine. Each workpiece stop station is equipped with a sensor and a shifting device. The sensor is configured to detect the workpiece traveling along the feeding line, and the shifting device is configured to pause the workpiece's travel and shift it to avoid the production line's path.

[0044] The assembly line workpiece inspection system also includes a host computer, which includes a memory and a processor. The memory is used to store executable instructions, and the processor is used to implement the workpiece scheduling method of the multi-objective particle swarm algorithm when running the executable instructions stored in the memory.

[0045] As a further improvement of the present invention, each of the workpiece stop stations is provided with a driving device capable of lifting and lowering. The driving device is provided with an interceptor. The driving device is configured to descend when a workpiece passes by the workpiece stop station in which it is located, intercept the workpiece through the interceptor, and then rise after interception.

[0046] As a further improvement of the present invention, the workpiece stop station is provided with a sensor, which is configured to detect whether a workpiece passes through the workpiece stop station.

[0047] As a further improvement of the present invention, a plurality of testing machines are evenly spaced on one side of the feeding line, and the workpiece stop station is evenly spaced on the feeding line corresponding to the testing machines.

[0048] As a further improvement of the present invention, the workpiece is placed on a carrier assembly line and moves along it. The shifting device includes two opposing lifting carrier plates. The opposing ends of the lifting carrier plates are provided with protruding bottom baffles. The area between the lifting carrier plates is used to accommodate the carrier. The ends of the lifting carrier plates are provided with intercepting baffles. When the carrier is located between the lifting carrier plates, the intercepting baffles prevent the carrier from moving along the feeding line. The bottom baffles support the bottom of the carrier.

[0049] The beneficial effects of this invention are as follows: This invention sets up workpiece stop stations on the production line corresponding to the testing machine, and applies a multi-objective particle swarm optimization algorithm to allocate carriers based on the relationship between the workpiece stop stations and the testing machine. The method used achieves optimal conditions between carriers, between carriers and the testing machine, that is, between individuals, and between individuals and the whole, based on the actual production situation. This can largely avoid material blockage on the feeding line, scientifically allocate carriers, and extend the service life of the equipment. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of a workpiece inspection system and a workpiece scheduling method based on a multi-objective particle swarm algorithm according to an embodiment of the present invention.

[0051] Figure 2 This is a schematic diagram of a shifting device according to one embodiment of the present invention.

[0052] Figure 3 This is a schematic diagram of the steps of a workpiece scheduling method based on a multi-objective particle swarm algorithm according to an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0054] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0055] This embodiment provides a production line workpiece inspection system and a workpiece scheduling method based on a multi-objective particle swarm optimization algorithm implemented through the system. The scheduling method is used to control the scheduling and allocation of workpieces traveling along the test section production line to different testing machines for testing. It can form a scientific allocation between the number of workpieces flowing into the system and the available testing machines, thereby realizing the adaptive allocation of the number of workpieces by the equipment, significantly enhancing the stability and robustness of the equipment, and effectively extending the service life of the equipment.

[0056] like Figure 1As shown, the assembly line workpiece inspection system provided in this embodiment includes a test section assembly line 1, a workpiece 2 traveling along the assembly line 1, and multiple test machines 4 arranged along one or both sides of the assembly line 1.

[0057] Specifically, in this embodiment, the workpiece 2 is a display screen. To avoid damage during product transfer, before testing, the workpiece 2 is typically placed on a carrier 3 and electrically connected to protect the product. The carrier 3 then transfers the workpiece and facilitates signal transfer. During the operation of the testing system, the workpiece 2 is placed on the carrier 3, and the carrier 3 carrying the workpiece 2 travels on the conveyor belt of the production line 1. The movement of the carrier 3 on the loading line 11 and unloading line 12, as described below, is also the movement of the workpiece 3. In other embodiments of the invention, when the workpiece 2 is another product, it can also be directly placed on the conveyor belt of the production line 1.

[0058] The production line 1 includes at least one feeding line 11, at least one unloading line 12, and motors for driving the feeding line 11 and the unloading line 12. Multiple testing machines 4 are arranged sequentially along the extension direction of the feeding line 11.

[0059] Specifically, in this embodiment, the assembly line workpiece inspection system includes two parallel feeding lines 11, namely the first feeding line 111 and the second feeding line 112. Multiple testing machines 4 are evenly spaced and arranged on the outer sides of the two feeding lines 11, and a discharging line 12 is provided on the inner sides of each feeding line 11, namely the first discharging line 121 and the second discharging line 122. Taking the first feeding line 111 as an example, during operation, the workpiece 2 is first placed on the carrier 3 and moves to the first feeding line 111 to await testing. After the workpiece 2 is fed into the testing machine 4 on the same side as the first feeding line 111 for testing, it is placed on the first discharging line 121 for unloading. By setting up two parallel testing lines, the system's working efficiency can be improved. In other embodiments of the present invention, the number of feeding lines 11 and discharging lines 12 can be adjusted as needed, and testing machines 4 can also be set on both sides of a feeding line 11 to meet higher testing requirements.

[0060] Furthermore, the production line 1 also includes a return line 13, which is used to return the carrier 3 to the front end for reuse. As described above, after testing is completed on the testing machine 4 in multiple testing stages, the workpiece 2 needs to be separated from the carrier 3, and the carrier 3 returns to the front end via the return line to achieve reuse. Specifically, in this embodiment, a first return line 131 and a second return line 132 are respectively provided inside the first unloading line 121 and the second unloading line 122. In other embodiments, if the workpiece does not require carrier support, a return line may not be provided.

[0061] The number of testing machines 4 can be adjusted according to factors such as the type of workpiece 2 being tested and the time required to test the workpiece 2. For example, in this embodiment, the workpiece 2 is a screen for devices such as smartwatches or VR glasses. Four testing machines 4 are respectively set on the outside of each feeding line 11, and each testing machine 4 has two testing channels. That is, in this embodiment, the detection system can test 16 workpieces 2 simultaneously under full load.

[0062] Multiple workpiece stop stations 5 are set on the feeding line 11. The workpiece stop stations 5 are respectively set at the test station entrance of the test machine 4. Sensors and displacement devices 51 are respectively set at the workpiece stop stations 5. The sensors are configured to detect the carrier 3 traveling along the feeding line 11. The displacement devices 51 are configured to pause the travel of the carrier 3 and shift it to avoid the travel path of the production line 1.

[0063] When an idle testing machine 4 is available, the carrier 3 travels to the workpiece stop station 5 corresponding to that testing machine 4, pauses its movement, and avoids the travel path of the production line 1, waiting to transfer the workpiece 2 to the testing machine 4 for testing. The carrier 3 can be moved out of the travel path of the production line 1 by the shifting device 51, thereby pausing its movement to wait for testing, and the waiting process will not affect the continued movement of subsequent carriers 3.

[0064] For example, in this embodiment, corresponding to the number of test channels of the testing machine 4, eight workpiece stop stations 5 are respectively set on the first feeding line 111 and the second feeding line 112.

[0065] Furthermore, each workpiece stop station 5 is equipped with a sensor, which is configured to detect whether a workpiece 2 has passed through the workpiece stop station 5.

[0066] Each workpiece stop station 5 has a shifting device 51 that includes a drive device capable of lifting and lowering. The drive device includes a drive motor and a cylinder, etc. An interceptor is provided on the drive device. The drive device is configured to descend when a carrier 3 passes by at the workpiece stop station 5 it is in, intercept the carrier 3 through the interceptor, and then rise after interception.

[0067] Specifically, such as Figure 2 As shown, in this embodiment, the shifting device 51 includes two opposing lifting plates 511. The opposite ends of the lifting plates 511 are provided with protruding bottom baffles 512. The area between the lifting plates 511 is used to accommodate the carrier 3. The ends of the lifting plates 511 are provided with intercepting baffles 513. When the carrier 3 is located between the lifting plates 511, the intercepting baffles 513 prevent the carrier 3 from moving along the feeding line 11, and the bottom baffles support the bottom of the carrier 3.

[0068] During operation, when the sensor detects that a carrier 3 is passing by, the motor-driven cylinder first controls the lifting plate 511 to move downward. After the carrier 3 moves between the lifting plates 511, the intercepting baffle 513 intercepts the carrier 3. Then, the motor-driven cylinder controls the lifting plate 511 to move upward, so that the carrier 3 is intercepted at the workpiece stop station 5, and it will not affect the subsequent carrier 3 from continuing to move on the feeding line 11.

[0069] In addition to moving up and down to avoid the path of the assembly line 1, in other embodiments of the present invention, the shifting device 51 can also be configured to move left and right or diagonally upward, as long as it can move the carrier 3 to an empty space to avoid the direction of the assembly line 1.

[0070] Furthermore, in this embodiment, the feeding line 11 is segmented, including multiple connected sub-feeding lines 11a. Each sub-feeding line 11a corresponds to a testing machine 4, and a workpiece stop station 5 is set at each sub-feeding line 11a. For example, in this embodiment, one side of the feeding line 11 consists of four connected sub-feeding lines 11a, and the testing machines 4 are respectively set at the ends of the sub-feeding lines 11a. Correspondingly, two workpiece stop stations 5 are set at the end of each feeding line segment. The device structure at the workpiece stop stations 5 is more complex. Setting it as a multi-segment feeding line 11a structure allows for convenient debugging and maintenance of each feeding line segment 11 and the workpiece stop stations 5.

[0071] In other embodiments of the present invention, the feeding line may be a single section, or a section of feeding line may be configured to correspond to multiple testing machines.

[0072] The assembly line workpiece inspection system also includes a robotic arm, which grips the workpiece 2 and moves it between the assembly line 1 and the testing machine 4. In this embodiment, the robotic arm performs two-dimensional movement in a planar direction. Therefore, a material feeding and shifting device is provided on the unloading line 12, which moves at the same height as the shifting device 51 on the loading line 11 and at the same height as the inlet of the testing machine 4. The robotic arm moves the carrier 3 and the workpiece 2 placed on it sequentially between the shifting device 51 on the loading line 11, the inlet position of the testing machine 4, and the shifting device 51 on the unloading line 12. After the carrier 3 and the workpiece 2 placed on it are transferred to the material feeding and shifting device on the unloading line 12, the material feeding and shifting device moves downward, causing the carrier 3 to fall back onto the unloading line 12.

[0073] like Figure 3 The diagram shows the steps of the task scheduling method based on the multi-objective particle swarm optimization algorithm. The particle swarm optimization algorithm has the advantages of fast convergence speed, few parameters, and simple algorithm implementation. It includes:

[0074] S1: Release a certain amount of carrier 3 into the test section production line 1. The carrier carries the workpiece 2 to be tested. Multiple workpiece stop stations 5 are set along the production line 1. The workpiece stop stations 5 are respectively set at the test channel entrance of the test machine 4.

[0075] S2: When there is an idle testing machine 4, the carrier 3 travels to the workpiece stop station 5 corresponding to the testing machine 4, stops traveling and avoids the travel route of the production line 1, and waits to transfer the workpiece 2 to the testing machine 4 for testing.

[0076] S3: Obtain the current number of available test machines 4, the current number of available workpiece stop stations 5, the number of tests performed by test machine 4, the initial position of workpiece 2, and the position of test machine 4. Construct an objective function, which represents the current number of workpieces 2 that can be carried.

[0077] S4: Calculate the time required for workpiece 2 to reach different workpiece stop stations 5, and establish constraints.

[0078] S5: Using the multi-objective particle swarm optimization algorithm, calculate the optimal number of workpieces to be released and the optimal number of test machines 4 to be assigned in the next time period based on the objective function and constraints, and then adjust the number of workpieces to be released and assign workpiece 2 to the optimal test machine 4 for testing.

[0079] In step S2, for the assembly line workpiece inspection system in this embodiment, it specifically includes:

[0080] S21: When there is an idle test machine 4, the sensor at the corresponding workpiece stop station 5 detects the operating status of the production line 1.

[0081] S22: When the sensor detects that a carrier 3 has passed by, the drive device located at the workpiece stop station 5 descends and intercepts the carrier 3 through the interceptor on it. The drive device rises and lifts the carrier 3, and the workpiece 2 waits to be transferred to the testing machine 4 for testing.

[0082] In step S3, based on the working status of the production line workpiece inspection system, the objective function is constructed according to the following formula:

[0083]

[0084] Where f represents the current number of workpieces 2 that can be carried, n represents the total number of testing stations of all testing machines 4, x represents the current number of idle testing machines 4, y represents the current number of idle workpiece stopping stations 5, and T i This represents the current cumulative number of tests performed on the i-th test machine 4, where M represents the total number of tests performed on test machine 4, and pos(X) i) represents the position of the test machine 4 represented by the i-th task, pos(X0) represents the initial position of workpiece 2, and K represents the proportion of the number of shielded test machines 4 to the total number. Each test machine has a shielding function. Setting this function means that the test machine is temporarily not present.

[0085] For example, in this embodiment, there are a total of 16 test stations, i.e., n=16.

[0086] The value of the objective function, also known as the fitness value, is used to evaluate the quality of a particle's position and determine whether to update the historical best position of an individual particle and the historical best position of the group, ensuring that the particle searches in the direction of the optimal solution.

[0087] In step S4, constraints are established according to the following formula:

[0088] pos(X o ) = pos (starting position of the test section pipeline) Equation 2

[0089]

[0090] Wherein, the initial position of workpiece 2 is the starting position of the test section production line 1, t represents the time it takes for workpiece 2 to travel to the corresponding test machine 4, and v represents the running speed of production line 1.

[0091] Equation 2 limits the space for particle search, i.e., the range of values ​​for the independent variable. Equation 3 sets a reasonable velocity range to limit the maximum velocity of the particles, thereby balancing the algorithm's exploration and development capabilities.

[0092] In step S5, the optimal number of vehicles 3 to be released and the optimal number of test vehicles 4 to be allocated are calculated for the next time period using a multi-objective particle swarm optimization algorithm based on the objective function and constraints. Specifically, this includes:

[0093] S51: Initialize the population particles, where each particle represents a set of optimal workpiece release quantities n that need to be optimized. c+1 And the optimal test machine 4 allocation number n t+1 .

[0094] S52: Calculate the fitness value of each particle according to the objective function, compare the dominance of each particle pairwise, and store the obtained non-dominated solutions through the Pareto solution set.

[0095] S53: Perform fitness value evaluation, calculate the objective function value of each particle, and store it in vector form.

[0096] S54: Update the position and velocity of the particle by updating the formula, and arbitrarily select a non-dominated solution that satisfies the constraints from the Pareto solution set as the historical best position and the global best position.

[0097] S55: Check the Pareto solution set dominance of each particle in the particle swarm. If it is a non-dominant solution compared to the particles in the solution set, store the particle in the Pareto solution set as a new Pareto solution set obtained at the moment, and dynamically update the Pareto solution set.

[0098] S56: Repeat the above steps until the iteration is complete, and obtain the final optimal number of workpieces released, n. c+1 And the optimal test machine 4 allocation number n t+1 .

[0099] In step S51, population particle initialization is performed, specifically including:

[0100] Within the initialization interval, the initial positions and initial velocities of the particle swarm are initialized, and it is checked whether they meet the constraints.

[0101] In this embodiment, the initialization parameters are selected as follows: inertia factor w = 0.5 + rand, acceleration constant c1 = c2 = 1.5, particle swarm size popsize = 50, and the number of iterations is set to 50.

[0102] The inertia factor *w* represents the influence of the velocity of the previous generation of particles on the velocity of the current generation of particles, i.e., the degree of confidence a particle has in its current state of motion, as particles move inertially based on their own velocities. *w* enables particles to maintain their inertia and tends to expand the search space. A larger *w* value indicates a stronger ability to explore new regions and a stronger global optimization capability, but a weaker local optimization capability. Conversely, a smaller *w* value indicates a weaker global optimization capability and a stronger local optimization capability. A larger *w* is beneficial for global search, escaping local extrema and avoiding getting trapped in local optima; a smaller *w* value is beneficial for local search, allowing the algorithm to converge quickly to the optimal solution. To achieve a balance between search speed and search accuracy, the inertia factor *w* should not be a fixed constant; therefore, in this implementation, the inertia factor *w* is a variable.

[0103] The acceleration constant c1 represents the weight of the particle's next action derived from its own experience, accelerating the particle towards its individual optimal position. The acceleration constant c2 represents the weight of the particle's next action derived from the experience of other particles, accelerating the particle towards the group's optimal position. To maintain a balance between convergence speed and search performance, in this embodiment, both c1 and c2 are set to 1.5.

[0104] Smaller particle swarm sizes are prone to getting trapped in local optima, while larger swarm sizes can improve convergence and find the global optimum faster, but the computational cost of each iteration will also increase accordingly. Therefore, in this implementation, the particle swarm size is set to 50.

[0105] A smaller number of iterations will make the final solution unstable, while a larger number of iterations will increase the computational load. Therefore, in this embodiment, the number of iterations is set to 50.

[0106] In other embodiments of the present invention, the initialization parameters may be adjusted according to the scale and computing power of the test system.

[0107] In step S54, the updated formula is:

[0108] V i (1+1)=w*V i (1)+c1*rand()*( (1)-X i (1))+c2*

[0109] rand()*((1)-X i (1) Equation 4

[0110] X i (t1+1)=X i (t1)+V i (t1) Equation 5

[0111] Among them, V i (t1+1) represents the particle's velocity value at the next update time, X i (t1+1) represents the particle's position value at the next update time, V i (t1) represents the particle's velocity value at the current update moment, X i (t1) represents the position value of the particle at the current update time, w represents the inertial weight factor distributed in the interval [0,1], c1 and c2 represent acceleration constants, rand() represents a random number distributed in the interval [0,1], pbesti(t1) represents the position of the individual optimal solution at the update time, gbesti(t1) represents the position of the global optimal solution at the update time, and t1 represents the current update time.

[0112] Furthermore, the task 2 scheduling method based on the multi-objective particle swarm optimization algorithm also includes:

[0113] When production line 1 is at full load, all workpiece stop stations 5 have carriers 3 stopped, and there are no idle testing machines 4, the release of carriers 3 to the testing section of production line 1 is stopped. In a specific embodiment, a buffer area can be set at the front end of the testing equipment. The buffer area is equipped with an intercepting cylinder (not shown). When the production line is at full load, the intercepting cylinder in the buffer area intercepts the workpieces and stops feeding. Thus, when the entire system is at full load, it can avoid continuing to input too many carriers 3 loaded with workpieces 2 into the system.

[0114] In summary, this invention sets up workpiece stop stations on the production line corresponding to the testing machine, and applies a multi-objective particle swarm optimization algorithm to allocate carriers based on the relationship between the workpiece stop stations and the testing machine. The method used achieves optimal conditions between carriers, between carriers and the testing machine, that is, between individuals, and between individuals and the whole, based on the actual production situation. This can largely avoid material blockage on the feeding line, scientifically allocate carriers, and extend the service life of the equipment.

[0115] It should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0116] The detailed descriptions listed above are merely specific descriptions of feasible implementations of the present invention and are not intended to limit the scope of protection of the present invention. All equivalent implementations or modifications made without departing from the spirit of the present invention should be included within the scope of protection of the present invention.

Claims

1. A workpiece scheduling method based on a multi-objective particle swarm optimization algorithm, used to control the scheduling and allocation of workpieces traveling along a test section production line to different testing machines for testing, wherein the workpieces travel along the production line, and multiple testing machines are arranged along one or both sides of the production line, characterized in that, Including the following steps: A certain number of workpieces are released into the testing section production line. Multiple workpiece stop stations are set along the production line, and the workpiece stop stations are respectively set at the entrance of the testing channel of the testing machine. When an idle testing machine is available, the workpiece stops moving at the workpiece stop station corresponding to that testing machine and avoids the production line's path, waiting to be transferred to the testing machine for testing. Specifically, when an idle testing machine is available, the sensor at the workpiece stop station corresponding to it detects the production line's operating status. When the sensor detects a workpiece passing by, the drive device at the workpiece stop station descends, intercepting the workpiece with an interceptor located thereon. The drive device then rises to lift the workpiece, which waits to be transferred to the testing machine for testing. The current number of available testing machines, the current number of available workpiece stop stations, the number of tests performed by the testing machines, the initial position of the workpiece, and the position of the testing machines are obtained. An objective function is then constructed, which represents the number of workpieces that can be carried at present. Calculate the time required for the workpiece to reach different workpiece stopping positions and establish constraints. Using a multi-objective particle swarm optimization algorithm, the optimal number of workpieces to be released and the optimal number of test machines to be allocated for the next time period are calculated based on the objective function and constraints. Then, the number of workpieces to be released and the workpieces to be allocated to the optimal test machines for testing are adjusted.

2. The workpiece scheduling method based on multi-objective particle swarm optimization algorithm according to claim 1, characterized in that, The specific components of the objective function to be constructed include: Construct the objective function based on the following formula: in, f This indicates the number of workpieces that can be carried at present. n This indicates the number of test stations for all test machines. x This indicates the number of currently available test machines. y This indicates the number of currently idle workstations. T i This represents the current cumulative number of tests performed by the i-th test machine. M This indicates the total number of tests performed by the testing machine. pos(X i ) This indicates the position of the test machine represented by the i-th task. pos(X 0 ) Indicates the initial position of the workpiece. K This indicates the percentage of shielded units on the test equipment out of the total number.

3. The workpiece scheduling method based on multi-objective particle swarm optimization algorithm according to claim 2, characterized in that, The established constraints specifically include: Establish constraints based on the following formula: The initial position of the workpiece is taken as the starting position of the test section assembly line. t This indicates the time it takes for the workpiece to travel to the corresponding testing machine. v This indicates the speed at which the production line operates.

4. The workpiece scheduling method based on multi-objective particle swarm optimization algorithm according to claim 3, characterized in that, The process of using a multi-objective particle swarm optimization algorithm to calculate the optimal number of workpieces released and the optimal number of testing machines allocated in the next time period, based on the objective function and constraints, specifically includes: Population particle initialization is performed, where each particle represents a set of optimal workpiece release quantities that need to be optimized. n c+1 And the optimal number of test machines allocated n t+1 ; The fitness value of each particle is calculated based on the objective function, the dominance of each particle is compared pairwise, and the obtained non-dominated solutions are stored in the Pareto solution set. Perform fitness value evaluation, calculate the objective function value of each particle, and store it in vector form; The position and velocity of the particles are updated by updating the formula, and non-dominated solutions that satisfy the constraints are arbitrarily selected from the Pareto solution set as the historical best position and the global best position. Check the Pareto solution set dominance of each particle in the particle swarm. If it is a non-dominant solution compared to the particles in the solution set, store the particle in the Pareto solution set as a new Pareto solution set obtained at the moment, and dynamically update the Pareto solution set. Repeat the above steps until the iteration is complete, and obtain the final optimal number of workpieces to release. n c+1 And the optimal number of test machines allocated n t+1 .

5. The workpiece scheduling method based on multi-objective particle swarm optimization algorithm according to claim 4, characterized in that, The update formula is: in, V i ( t1+1 () represents the particle's velocity value at the next update time. X i ( t1+1 () represents the particle's position value at the next update time. V i ( t1 () represents the particle's velocity value at the current update moment. X i ( t1 () represents the particle's position value at the current update time. w c1 and c2 represent the inertial weighting factors distributed in the interval [0,1], and the acceleration constants. rand () represents a random number distributed in the interval [0,1]. pbesti ( t1 ) represents the position of the individual optimal solution at the update time. gbesti ( t1 ) represents the position of the global optimal solution at the update time. t1 Indicates the current update time.

6. A workpiece inspection system for an assembly line, comprising a testing section assembly line, workpieces traveling along the assembly line, and a plurality of testing machines arranged along one or both sides of the assembly line, characterized in that: The production line includes at least one loading line, at least one unloading line, and a motor that drives the loading line and the unloading line; Multiple testing machines are arranged sequentially along the extension direction of the feeding line; Multiple workpiece stop stations are set on the feeding line. Each workpiece stop station corresponds to the entrance of the testing machine. Each workpiece stop station is equipped with a sensor and a shifting device. The sensor is configured to detect the workpiece traveling along the feeding line, and the shifting device is configured to pause the workpiece's travel and shift it to avoid the production line's path. Each of the workpiece stop stations is provided with a drive device capable of lifting and lowering. The drive device is provided with an interceptor. The drive device is configured to descend when a workpiece passes by the workpiece stop station it is in, intercept the workpiece through the interceptor, and then rise after interception. The assembly line workpiece inspection system further includes a host computer, which includes a memory and a processor. The memory is used to store executable instructions, and the processor is used to implement the workpiece scheduling method of the multi-objective particle swarm algorithm according to any one of claims 1 to 5 when running the executable instructions stored in the memory.

7. The assembly line workpiece inspection system according to claim 6, characterized in that, Multiple testing machines are evenly spaced on one side of the feeding line, and the workpiece stop station is evenly spaced on the feeding line corresponding to the testing machines.

8. The assembly line workpiece inspection system according to claim 6, characterized in that, The workpiece is placed on a carrier assembly line and moves along it. The shifting device includes two opposing lifting carrier plates. The opposite ends of the lifting carrier plates are provided with protruding bottom baffles. The area between the lifting carrier plates is used to accommodate the carrier. The ends of the lifting carrier plates are provided with intercepting baffles. When the carrier is located between the lifting carrier plates, the intercepting baffles prevent the carrier from moving along the feeding line. The bottom baffles support the bottom of the carrier.

9. The assembly line workpiece inspection system according to claim 6, characterized in that, The production line also includes a return line for returning the vehicle to the front end for reuse.

10. The assembly line workpiece inspection system according to claim 6, characterized in that, The unloading line is equipped with an unloading shifting device. The movement height of the unloading shifting device is consistent with that of the shifting device on the loading line and is consistent with the height of the inlet of the testing machine.

Images