3D printing concrete path planning method

Abstract

The invention discloses a 3D printing concrete path planning method, and particularly relates to the technical field of 3D printing concrete paths, and the method comprises the following steps: 1, model preprocessing: carrying out meshing processing on an input 3D model, converting the input 3D model into a triangular mesh model with proper precision, carrying out topology analysis on the mesh model, and carrying out modeling; identifying the boundary, the internal structure and different regional geometric features of the model, and providing basic data for path planning; 2, generating an initial path, discretizing the model into a plurality of layer sheets along the height direction according to the height information of the model, determining the printing sequence of each layer sheet, and generating the initial path for each layer sheet; 3, optimizing an interlayer path, and establishing an interlayer connection model; by using an energy-optimized interlayer path optimization method, the interlayer bonding energy and the printing path smoothing energy are comprehensively considered, so that the paths of adjacent layers reach the optimal connection state, the interlayer bonding strength is enhanced, and the stress concentration is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of 3D printing concrete path, more particularly, the present application relates to a 3D printing concrete path planning method. BACKGROUND

[0002] In the field of construction, 3D printing concrete technology is gradually emerging, which can realize the integrated forming of complex building components. At present, the common 3D printing concrete path planning method is mainly based on the traditional layer-by-layer accumulation method, and the printing head extrudes concrete layer by layer according to the predetermined plane path. The path of each layer is usually a simple straight line or curve filling, and the paths between adjacent layers are relatively independent, without fully considering the influence of layer connection on printing quality and efficiency. Another method is to generate a path based on model slicing, which slices the 3D model along the height direction, and then plans a path for each slice layer. However, this method is prone to problems such as discontinuous and uneven paths when dealing with complex geometrical shapes and internal structures.

[0003] In the invention patent with publication number CN116432452A, a 3D printing concrete path planning method based on Hamilton path is disclosed. By reasonably planning the starting point and ending point of the printing head, the printing path is optimized, the number of start-stop of the printing head in the printing process is reduced, the problem of repeated passing through the same vertex by the Euler circuit printing head is avoided, the continuity of the printing process is improved, and the inflection point in the printing process is reduced. However, when layer-by-layer accumulation path is used, the lack of correlation between adjacent layers leads to low interlayer bonding strength, which easily causes delamination phenomenon in the printing process, affecting the integrity and mechanical properties of the printed part. SUMMARY

[0004] In order to overcome the defects that the lack of correlation between adjacent layers leads to low interlayer bonding strength, which easily causes delamination phenomenon in the printing process, affecting the integrity and mechanical properties of the printed part when layer-by-layer accumulation path is used, the embodiments of the present application provide a 3D printing concrete path planning method to solve the problems raised in the background art.

[0005] To achieve the above purpose, the present application provides the following technical scheme:

[0006] Comprising the following steps:

[0007] Step one: model preprocessing, the input 3D model is subjected to meshing treatment, which is converted into a triangular mesh model with appropriate accuracy, and the mesh model is subjected to topological analysis to identify the boundary, internal structure and different regional geometric features of the model, providing basic data for path planning;

[0008] Step two: initial path generation, discretize the model into multiple slices along the height direction according to the height information of the model, determine the printing order of each slice, and generate the initial path for each slice;

[0009] Step three: inter-slice path optimization, establish an inter-slice connection model based on the initial path, analyze the connection relationship between the paths of adjacent slices, and adjust and optimize the inter-slice path;

[0010] Step four: local path optimization and correction, during the printing process, local path correction is performed on the path deviation or defects caused by the complexity of the model and environmental factors;

[0011] Step five: parameter self-adaptive adjustment, automatically adjust the related parameters of path planning.

[0012] In a preferred embodiment, the step two considers the rheological properties of concrete and the kinematic parameters of the printing head when generating the initial path, performs preliminary optimization on the path, and generates the initial path using an improved filling algorithm.

[0013] In a preferred embodiment, the improved filling algorithm is used for curvature-driven path spacing adjustment and primary curvature direction pre-growth. The curvature-driven path spacing adjustment is used to reduce the spacing in high-curvature areas to improve surface quality and increase the spacing in low-curvature areas to improve efficiency. The primary curvature direction pre-growth is used to expand the path along the maximum principal curvature direction to reduce the number of printing head turns.

[0014] In a preferred embodiment, the mathematical model and parameter definition of the improved filling algorithm are as follows:

[0015] Input parameters:

[0016] The vertex set V = {v i} and the face set F = {f j} of the grid model;

[0017] Initial path spacing d0, curvature sensitivity coefficient α, and growth step ΔL;

[0018] Curvature threshold κ th (distinguishing high / low curvature areas);

[0019] Path generation steps:

[0020] (1) Curvature calculation: for vertex v i , calculate the principal curvature κ(v i ) = max{|λ1|, |λ2|}, where λ1, λ2 are the eigenvalues of the Weingarten matrix;

[0021] (2) Dynamic spacing adjustment:

[0022] (3) Region growing: initialize seed point set S as model contour edge;

[0023] Iterative expansion: for seed point s∈S, extend along principal curvature direction e max (s) generate new path;

[0024] Point s' = s + ΔL·e max (s);

[0025] Update seed set S←S∪{s′}, until cover the entire region.

[0026] In a preferred embodiment, step three uses an energy optimization method to calculate the optimization function of interlayer bonding performance and printing path smoothing performance, so that the paths of adjacent layers reach the best connection state in starting point, ending point and direction.

[0027] In a preferred embodiment, the interlayer bonding performance includes the bonding performance of concrete at the interlayer interface, which is related to the material properties of concrete, extrusion speed, path spacing and other factors, and the interlayer path optimization expression is:

[0028]

[0029] Wherein:

[0030] f bond is the interlayer bonding performance function;

[0031] V extrusion is the extrusion speed (mm / s);

[0032] s is the interlayer path spacing (mm);

[0033] t cure is the concrete curing time (s);

[0034] β is the material curing coefficient;

[0035] ω1, ω2 are weight coefficients, it should be noted that ω1, ω2 are set according to actual conditions, for example, expert weighting method is adopted, that is, experts in the relevant field are invited to determine the weight coefficients of various parameters through professional opinion investigation and comprehensive evaluation.

[0036] In a preferred embodiment, the optimization function of printing path smoothing performance is used to measure the smoothness of the path, avoid sharp turns or breakpoints in the path, and ensure the smoothness of the printing head movement, and the printing path smoothing performance expression is:

[0037]

[0038] Wherein:

[0039] f smooth is a path smoothing function;

[0040] k(l) is a path curvature;

[0041] is a path direction angle; i is a path direction angle;

[0042] is a direction change penalty coefficient;

[0043] is a total optimization objective function F total is expressed as:

[0044] F total = λ · f bond + (1 - λ) · f smooth ;

[0045] By adjusting λ, the bonding strength and smoothness are balanced;

[0046] In the inter-layer path optimization process, real-time concrete hardening monitoring data is also combined, and the inter-layer path connection mode and speed parameters are dynamically adjusted according to the hardening degree of the concrete, so as to ensure that the concrete between different layers can be fully bonded and fused.

[0047] In a preferred embodiment, the hardening program dynamic adjustment logic is as follows:

[0048] Hardening monitoring: The concrete temperature T(t) is measured in real time by the thermocouple embedded in the print head, and the solidification degree c(t) = 1 - e -k·T(t)·t , where k is the reaction rate constant;

[0049] Adjustment strategy:

[0050] Path connection mode: when c(t) < 0.5, staggered overlapping connection (overlapping length L overlap = 0.2d0) is adopted;

[0051] When c(t) ≥ 0.5, switch to end-point alignment connection;

[0052] Speed parameter: the printing speed v is inversely proportional to the solidification degree c(t):

[0053] In a preferred embodiment, the step four uses a visual sensor installed on the printing device to take real-time photos and collect images of the printing result after the print head completes printing of a path segment, processes and analyzes the collected images, uses OpenCV edge detection in the image processing algorithm, extracts the geometric information of the actual printing path, and compares it with the pre-planned path. According to the comparison result, an error feedback control algorithm is used to locally correct and optimize the subsequent printing path;

[0054] Error feedback control algorithm logic as follows:

[0055] Image acquisition: industrial camera acquires print cross-section images at ≥ 30 FPS;

[0056] Error detection: actual path profile is extracted using Canny edge detection;

[0057] Calculate path centerline deviation: e(t) = ‖x stress (t) - x adapt (t)‖;

[0058] Feedback correction: PID controller is used to calculate the correction amount: Update next path segment parameters: x dotps (t+Δt) + u(t)·Δt;

[0059] Delay compensation: through feedforward control to estimate the system response within Δt = 50 ms, adjust the print head position in advance.

[0060] In a preferred embodiment, the parameter adaptive adjustment in step five includes a multi-objective optimization model and dynamic adjustment logic;

[0061] Multi-objective optimization model logic as follows:

[0062] Input variables:

[0063] (a) Model complexity C (number of vertices / number of patches);

[0064] (b) Concrete rheological parameters η (viscosity), τ (yield stress);

[0065] (c) Equipment performance P (maximum acceleration, extrusion pressure range);

[0066] Output parameters:

[0067] (d) Filling density ρ ∈ [0.1, 0.9];

[0068] (e) Path spacing s ∈ [1, 10] mm;

[0069] (f) Print speed v ∈ [50, 300] mm / s;

[0070] (g) Extrusion pressure p ∈ [0.1, 1.0] MPa;

[0071] Fitness function:

[0072] Where:

[0073] T max is the printing time;

[0074] σ stress Stress concentration factor (estimated by finite element analysis);

[0075] ΔV is the volume error (the difference between the target volume and the actual extrusion volume);

[0076] Process optimization:

[0077] S1, initialize population: randomly generate N sets of parameter combinations;

[0078] S2, evaluate fitness: calculate F for each set of parameters; adapt ;

[0079] S3, selection / crossover / variation: retain high-quality individuals, generate a new population through simulated binary crossover (SBX) and polynomial mutation;

[0080] S4, iterate to convergence or reach the maximum number of iterations;

[0081] The dynamic adjustment logic is as follows:

[0082] High complexity model (C>105): reduce p, v, and increase s to avoid device overload;

[0083] High viscosity concrete (η>104Pa\c dotps ): increase p and reduce v to ensure continuous extrusion;

[0084] Device performance limitations (such as maximum acceleration a max ): limit v and a to prevent motion from falling out of step;

[0085] The parameter self-adaptive adjustment process adopts a multi-objective optimization strategy based on genetic algorithm, which finds the optimal parameter combination under the premise of ensuring printing quality and efficiency, so that the path planning algorithm can perform best under different conditions.

[0086] Technical effects and advantages of the present application:

[0087] 1. In the existing device, the lack of correlation between adjacent layers of paths leads to low interlayer bonding strength, and the phenomenon of layering easily occurs during printing, affecting the overall integrity and mechanical properties of the printed part. The present application uses an energy-optimized interlayer path optimization method, which considers the interlayer bonding energy and the smoothness of the printing path, so that the paths of adjacent layers reach the best connection state, enhancing the interlayer bonding strength and reducing stress concentration, effectively reducing the layering phenomenon and improving the overall integrity and mechanical properties of the printed part.

[0088] 2、The path planning method based on model slice of the existing equipment has low path planning efficiency when facing complex shapes, and the generated path is easy to cause uneven extrusion of concrete, causing local defects, affecting the surface quality and dimensional accuracy of the printed part. The improved filling algorithm combined with curve offset and region growing strategy can generate an initial path that fits the geometric features of the model, reducing path intersection and overlap, improving the efficiency and quality of path planning. And using the local path correction strategy of visual feedback, combined with error feedback control algorithm, real-time correction of path deviation and defects in the printing process is realized, ensuring the dimensional accuracy and geometric shape of the printed part, to avoid the occurrence of local defects. And the parameter adaptive adjustment step adopts a multi-objective optimization strategy based on genetic algorithm, which can automatically adjust the path planning parameters according to real-time information, adapt to different printing tasks and material characteristics, and improve the universality and adaptability of the algorithm.

[0089] 3、In terms of path planning efficiency, the improved filling combined with the parameter adaptive adjustment step shortens the path planning time, enabling rapid response to different model and task requirements, and improving the production efficiency of 3D printed concrete. At the same time, due to the application of path optimization and local correction strategy, the surface quality and dimensional accuracy of the printed part are also significantly improved, the surface roughness is reduced, and the high-quality printing requirements of complex concrete structures are met. BRIEF DESCRIPTION OF DRAWINGS

[0090] In order to facilitate the understanding of those skilled in the art, the present application will be further described below with reference to the accompanying drawings;

[0091] Fig. 1 The flowchart of the present application;

[0092] Fig. 2 The interlayer path optimization flowchart of the present application;

[0093] Fig. 3 The parameter adaptive adjustment flowchart of the present application. DETAILED DESCRIPTION

[0094] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0095] Embodiment: Figs. 1-3 A 3D printed concrete path planning method is given, comprising the following steps:

[0096] Step one: model preprocessing, the input 3D model is meshed to convert it into a triangular mesh model with appropriate accuracy, and topological analysis is performed on the mesh model to identify the model's boundaries, internal structure, and different regional geometric features such as curvature, height, direction, etc., providing basic data for path planning;

[0097] Step two: initial path generation, according to the height information of the model, the model is discretized into multiple slices along the height direction, and the printing order of each slice is determined. For each slice, an initial path is generated;

[0098] Step three: inter-layer path optimization, based on the initial path, an inter-layer connection model is established, and the connection relationship between adjacent slices is analyzed, including the relative position of the path start and end points, the continuity of the path direction, etc., and the inter-layer path is adjusted and optimized;

[0099] Step four: local path optimization and correction, during the printing process, local path correction is performed for path deviations or defects caused by the complexity of the model and environmental factors;

[0100] Step five: parameter adaptive adjustment, according to the model complexity, concrete rheological parameters, printing device performance, etc., real-time information, automatically adjust the relevant parameters of path planning, such as filling density, path spacing, printing speed, extrusion pressure, etc.

[0101] Step two, when generating the initial path, the rheological properties of the concrete and the kinematic parameters of the printing head are considered simultaneously to preliminarily optimize the path, ensuring that the concrete is evenly distributed during extrusion and avoiding local accumulation or material shortage. An improved filling algorithm is used to generate the initial path. This filling algorithm combines curve offset and region growing strategies to ensure path integrity while automatically adjusting path spacing and direction based on the model's geometric features, making the path better fit the model's shape and reducing path intersection and overlap.

[0102] The improved filling algorithm is used for curvature-driven path spacing adjustment and principal curvature direction primary growth. Curvature-driven path spacing adjustment is used to reduce spacing in high-curvature areas to improve surface quality and increase spacing in low-curvature areas to improve efficiency. Principal curvature direction primary growth is used to extend the path along the maximum principal curvature direction to reduce the number of printing head turns.

[0103] Improved filling algorithm description:

[0104] Traditional parallel scanning or contour offset algorithms are prone to path intersection and uneven spacing in complex curved surface areas. This invention proposes a filling algorithm that combines dynamic curvature perception and adaptive region growth. The core improvements include:

[0105] (1) Curvature-driven path spacing adjustment: reduce spacing in high-curvature regions to improve surface quality, increase spacing in low-curvature regions to improve efficiency.

[0106] (2) Principal curvature direction optimization: extend paths along the direction of maximum principal curvature, reducing the number of print head turns.

[0107] Improved filling algorithm mathematical model and parameter definition:

[0108] Input parameters:

[0109] Grid model vertex set V = {v i}, face set F = {f j};

[0110] Initial path spacing d0, curvature sensitivity coefficient α, growth step ΔL;

[0111] Curvature threshold k th (differentiate high / low curvature regions);

[0112] Path generation steps:

[0113] (1) Curvature calculation: for vertex v i , calculate the principal curvature κ(v i ) = max{|λ1|, |λ2|}, where λ1, λ2 are the eigenvalues of the Weingarten matrix;

[0114] (2) Dynamic spacing adjustment:

[0115] (3) Region growth: initialize seed point set S as the model contour edge;

[0116] Iterative expansion: for seed point s ∈ S, generate a new path along the principal curvature direction e max (s);

[0117] Point s' = s + ΔL·e max (s);

[0118] Update seed set S ← S ∪ {s′}, until covering the entire region.

[0119] Algorithm advantages:

[0120] (1) Adaptive spacing: adjust spacing by curvature weighting, reduce material accumulation risk in high-curvature regions.

[0121] (2) Direction continuity: grow paths along the principal curvature direction, reduce print head sudden stop and start frequency, improve motion stability.

[0122] The third step is to calculate the optimization function of the interlayer bonding performance and the printing path smoothing performance by using the energy optimization method, so that the paths of adjacent layers are optimally connected in the starting point, the ending point and the direction. The energy optimization method is used to adjust and optimize the interlayer path. By introducing the optimization function of the interlayer bonding performance and the printing path smoothing performance, the paths of adjacent layers are optimally connected in the starting point, the ending point and the direction, thereby enhancing the interlayer bonding strength and reducing the stress concentration phenomenon in the printing process. The interlayer bonding performance mainly considers the bonding performance of concrete at the interlayer interface, which is related to the material properties of concrete, the extrusion speed, the path spacing and other factors; the printing path smoothing performance is used to measure the smoothness of the path, so as to avoid sharp turning or breakpoints of the path and ensure the stability of the movement of the printing head.

[0123] The interlayer bonding performance includes the bonding performance of concrete at the interlayer interface, which is related to the material properties of concrete, the extrusion speed, the path spacing and other factors, and the interlayer path optimization expression is:

[0124]

[0125] wherein:

[0126] f bond is the interlayer bonding performance function;

[0127] V extrusion is the extrusion speed (mm / s);

[0128] s is the interlayer path spacing (mm);

[0129] t cure is the concrete solidification time (s);

[0130] β is the material solidification coefficient;

[0131] ω1 and ω2 are weight coefficients, and it should be noted that ω1 and ω2 are set according to the actual situation, for example, the expert weighting method is used, that is, the experts in the relevant field are invited to determine the weight coefficients of various parameters through professional opinion investigation and comprehensive evaluation.

[0132] The optimization function of the printing path smoothing performance is used to measure the smoothness of the path, so as to avoid sharp turning or breakpoints of the path and ensure the stability of the movement of the printing head, and the printing path smoothing performance expression is:

[0133]

[0134] wherein:

[0135] f smooth is the path smoothing function;

[0136] k(l) is the path curvature;

[0137] θ i is the i-th segment path direction angle;

[0138] ω3 is the direction change penalty coefficient;

[0139] Total optimization objective F total The expression is:

[0140] F total = λ · f bond + (1 - λ) · f smooth ;

[0141] Balancing the bonding strength and smoothness by adjusting λ;

[0142] In the inter-layer path optimization process, real-time concrete hardening monitoring data is also combined, and the inter-layer path connection mode and speed parameters are dynamically adjusted according to the hardening degree of the concrete, so as to ensure that the concrete between different layers can be fully bonded and fused.

[0143] The hardening program dynamic adjustment logic is as follows:

[0144] Hardening monitoring: The thermocouple embedded in the print head measures the concrete temperature T(t) in real time, and calculates the solidification degree c(t)=1-e -k·T(t)·t , where k is the reaction rate constant;

[0145] Adjustment strategy:

[0146] Path connection mode: when c(t)<0.5, staggered overlapping connection (overlapping length L overlap =0.2d0) is adopted;

[0147] When c(t)≥0.5, switch to end-point alignment connection;

[0148] Speed parameter: the printing speed v is inversely proportional to the solidification degree c(t):

[0149] Step four, after the print head completes the printing of each path, a visual sensor installed on the printing device is used to take real-time photos and collect images of the printing results, the collected images are processed and analyzed, the image processing algorithm uses OpenCV edge detection to extract the geometric information of the actual printing path, such as position, width, height, etc., and compares it with the pre-planned path, according to the comparison result, uses error feedback control algorithm to make local correction and optimization on the subsequent printing path, to ensure that the size accuracy and geometric shape of the printed part meet the design requirements, and the feedback delay is ≤50ms;

[0150] The error feedback control algorithm logic is as follows:

[0151] Image acquisition: industrial camera acquires print cross-section images at ≥ 30 FPS;

[0152] Error detection: actual path profile is extracted using Canny edge detection;

[0153] Compute path centerline deviation: e(t) = ‖x target (t) - x actual (t)‖;

[0154] Feedback correction: PID controller is adopted to compute correction amount: Update next path segment parameters: x next (t+Δt) + u(t)·Δt;

[0155] Delay compensation: feedforward control is used to predict system response within Δt = 50 ms and adjust printhead position in advance.

[0156] Parameter self-adaptive adjustment in step five includes multi-objective optimization model and dynamic adjustment logic;

[0157] Multi-objective optimization model logic is as follows:

[0158] Input variables:

[0159] (a) model complexity C (number of vertices / number of facets);

[0160] (b) concrete rheological parameters η (viscosity), τ (yield stress);

[0161] (c) equipment performance P (maximum acceleration, extrusion pressure range);

[0162] Output parameters:

[0163] (d) fill density ρ ∈ [0.1, 0.9];

[0164] (e) path spacing s ∈ [1, 10] mm;

[0165] (f) print speed v ∈ [50, 300] mm / s;

[0166] (g) extrusion pressure p ∈ [0.1, 1.0] MPa;

[0167] Fitness function:

[0168] Where:

[0169] T print is the printing time;

[0170] σ stress is the stress concentration factor (estimated by finite element analysis);

[0171] AV is the volume error (difference between target volume and actual extrusion volume);

[0172] Flow optimization:

[0173] S1, initialize population: randomly generate N sets of parameter combinations;

[0174] S2, evaluate fitness: calculate F for each set of parameters adapt ;

[0175] S3, selection / crossover / mutation: retain high-quality individuals, generate a new population through simulated binary crossover (SBX) and polynomial mutation;

[0176] S4, iterate to convergence or reach the maximum number of iterations;

[0177] The dynamic adjustment logic is as follows:

[0178] High complexity model (C>105): reduce p, v, and increase s to avoid device overload;

[0179] High viscosity concrete (η>104Pa\c dotps ): increase p and reduce v to ensure continuous extrusion;

[0180] Device performance limitations (such as maximum acceleration a max ): limit v and a to prevent motion from falling out of step;

[0181] The parameter self-adaptive adjustment process adopts a multi-objective optimization strategy based on genetic algorithms, which finds the optimal parameter combination under the premise of ensuring printing quality and efficiency, so that the path planning algorithm can perform optimally under different conditions.

[0182] The above formulas are dimensionless and their numerical values are calculated. The formulas are obtained by collecting a large amount of data to simulate the most recent real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0183] It should be understood that the size of the serial number of each process described above in various embodiments of the present application does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0184] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A 3D printed concrete path planning method, characterized by: The method comprises the following steps: Step 1: model preprocessing, the input 3D model is meshed to convert it into a triangular mesh model with appropriate accuracy, and topological analysis is performed on the mesh model to identify the boundaries, internal structure and different regional geometric features of the model, providing basic data for path planning; Step 2: initial path generation, the model is discretized into multiple slices along the height direction according to the height information of the model, and the printing order of each slice is determined, and an initial path is generated for each slice; Step 3: inter-layer path optimization, an inter-layer connection model is established according to the initial path, the connection relationship between adjacent slices is analyzed, and the inter-layer path is adjusted and optimized; Step 4: local path optimization and correction, local path correction is performed on the path deviation or defects caused by the complexity of the model and environmental factors during printing; Step 5: adaptive adjustment of parameters, automatically adjusting the related parameters of path planning.

2. The 3D printed concrete path planning method of claim 1, wherein: The step 2 considers the rheological properties of concrete and the kinematic parameters of the printing head when generating the initial path, and preliminarily optimizes the path by using an improved filling algorithm to generate the initial path.

3. The 3D printed concrete path planning method of claim 2, wherein: The improved filling algorithm is used for curvature-driven path spacing adjustment and primary curvature direction optimization, the curvature-driven path spacing adjustment is used to reduce the spacing in high-curvature areas to improve surface quality, and increase the spacing in low-curvature areas to improve efficiency, and the primary curvature direction optimization is used to expand the path along the maximum principal curvature direction to reduce the number of printing head turns.

4. The 3D printed concrete path planning method of claim 3, wherein: The mathematical model and parameter definition of the improved filling algorithm are as follows: Input parameters: A mesh model vertex set V = {v i}, a face sheet set F = {f j}. Initial path spacing d0, curvature sensitivity coefficient α, growth step ΔL; Curvature threshold K th ; Path generation steps: (1) Curvature computation: For a vertex v i , compute the principal curvature κ(v i ) = max{|λ1|, |λ2|}, where λ1, λ2 are eigenvalues of the Weingarten matrix; (2) Dynamic pitch adjustment: (3) Region growth: initialize the seed point set S as the contour edge of the model; Iterative expansion: for a seed point s ∈ S, expand along the principal curvature direction e max (s) generating a new path; Point s' = s + AL e max (s); Update the seed set S←S∪{s′}, until the entire region is covered.

5. The 3D printed concrete path planning method of claim 1, wherein: The step 3 uses an energy optimization method to calculate the optimization function of the inter-layer bonding performance and the printing path smoothness, so that the paths of adjacent slices reach the best connection state in terms of starting point, ending point and direction.

6. The 3D printed concrete path planning method of claim 5, wherein: The inter-layer bonding performance includes the bonding performance of concrete at the inter-layer interface, and the inter-layer path optimization expression is: Where: f bond is the interlayer adhesion function; V extrusion for extrusion speed; s is the inter-layer path spacing; t cure t is the solidification time of the concrete; β is the material solidification coefficient; ω1, ω2 are weight coefficients.

7. The 3D printed concrete path planning method of claim 6, wherein: The optimization function of the printing path smoothness is used to measure the smoothness of the path to avoid sharp turns or breakpoints in the path, and to ensure the smoothness of the printing head movement, and the printing path smoothness expression is: Where: f smooth is a path smoothing function; k(l) is the path curvature; θ i is the i-th path direction angle; ω3 is the direction change penalty coefficient; Total optimization objective F total The expression is: F total = λ · f bond + (1 - λ) · f smooth ; The bonding strength and smoothness are balanced by adjusting λ; In the inter-layer path optimization process, real-time concrete hardening monitoring data is also combined to dynamically adjust the inter-layer path connection mode and speed parameters according to the hardening degree of the concrete.

8. The 3D printed concrete path planning method of claim 7, wherein: The dynamic adjustment logic of the hardening program is as follows: Curing monitoring: The concrete temperature T(t) is measured in real time by thermocouples embedded in the print head, and the degree of curing c(t) = 1 - e -k·T(t)·t where k is the reaction rate constant; Adjustment strategy: Path connection mode: when c(t) < 0.5, staggered overlap connection is adopted, and the overlap length L overlap = 0.2d0; When c(t)≥0.5, switch to endpoint alignment connection; Speed parameter: the printing speed v is inversely proportional to the solidification degree c(t):

9. The 3D printed concrete path planning method of claim 1, wherein: The step four uses a visual sensor installed on the printing device to take real-time photos and collect images of the printing result after the printing head completes a path of printing, processes and analyzes the collected images, uses OpenCV edge detection in the image processing algorithm to extract the geometric information of the actual printing path, and compares the geometric information with the pre-planned path, according to the comparison result, uses an error feedback control algorithm to locally correct and optimize the subsequent printing path; The error feedback control algorithm logic is as follows: Image collection: an industrial camera collects printing cross-section images at ≥ 30 FPS; Error detection: use Canny edge detection to extract the actual path contour; Compute path centerline deviation: e(t) = ||x target (t) - x actual (t)‖; Feedback correction: The PID controller is used to calculate the correction amount: Update next path segment parameters: x next (t + Δt) + u(t) · Δt; Delay compensation: estimate the system response within Δt = 50 ms through feedforward control, and adjust the printing head position in advance.

10. The 3D printed concrete path planning method of claim 1, wherein: The parameter adaptive adjustment in the step five includes a multi-objective optimization model and a dynamic adjustment logic; The multi-objective optimization model logic is as follows: Input variables: (a) model complexity C; (b) concrete rheological parameters η, τ; (c) device performance P; Output parameters: (d) filling density ρ ∈ [0.1, 0.9]; (e) path spacing s ∈ [1, 10] mm; (f) printing speed v ∈ [50, 300] mm / s; (g) extrusion pressure p ∈ [0.1, 1.0] MPa; Fitness function: Wherein: T print is the printing time; σ stress σ is the stress concentration factor; ΔV is the volume error; Process optimization: S1, initialize population: randomly generate N sets of parameter combinations; S2, evaluate fitness: compute F for each set of parameters adapt ; S3, selection / crossover / variation: retain high-quality individuals, generate a new population through simulated binary crossover and polynomial mutation; S4, iterate to convergence or reach the maximum number of iterations; The dynamic adjustment logic is as follows: High complexity model: reduce ρ and v, and increase s to avoid device overload; High viscosity concrete: increase p and reduce v to ensure extrusion continuity; Device performance limit: limit v and a to prevent motion out of step; The parameter adaptive adjustment process adopts a multi-objective optimization strategy based on genetic algorithm, which finds the optimal parameter combination under the premise of ensuring printing quality and efficiency, so that the path planning algorithm can perform best under different conditions.

Images