A circuit board laser drilling path automatic optimization method

Abstract

The present application relates to the field of data processing, in particular to a kind of circuit board laser drilling path automatic optimization method, for the defect that traditional algorithm only pursues shortest path and ignores thermal accumulation effect, introduces double pheromone ant colony algorithm. By constructing two independent pheromones of path length and heat distribution, and defining hole site neighborhood heat influence factor and path global heat entropy to quantify heat dissipation performance, guide algorithm to consider distance and heat at the same time when selecting path, so as to effectively balance processing efficiency and heat control, avoid algorithm to fall into suboptimal solution due to local heat concentration, finally find the comprehensive optimal path considering path length and heat dissipation effect. The present application balances drilling path length and heat control by adopting double pheromone ant colony algorithm, constructs and independently updates two pheromones of path length and heat distribution, to find optimal solution.

Description

Technical Field

[0001] This invention relates to the field of data processing, and more specifically to an automatic optimization method for laser drilling paths on circuit boards. Background Technology

[0002] In the high-density manufacturing process of printed circuit boards, laser drilling is a key technology, and its efficiency and quality directly affect the performance of the final product. To improve production efficiency, it is usually necessary to optimize the processing sequence of tens of thousands of drilling points. Optimizing the drilling path aims to find a shortest path that traverses all holes to shorten the total travel of the laser head and reduce processing time.

[0003] However, laser drilling involves the instantaneous injection of high energy, generating significant heat at and around the drilling point. If the drilling path planning only aims for the shortest distance, the laser head may operate continuously and densely in a localized area, causing severe heat accumulation. This localized overheating can damage the circuit board substrate material and potentially affect the hole wall quality and electrical performance. Therefore, an ideal drilling path must consider not only the economy of path length but also the uniformity of heat distribution during processing to avoid localized hotspots. Balancing the conflicting goals of shortest path length and heat dispersion constitutes a major challenge in the current technological field.

[0004] For path optimization problems like this, ant colony optimization (ACO) is a widely used heuristic search algorithm. Traditional ACO uses a single pheromone based on path length to guide the search direction, effectively finding shorter paths. However, in laser drilling scenarios, its inherent limitations are quite apparent: because the algorithm's incentive mechanism relies entirely on path length, it inevitably tends to select path segments with short local distances, leading to an overly concentrated drilling sequence in space. This optimization tendency makes the algorithm prone to converging to a suboptimal solution—a shorter path but with a serious risk of local heat accumulation—failing to meet the requirements of high-quality, high-reliability production. Summary of the Invention

[0005] To address the problem that the ant colony algorithm tends to converge to a suboptimal solution with a short path but a serious risk of local heat accumulation, this invention proposes an automatic optimization method for laser drilling paths on circuit boards. The method includes: obtaining the two-dimensional coordinates of all drilling points on the circuit board to be processed; obtaining a neighborhood thermal influence factor for the hole location, which characterizes the risk of local heat concentration during drilling, and whose value is inversely proportional to the Euclidean distance between any two hole locations and directly proportional to the sum of the local densities of the two hole locations; and obtaining the global thermal entropy of the path, which quantifies the heat dissipation performance of the entire drilling path, based on information entropy. The theory calculates the uniformity of the distribution of the thermal influence factor in the neighborhood of the aperture in each step of the path. A dual-pheromone ant colony algorithm is used for path search. The dual pheromones include a path length pheromone to guide the search for short paths and a heat distribution pheromone to guide the search for paths with high heat dissipation performance. After each iteration of the algorithm, the dual pheromones are independently decoupled and updated. The update amount of the path length pheromone is inversely proportional to the total length of the paths generated in that iteration, and the update amount of the heat distribution pheromone is directly proportional to the global thermal entropy of the path. The update process is repeated until the termination condition is met, and the shortest processing path that satisfies the preset minimum global thermal entropy among all iterations is output.

[0006] Existing technologies for planning laser drilling paths on circuit boards typically focus solely on finding the shortest path. While this improves processing efficiency, it overlooks the risk of localized heat buildup due to overly concentrated hole locations, potentially leading to defects such as delamination and carbonization, thus impacting product yield. This invention introduces two metrics—the hole location neighborhood thermal influence factor and the path's global thermal entropy—to evaluate the heat dissipation performance of the drilling path. It employs a dual-pheromone ant colony algorithm to simultaneously optimize path length and heat distribution uniformity. By decoupling the updates of the path length pheromone and heat distribution pheromone, an effective balance is achieved between pursuing high efficiency and ensuring high quality. Ultimately, it can automatically plan an optimal drilling path that guarantees processing efficiency while effectively avoiding the risk of localized overheating, significantly improving the quality and reliability of circuit board processing.

[0007] Furthermore, the specific method for calculating the thermal influence factor of the pore location neighborhood is as follows:

[0008] ;

[0009] in This indicates the thermal influence factor of the pore location neighborhood; and These are respectively represented as hole positions. and hole position Local density; This represents a set distance reference value; Indicates hole position and The Euclidean distance between them; It represents a very small positive real number, used to prevent the denominator from being zero.

[0010] Furthermore, set a radius , to the target hole position - The number of existing pores in the neighborhood is used as its local density.

[0011] Further, the average nearest neighbor distance of all holes on the circuit board is calculated, and the radius is... Set to the average of the nearest neighbor distances times, of which This is a preset constant, and its value ranges from 2 to 5.

[0012] In existing technologies, similar parameters often require manual setting and repeated adjustments, lacking universality. This technology improves performance by adjusting the radius... By correlating the parameter with the average of the nearest neighbor distances of all holes on the circuit board, the parameters can be automatically adjusted according to the density of different circuit board layouts, avoiding the tediousness and uncertainty of manual parameter tuning, and significantly enhancing the algorithm's adaptability and robustness to different drilling tasks.

[0013] Furthermore, the method for calculating the global thermal entropy of the path is as follows:

[0014] ;

[0015] in Represents the global thermal entropy of the path; Indicates the first The normalized proportion of the thermal influence factor of the pore site neighborhood in the total thermal influence factor of all pore sites neighborhoods along the entire path.

[0016] Unlike existing technologies that lack a comprehensive evaluation method for the heat dissipation performance of the entire path, this technology utilizes information entropy theory to define the uniformity of the distribution of local thermal influence factors as a single thermal entropy value. This provides a scientific and unified standard for evaluating and comparing the overall heat dissipation performance of different paths, making the algorithm's optimization objective clearer and effectively guiding the search towards paths with more uniform heat distribution.

[0017] Furthermore, in the ant colony algorithm, the probability model for an ant selecting the next unvisited hole is as follows:

[0018] ;

[0019] in Indicates the probability of selection; Indicates the first In the next iteration, the path Path length and pheromone concentration; Represents the reciprocal of the distance to the heuristic information; Indicates the first In the next iteration, the path The concentration of pheromones distributed on the heat surface; These represent setting three weight parameters respectively; Ants A set of unvisited apertures.

[0020] Compared to traditional ant colony algorithms that only consider distance or a single pheromone, this invention combines path length pheromone, heat distribution pheromone, and the inverse of distance, and adjusts them through weight parameters. This multi-factor decision model makes the path search process more intelligent, enabling flexible trade-offs between path length and heat dissipation performance, thereby more efficiently finding excellent paths that satisfy both optimization objectives in a vast solution space.

[0021] Furthermore, the amount of update for the hot distribution pheromone is also proportional to a scaling factor used to balance the reward levels of the two pheromone types.

[0022] Furthermore, the method for determining the scaling factor includes: generating multiple random drilling paths before the main loop of the algorithm begins; calculating the average path length and average global thermal entropy of the multiple random drilling paths; and using the ratio of 1 to the product of the average path length and the average global thermal entropy as the scaling factor.

[0023] By statistically analyzing the average length and average thermal entropy of random paths before the algorithm begins, the scaling factor can be calculated, avoiding the need for manual setting of this key parameter. This further enhances the automation level of the entire optimization method and its adaptability to different problems, making the algorithm more robust in practical applications.

[0024] Furthermore, the decoupled update of the pheromone also includes a volatilization process, which is achieved by multiplying the current pheromone concentration by a volatilization coefficient in the range of (0,1).

[0025] Furthermore, the termination condition is reaching a preset maximum number of iterations.

[0026] The technical effects of this invention are as follows:

[0027] Traditional methods only pursue the shortest path to improve efficiency, but this invention introduces the concepts of hole location neighborhood thermal influence factor and path global thermal entropy to assess the risk of heat accumulation during processing, and treats these as optimization objectives of equal importance to path length. Through a dual-pheromone ant colony algorithm, a synergistic optimization and intelligent balance between processing efficiency and processing quality is achieved, ultimately obtaining the optimal drilling path that balances speed and reliability, effectively avoiding yield reduction caused by localized overheating. Attached Figure Description

[0028] Figure 1 This is a schematic flowchart illustrating an embodiment of the automatic optimization method for laser drilling paths on a circuit board according to the present invention.

[0029] Figure 2 This is a schematic diagram illustrating the local density distribution of pore sites used to calculate the thermal influence factor in an embodiment of the present invention;

[0030] Figure 3 This is a schematic diagram illustrating the iterative process curves of the two optimization objectives, path length and global thermal entropy, in an embodiment of the present invention.

[0031] Figure 4 This is a schematic diagram illustrating the superposition analysis of the relationship between the final optimized path and the local density of the pore positions, as well as the path step size distribution characteristics in an embodiment of the present invention.

[0032] Figure 5 This is a schematic diagram illustrating the optimal drilling path and its corresponding path thermal influence factor distribution and global thermal entropy results in the final output of an embodiment of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0035] An embodiment of an automatic optimization method for laser drilling paths on circuit boards:

[0036] like Figure 1 As shown, the automatic optimization method for laser drilling path of circuit boards according to the present invention includes:

[0037] S1. Read the two-dimensional coordinates of the drilling points on the circuit board and the physical properties of the material, calculate the Euclidean distance matrix between the holes, and complete the initialization of the dual pheromone matrix for path length and heat distribution optimization.

[0038] In this embodiment, the drilling data of the circuit board to be processed is first obtained, which includes a set of two-dimensional coordinates of all drilling points. ,in This represents the total number of holes. Simultaneously, the physical properties of the circuit board are obtained. These parameters can be directly read from a process database or material specification sheet, and, for example, include the thermal conductivity, specific heat capacity, and density of the board material, as well as the injection energy for a single laser drilling operation.

[0039] Based on the acquired coordinate data, calculate the coordinates of any two borehole points. and Euclidean distance between This is used to construct the initial distance matrix. Understandably, to support the dual-pheromone mechanism of the ant colony algorithm in subsequent steps, two independent distance matrices, each with dimension [missing information], need to be initialized. Pheromones matrix: Path length pheromone matrix and thermal distribution pheromone matrix As a preferred approach, the initial value of all pheromones can be set to a small positive real number. and This ensures that all paths have the potential to be selected in the early stages of the algorithm.

[0040] S2. The neighborhood thermal influence factor is calculated based on the local density of the hole and the hole spacing to assess the local thermal risk, and the information entropy theory is introduced to define the global thermal entropy of the path to quantify the heat dissipation uniformity of the overall processing sequence.

[0041] In the path construction process of the ant colony algorithm, considering only the instantaneous temperature of the next target point is insufficient to fully evaluate the heat accumulation effect, because heat accumulation is a dynamic and regional process. Therefore, in this embodiment, the degree of heat concentration of a path is determined from both local and global levels, and an evaluable global index is calculated.

[0042] First, for any drilled point... After a period of time Then, for any point in space The resulting temperature contribution can be obtained using a simplified thermal diffusion model. In this embodiment, the thermal influence factor of the pore site neighborhood is calculated to characterize the temperature contribution after processing. After the first point, its neighboring points are then processed. This leads to a risk of concentrated heat. Analysis of the scenario shows that this risk is not only related to the distance between the two points. Related to, and also with The density of machined holes around a point is related to the density of holes around that point; the more densely machined holes are around a point, the less easily heat dissipates. Therefore, for each hole... Define its local density The local density By calculating its The number of existing pores in the neighborhood is used to obtain the local density. The specific implementation method for calculating the local density is a well-known technology and will not be described in detail here.

[0043] The above This is a radius set according to the average hole spacing. A preferred embodiment is to set the radius... It is set to be proportional to the average nearest neighbor distance of the circuit board. The specific calculation method is as follows:

[0044] First, examine each hole on the circuit board. Calculate its distance from all other holes, and find the minimum distance among them; this is the hole location. nearest neighbor distance Then, the average nearest neighbor distance of all holes is calculated to obtain the average nearest neighbor distance of the holes on the circuit board. This objectively reflects the overall sparseness or density of the circuit board; finally, the radius... Set to this average value times, that is Among them, multiples It is an empirical constant, preferably, The value ranges from 2 to 5. For example, in a specific embodiment, it can be taken as... .

[0045] like Figure 2 As shown, the two-dimensional coordinates of all 50 drill points on the circuit board are displayed in the form of a scatter plot. The color of each hole is rendered based on its local density value, transitioning from cool to warm tones, with redder colors indicating areas around the hole. The denser the pores in the neighborhood.

[0046] Based on the above analysis, the specific formula for calculating the thermal influence factor of the pore site neighborhood is as follows:

[0047] ;

[0048] in This represents the thermal influence factor in the neighborhood of the pore site, i.e., from the pore site Move to the hole The risk of concentrated heat; and These are respectively represented as hole positions. and hole position The local density, by pre-calculating its -The number of holes in the neighborhood is obtained; This represents a distance reference value used to ensure that the calculated thermal influence factor of the pore location neighborhood is dimensionless. In one embodiment, it can be set to the aforementioned average nearest neighbor distance. ; Indicates hole position and The Euclidean distance between them; Represent a very small positive real number, for example This is used to prevent the denominator from being zero.

[0049] When ants move from a high-density area Move to an area with the same high density At that time, the numerator It will be very large; at the same time, if the distance between these two points is... The distance is still very short, so the denominator will be very small. This collectively leads to a small thermal impact factor. A significant increase indicates that this movement strategy carries an extremely high risk of heat accumulation. Conversely, jumping from a dense region to a sparse region, or even within a dense region but with a long jump distance, indicates a significantly higher risk of heat accumulation. The values ​​will be relatively small, indicating lower risk.

[0050] After a complete path is constructed, it is necessary to evaluate the global heat dissipation performance of the entire path. In this embodiment, the global thermal entropy of the path is defined. The higher the thermal entropy of a path, the more uniform the spatial distribution of its processing sequence and the more dispersed the heat. It utilizes information entropy theory to measure the uniformity of the distribution of thermal influence factors at each step of the path, specifically:

[0051] ;

[0052] ;

[0053] in This represents the global thermal entropy of the path; a larger value indicates a more uniform global heat distribution along the path. Indicates the first path Step, that is, from the first The hole moves to the first Thermal influence factor of the neighborhood of each pore; Indicates the first path Step, that is, from the first The hole moves to the first Thermal influence factor of the neighborhood of each pore; Indicates the first The normalized proportion of the thermal impact factor of a step in the total thermal impact factor of the entire path is the probability of thermal concentration risk occurring in that step.

[0054] According to the above formula, if the thermal influence factor generated at each step of a path... They are all about the same size, meaning the ants alternate between moving at different risks. The value of information entropy will tend to be uniformly distributed, at which point the information entropy... Achieving the maximum value indicates excellent heat dissipation performance of the path. Conversely, if the path contains several steps... If the value is particularly large, meaning drilling occurs continuously within a small area, while the values ​​for other steps are very small, then... The distribution will be very sharp, leading to A very small value indicates poor heat dissipation performance of the path, resulting in severe local heat accumulation.

[0055] S3. By integrating path length pheromones, heuristic information, and heat distribution pheromones, and by setting weight parameters, a dynamic path selection probability model for ants is constructed to comprehensively evaluate path distance and heat dissipation performance.

[0056] When ants choose their next drilling point, their decision is no longer dominated by a single pheromone, but is simultaneously guided by both path length pheromones and heat distribution pheromones. In the... In the next iteration, located at the hole position ants Select the next unvisited hole. probability Defined as:

[0057] ;

[0058] in Indicates the first In the next iteration, the path Path length and pheromone concentration; Heuristic information is usually represented by . This is used to indicate a preference for short distances; Indicates the first In the next iteration, the path The concentration of pheromones distributed on the heat surface; These represent three weight parameters, which respectively control the influence of path length pheromone, heuristic information, and heat distribution pheromone on decision-making; Ants A set of unvisited apertures.

[0059] A segment that guides ants to take the shortest path. It provides a greedy tendency for finding the shortest distance in the immediate local area, while This guides the ants towards segments of a path historically proven to constitute high thermal entropy, i.e., good heat dissipation. As a preferred solution, users can adjust... and The relative size of the values ​​allows for flexible control over the algorithm's emphasis on efficiency and quality. For example, when... When the heat dissipation is good, the algorithm tends to find a path with good heat dissipation performance; otherwise, it focuses more on the path length.

[0060] S4. After each iteration, based on two independent indicators, the total path length and the global path heat entropy, perform independent and decoupled reward and evaporation updates for the path length pheromone and the heat distribution pheromone.

[0061] After all ants have completed one round of path construction, the two pheromones are updated independently and decoupled. First, the path length pheromone is updated. The process involves updating the ant colony algorithm. This is similar to the standard ant colony algorithm, but only considers path length as a metric. The update is based on rewarding ants that construct shorter paths, specifically:

[0062] ;

[0063] ;

[0064] ;

[0065] in Indicates the first In the next iteration, the path Path length and pheromone concentration; The volatility coefficient of the pheromone representing the path length, with a value range of [value missing]. ; This indicates that in this iteration, all ants are on the path The total increase in pheromone levels left on the surface; This indicates the total number of ants; Ants Total length of the constructed path; This represents a constant used to adjust the reward based on path length. In this embodiment, to simplify the model and reduce the number of parameters to be tuned, it can be set to... The preferred value is 1.

[0066] The shorter the path of the ant, the better. The smaller the value, the more pheromones it leaves along its path, thus enhancing the exploration of shorter paths.

[0067] Next, the thermal distribution pheromone was analyzed. An update will be performed. This update does not consider path length, but only the global thermal entropy of the path. The update is based on rewarding ants that construct paths with better heat dissipation performance, specifically:

[0068] ;

[0069] ;

[0070] ;

[0071] in Indicates the first In the next iteration, the path The concentration of pheromones distributed on the heat surface; The volatility coefficient of the heat distribution pheromone is represented, and its value ranges from [value missing]. ; This indicates that in this iteration, all ants are on the path The incremental heat distribution pheromone left on the surface; Ants The global thermal entropy of the constructed path is calculated in step S2; This represents a constant used to adjust the reward size based on the path's global thermal entropy.

[0072] Because path length and global thermal entropy have different dimensions and numerical ranges, if their rewards are not scaled, the update of one of the pheromones may become almost negligible. As a normalization scaling factor, it stretches or compresses the thermal entropy reward to the same order of magnitude as the length reward, therefore It is dynamically adaptive to ensure a balance between the two types of rewards.

[0073] A preferred implementation involves a preprocessing step before the main loop of the algorithm begins. This involves generating a suitable number of completely random drilling path samples. For example, generating 100 random paths. For each of these 100 random paths, its total path length is calculated. and path global thermal entropy Then calculate the average path length of these 100 samples. and average global thermal entropy Next, the average of the inherent scales of these two randomly sampled instances is used to calculate a unique, fixed... .because And in this embodiment, it is set At this point, we have: However, due to the final average path length There is a unit of length, therefore the final value is set. The calculation method is as follows: ;in This represents a reference length value, which can be set to 1mm in this embodiment, for use in outputting... The value is dimensionless; the implementer needs to determine the unit of this reference distance based on the actual path distance, and finally use this calculated value. The value is substituted into the entire iteration process of the ant colony algorithm.

[0074] The logic behind this update rule is: the global thermal entropy of the path. Larger ants, meaning those with better path heat dissipation, leave more heat pheromones along their paths. This incentivizes subsequent ants to explore segments that form high thermal entropy paths, even if these segments aren't locally shortest. Through this updating mechanism, even if a path contains long jumps, it can still achieve its goals. Larger, but as long as its If it is high enough, it can still be found in the thermal distribution pheromone matrix. It retains more pheromones, thus avoiding the fate of being prematurely eliminated by the traditional single pheromone mechanism.

[0075] S5. Repeat the iterative process of path construction and pheromone update until the termination condition is met, and finally output the comprehensive optimal drilling path scheme with the shortest path length under the preset thermal entropy threshold.

[0076] Repeat steps S3 and S4 until the preset maximum number of iterations is reached.

[0077] like Figure 3 As shown, the graph contains two subgraphs recording the performance changes of the algorithm over 100 iterations. The upper graph shows the convergence curves of the optimal path length and the average path length. The solid line represents the length of the optimal solution in each generation, and the dashed line represents the average solution length of all ants in the current generation. The lower graph shows the evolution curves of the optimal path heat entropy and the average path heat entropy. The solid line represents the heat entropy of the optimal solution in each generation, and the dashed line represents the heat entropy of the average solution of all ants in the current generation. The blue solid line in the upper graph steadily decreases and eventually converges, indicating that the algorithm successfully found a shorter path. The red solid line in the lower graph steadily increases amidst fluctuations, indicating that the algorithm is also continuously searching for a path with a more uniform heat distribution.

[0078] When the algorithm terminates, it outputs the path with the best overall evaluation across all iterations as the final laser drilling path scheme. For example, the optimal path can be one that satisfies a preset minimum global thermal entropy. Given the threshold, path length The shortest solution. For example... Figure 4 and Figure 5 As shown:

[0079] in Figure 4 The left image shows the optimized blue path line superimposed on the local density map of the boreholes; Figure 4 The right figure shows the distribution of the distance between adjacent holes in the optimal path in the form of a histogram, and the average step length is marked with a red dashed line.

[0080] Figure 5 The left figure shows the optimal borehole path of the final output, and marks the start point, end point and total path length; Figure 5 The right figure shows the sequential distribution of the thermal impact factor magnitude at each step of the path, along with the global thermal entropy value of the entire path. Except for a few initial steps with higher thermal impact, the thermal impact factors for the vast majority of subsequent steps remain at very low and uniform levels. This uniform risk distribution reflects high information entropy, demonstrating the path's heat dissipation performance.

[0081] Compared with existing technologies, the method proposed in this invention fundamentally solves the inherent conflict between the two objectives of shortest path and heat distribution in traditional ant colony algorithms by establishing two independent evaluation dimensions—path length pheromone and heat distribution pheromone—and decoupling and updating them. This allows the algorithm to retain and develop solutions with excellent heat dissipation performance but slightly longer paths during the search process, effectively preventing the algorithm from prematurely converging to suboptimal solutions that are only locally optimal in terms of path length but have poor heat control. This significantly improves the global search capability and the overall quality of the final path solution.

Claims

1. A method for automatically optimizing laser drilling paths on circuit boards, characterized in that, The method includes: obtaining the two-dimensional coordinates of all drilling points on the circuit board to be processed; Obtain a borehole neighborhood thermal influence factor to characterize the risk of local heat concentration during drilling. Its value is inversely proportional to the Euclidean distance between any two boreholes and directly proportional to the sum of the local densities of the two boreholes. The path global thermal entropy used to quantify the heat dissipation performance of the entire drilling path is obtained. Its value is calculated based on the information entropy theory to determine the uniformity of the distribution of the thermal influence factor of the borehole location neighborhood at each step in the path. A dual-pheromone ant colony algorithm is used for path search. The dual pheromones include a path length pheromone for guiding the search for short paths and a heat distribution pheromone for guiding the search for paths with high heat dissipation performance. After each iteration of the algorithm, the dual pheromones are independently decoupled and updated. The update amount of the path length pheromone is inversely proportional to the total length of the paths generated in that iteration, and the update amount of the heat distribution pheromone is directly proportional to the global thermal entropy of the path. Repeat the update process until the termination condition is met, and output the shortest processing path that satisfies the preset minimum global thermal entropy among all iterations.

2. The automatic optimization method for laser drilling path on a circuit board according to claim 1, characterized in that, The specific method for calculating the thermal influence factor of the pore site's neighborhood is as follows: ； in This indicates the thermal influence factor of the pore location neighborhood; and These are respectively represented as hole positions. and hole position Local density; This represents a set distance reference value; Indicates hole position and The Euclidean distance between them; It represents a very small positive real number, used to prevent the denominator from being zero.

3. The automatic optimization method for laser drilling paths on circuit boards according to claim 2, characterized in that, Set a radius , to the target hole position - The number of existing pores in the neighborhood is used as its local density.

4. The automatic optimization method for laser drilling path on a circuit board according to claim 3, characterized in that, Calculate the average nearest neighbor distance for all holes on the circuit board, and then use the radius... Set to the average of the nearest neighbor distances times, of which This is a preset constant, and its value ranges from 2 to 5.

5. The automatic optimization method for laser drilling path on a circuit board according to claim 1, characterized in that, The specific method for calculating the global thermal entropy of the path is as follows: ； in Represents the global thermal entropy of the path; Total number of holes; Indicates the first The normalized proportion of the thermal influence factor of the pore site neighborhood in the total thermal influence factor of all pore sites neighborhoods along the entire path.

6. The automatic optimization method for laser drilling path on a circuit board according to claim 1, characterized in that, In the ant colony algorithm, the probability model for an ant choosing the next unvisited hole is as follows: ； in Indicates the probability of selection; Indicates the first In the next iteration, the path Path length and pheromone concentration; Represents the reciprocal of the distance to the heuristic information; Indicates the first In the next iteration, the path The concentration of pheromones distributed on the heat surface; These represent setting three weight parameters respectively; Ants A set of unvisited apertures.

7. The automatic optimization method for laser drilling path on a circuit board according to claim 1, characterized in that, The amount of update to the thermally distributed pheromone is also proportional to a scaling factor used to balance the reward levels of the two pheromones.

8. The automatic optimization method for laser drilling path on a circuit board according to claim 7, characterized in that, The method for determining the scaling factor includes: Before the main loop of the algorithm begins, multiple random drilling paths are generated; Calculate the average path length and average global thermal entropy of the multiple random borehole paths; The scaling factor is the ratio of 1 to the product of the average path length and the average global thermal entropy.

9. The automatic optimization method for laser drilling path of circuit boards according to claim 1, characterized in that, The decoupled update of the pheromone also includes a volatilization process, which is achieved by multiplying the current pheromone concentration by a volatilization coefficient in the range of (0,1).

10. The automatic optimization method for laser drilling path of circuit boards according to claim 1, characterized in that, The termination condition is reaching the preset maximum number of iterations.

Images