3D printing parameter optimization method and system based on artificial intelligence

Abstract

The invention discloses a 3D printing parameter optimization method and system based on artificial intelligence, and the method comprises the steps: scanning a printing object, determining the size and shape of the printing object, and obtaining a digital model of the printing object; the three-dimensional digital model of the printing object is divided into a plurality of layered sheets according to a preset rule, and a single-layer printing path, corresponding to each sheet, in the printing process is obtained; constructing a dynamic error analysis model, and calculating a geometric error index of the current slice; when any index exceeds a threshold value, triggering a parameter optimization process; inputting the real-time data and the error indexes into a pre-trained artificial intelligence optimization model to obtain a parameter adjustment scheme; actual forming data are obtained through a three-dimensional scanner, accumulative errors are calculated, and weight parameters of the artificial intelligence optimization model are updated. The method has the advantages that the real-time printing data is compared with ideal data, the printing parameters are dynamically adjusted through the artificial intelligence optimization model, the path is corrected according to errors, and the precision and quality of the printing process are improved.

Description

Technical Field

[0001] The present invention relates to parameter optimization technology, and particularly to a 3D printing parameter optimization method and system based on artificial intelligence. Background Art

[0002] As an advanced manufacturing process, 3D printing technology has developed rapidly in recent years and has been widely applied in multiple industries such as aerospace, automotive, medical, and construction. 3D printing has obvious advantages in terms of design flexibility and processing cycle. It constructs objects by adding materials layer by layer, and can achieve complex and delicate geometric shapes, even structures that cannot be completed by traditional manufacturing processes, which makes it have great potential in innovative design and personalized customization.

[0003] Most of the current 3D printing parameter optimization methods and systems on the market rely on static parameter settings and cannot adjust printing parameters in real time during the printing process, resulting in difficulty in timely correcting errors during the printing process. This method usually can only analyze and correct errors after printing is completed, and cannot achieve adaptive adjustment during the printing process, easily causing unstable printing quality. Secondly, traditional 3D printing methods for error monitoring and adjustment are relatively simple, often relying only on the preset parameters of the equipment and manual adjustment, ignoring the dynamic influence of factors such as temperature changes and material properties on printing accuracy. While the optimization method based on artificial intelligence can collect real-time sensor data, monitor geometric errors during the printing process in real time, and perform dynamic correction, significantly improving printing accuracy and quality. In addition, most of the current technologies on the market also lack a comprehensive analysis of the entire printing process and are difficult to achieve precision optimization for different thin slices and different layers. Summary of the Invention

[0004] In order to improve the existing 3D printing parameter optimization methods and systems, a 3D printing parameter optimization method and system based on artificial intelligence are provided. This method dynamically adjusts printing parameters by collecting printing data in real time and comparing it with ideal data to ensure high-precision printing. The artificial intelligence optimization model significantly improves the accuracy and quality of the printing process according to the error correction path.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A 3D printing parameter optimization method based on artificial intelligence, including: Scanning the printing object to determine its size and shape, and obtaining the digital model of the printing object; Dividing the three-dimensional digital model of the printing object into multiple layered thin slices according to preset rules, and obtaining the single-layer printing path corresponding to each thin slice during the printing process; Based on sensors, collecting the printing data of each thin slice in real time, and constructing a dynamic error analysis model to calculate the geometric error index of the current thin slice; Compare based on geometric error metrics and a preset tolerance threshold. When any metric exceeds the threshold, trigger the parameter optimization process; Input the real-time data and error metrics into a pre-trained artificial intelligence optimization model to obtain a parameter adjustment plan, dynamically update the current printing instructions, and correct the printing path for subsequent thin slices; Based on after each layer is printed, obtain the actual formed data through a 3D scanner, calculate the cumulative error, and update the weight parameters of the artificial intelligence optimization model.

[0006] Preferably, the process of dividing the three-dimensional digital model of the printing object into multiple layered thin slices according to a preset rule and obtaining the single-layer printing path corresponding to each thin slice during printing specifically includes: The preset rule is specifically to determine the layer thickness according to the printing accuracy and material properties; Perform cutting along the Z-axis from the bottom to the top of the model based on the determined layer thickness; Obtain the outer boundary and inner filling path of each layer of the printing object through the convex hull algorithm and the inner filling algorithm respectively, and calculate and obtain all the contour functions of each layer; Move the print head from the end point of the previous layer to the start point of the current layer, and generate the printing path based on the contour function of each layer through the linear scanning algorithm.

[0007] Preferably, the process of collecting the printing data of each thin slice in real time based on the sensor and constructing a dynamic error analysis model to calculate the geometric error metrics of the current thin slice specifically includes: Execute the printing program based on the printing path, and collect the printing data of each layer in real time through the sensor; Construct a dynamic error analysis model based on possible influencing sources, compare the obtained real-time printing data of each layer with the ideal printing data of each layer, and obtain the geometric error of each layer of thin slices through finite element analysis; The possible influencing sources include: temperature change, material deformation, equipment error, and kinetic characteristics during the printing process; The geometric error includes: height error, radial error, and contour error.

[0008] Preferably, the process of comparing the geometric error metrics with the preset tolerance threshold and triggering the parameter optimization process when any metric exceeds the threshold specifically includes: Based on the height error, radial error, and contour error obtained from the dynamic error analysis model, compare them with the preset tolerance threshold one by one. If it is within the threshold, proceed to print the next layer. If it exceeds the threshold, trigger the optimization process.

[0009] Preferably, the steps of inputting the real-time data and error metrics into a pre-trained artificial intelligence optimization model, obtaining a parameter adjustment plan, dynamically updating the current printing instruction, and correcting the printing path of subsequent thin slices specifically include: Based on historical printing error data, input it into a parameter optimization model for training to obtain a parameter optimization model; Based on the situation of triggering the optimization process, input the real-time collected printing data and error metric data into the parameter optimization model to obtain the output parameter adjustment plan; Based on the obtained parameter adjustment plan, automatically update the current printing instruction; Through local correction and comprehensive correction, correct the subsequent printing path to ensure the geometric accuracy of each layer of thin slices.

[0010] Preferably, after each layer is printed, the steps of obtaining actual forming data through a 3D scanner, calculating the cumulative error, and updating the weight parameters of the artificial intelligence optimization model specifically include: Based on the printed object after printing, scan the whole through a 3D scanner to obtain the overall forming data; Compare the corrected printing data of each layer of thin slices with the data of each layer of thin slices in the digital model to obtain the parameter optimization error; Accumulate the parameter optimization errors of each layer of thin slices to obtain the overall cumulative error data; Based on the overall cumulative error data, input it into the parameter optimization model, and update the weight parameters of the model through the gradient descent method; Update the updated weight parameters to the parameter optimization model in real time.

[0011] Furthermore, a 3D printing parameter optimization system based on artificial intelligence is proposed, including: Scanning and Modeling Module: The scanning and modeling module is mainly used to scan the printing object and determine its size and shape to generate an accurate digital model for subsequent printing path planning; Layering and Path Generation Module: The layering and path generation module is mainly used to cut the digital model into multiple layered thin slices according to preset rules and generate the printing path of each thin slice to ensure the accuracy and stability of the printing process; Dynamic Error Analysis Module: The dynamic error analysis module is mainly used to use sensors to collect real-time printing data of each layer, construct a dynamic error analysis model, and evaluate the geometric error during the printing process; Error Threshold Comparison Module: The error threshold comparison module is mainly used to compare the real-time collected geometric error metrics with the preset tolerance threshold, and trigger the printing parameter optimization process when the error exceeds the threshold; Artificial intelligence optimization module: The artificial intelligence optimization module is mainly used to input real-time data and error metrics into a pre-trained artificial intelligence optimization model, generate a parameter adjustment plan, and dynamically update the printing instructions; Cumulative error calculation module: The cumulative error calculation module is mainly used to calculate the overall cumulative error of the printed object scanned by the 3D scanner and update the weight parameters of the optimization model; Processor: The processor is mainly used for the calculation process of each formula and the construction calculation process of each model.

[0012] Compared with the prior art, the advantages of the present invention are as follows: Through real-time data acquisition and dynamic error analysis, the printing accuracy and quality can be significantly improved. First, the system generates an accurate digital model through scanning and modeling, and combines layer-by-layer cutting and path planning to ensure the precise execution of each layer's printing path. Second, the printed data collected in real time is compared with the ideal data, which can timely detect and correct geometric errors caused by factors such as temperature changes and material deformation, avoiding quality problems caused by error accumulation in traditional 3D printing. Through the artificial intelligence optimization model, the system can automatically adjust the printing parameters according to real-time errors and dynamically correct the subsequent printing paths to ensure that each layer can be accurately completed. After each layer of printing is completed, the 3D scanner compares the actual formed data with the digital model, calculates the cumulative error, and updates the weight parameters of the optimization model, further improving the overall printing accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the method proposed by the present invention; Figure 2 It is a schematic diagram of the layer-by-layer division of the printed object proposed by the present invention; Figure 3 It is a schematic diagram of obtaining geometric error indicators proposed by the present invention; Figure 4 It is a schematic diagram of parameter adjustment and path optimization proposed by the present invention; Figure 5 It is a schematic diagram of weight parameter optimization proposed by the present invention; Figure 6 It is an architecture diagram of the electronic device in this solution; Figure 7 It is a schematic diagram of the structure of the computer-readable storage medium in this solution. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations.

[0015] A 3D printing parameter optimization system based on artificial intelligence, comprising: Scanning and Modeling Module: The scanning and modeling module is mainly used to generate an accurate digital model by scanning the printing object and determining its size and shape for subsequent printing path planning; Layering and Path Generation Module: The layering and path generation module is mainly used to cut the digital model into multiple layered slices according to preset rules and generate the printing path for each slice to ensure the accuracy and stability of the printing process; Dynamic Error Analysis Module: The dynamic error analysis module is mainly used to collect the printing data of each layer in real time by using sensors, construct a dynamic error analysis model, and evaluate the geometric error in the printing process; Error Threshold Comparison Module: The error threshold comparison module is mainly used to compare the geometric error index collected in real time with the preset tolerance threshold. When the error exceeds the threshold, it triggers the printing parameter optimization process; Artificial Intelligence Optimization Module: The artificial intelligence optimization module is mainly used to input the real-time data and error index into the pre-trained artificial intelligence optimization model, generate a parameter adjustment scheme, and dynamically update the printing instructions; Cumulative Error Calculation Module: The cumulative error calculation module is mainly used to scan the printed object completed by the 3D scanner, calculate the overall cumulative error, and update the weight parameters of the optimization model; Processor: The processor is mainly used for the calculation process of each formula and the construction and calculation process of each model.

[0016] Refer to Figure 1 As shown, an artificial intelligence-based 3D printing parameter optimization method includes: Step 1: Scan the printing object to determine its size and shape, and obtain the digital model of the printing object; Step 2: Divide the three-dimensional digital model of the printing object into multiple layered slices according to preset rules, and obtain the single-layer printing path corresponding to each slice during the printing process; Step 3: Based on sensors, collect the printing data of each slice in real time, construct a dynamic error analysis model, and calculate the geometric error index of the current slice; Step 4: Based on the comparison between the geometric error index and the preset tolerance threshold, when any index exceeds the threshold, trigger the parameter optimization process; Step 5: Input the real-time data and error index into the pre-trained artificial intelligence optimization model, obtain a parameter adjustment scheme, dynamically update the current printing instructions, and correct the printing path of the subsequent slices; Step 6: After each layer is printed, obtain the actual forming data through a 3D scanner, calculate the cumulative error, and update the weight parameters of the artificial intelligence optimization model.

[0017] Refer to Figure 2As shown, the three-dimensional digital model of the printing object is segmented into multiple layered thin slices according to a preset rule, and obtaining the single-layer printing path during the printing process corresponding to each thin slice specifically includes: The preset rule is specifically to determine the layer thickness according to the printing accuracy and material properties; Based on the determined layer thickness, cut from the bottom to the top of the model along the Z-axis; Obtain the outer boundary and inner filling path of each layer of the printing object through the convex hull algorithm and the inner filling algorithm respectively, and calculate and obtain all contour functions of each layer; Move the print head from the end point of the previous layer to the start point of the current layer, and generate the printing path through the linear scanning algorithm based on the contour function of each layer.

[0018] Specifically, the determination of the layer thickness is usually determined by the printing accuracy and material properties. The higher the accuracy, the smaller the layer thickness. According to the determined layer thickness, cut the model from the bottom to the top along the Z-axis, and the cutting position interval of each layer is the layer thickness; Extract the outer boundary path of each layer through the convex hull algorithm. Given a set of boundary point sets, sort the points according to the x coordinate, traverse the sorted point set using the stack structure, calculate the convex hull boundary, and the calculation result is the convex hull boundary point set, which constitutes the outer boundary path; The inner filling path is generated through the inner filling algorithm. Fill the outer boundary of each layer. First, determine the filling interval, and use the horizontal scan line to cut the filling area from top to bottom to generate the inner filling path of each layer; Combine the outer boundary path and the inner filling path into a contour function. Move the print head from the end point position of the previous layer to the start point position of the current layer. According to the contour of the contour function and the inner filling requirements, generate a series of continuous printing paths, and according to the control strategy of the printer, determine the order of path scanning. Usually, strategies such as from the outside to the inside or serpentine scanning are adopted. According to the generated path, control the print head to move along the predetermined trajectory.

[0019] Refer to Figure 3 As shown, based on the sensor, real-time collect the printing data of each thin slice, and construct a dynamic error analysis model to calculate the geometric error index of the current thin slice, which specifically includes: Execute the printing program based on the printing path, and real-time collect the printing data of each layer through the sensor; Construct a dynamic error analysis model based on possible influencing sources, compare the obtained real-time printing data of each layer with the ideal printing data of each layer, and obtain the geometric error of each layer of thin slices through finite element analysis; The possible influencing sources include: temperature change, material deformation, equipment error, and dynamic characteristics during the printing process; The geometric error includes: height error, radial error, and contour error.

[0020] Specifically, the actual printing data of each layer during the printing process is collected in real time by sensors. These data include information such as the spatial coordinates, temperature, and mechanical state of each point; Based on possible influencing sources (temperature changes, material deformation, equipment errors, and dynamic characteristics), a dynamic error analysis model is established. Among them, temperature changes will cause thermal expansion of the material, thereby affecting the geometry during the printing process. The influence of temperature on the printed object can be expressed as: , where, is the height error caused by temperature changes, is the coefficient of thermal expansion, is the temperature change, is the length of the printing path; Physical properties such as the viscosity and hardness of the material will affect the deformation during the printing process, which can be expressed by the elastic modulus E and Poisson's ratio v. The formula is: , where, is the strain of the material, is the applied stress, is the elastic modulus of the material, is Poisson's ratio; Equipment errors include printer accuracy, drive system errors, etc., which can usually be obtained through calibration measurement. This part of the error can be compensated using an empirical model or the system error measured through experiments; During the printing process, the mechanical system of the printer will be affected by dynamic characteristics (such as inertia, vibration, etc.), resulting in slight changes in the position of the print head.

[0021] Based on the comparison between the geometric error index and the preset tolerance threshold, when any index exceeds the threshold, the parameter optimization process is triggered, specifically including: Based on the height error, radial error, and contour error obtained from the dynamic error analysis model, they are compared with the preset tolerance threshold one by one. If it is within the threshold, the next layer of printing is performed. If it exceeds the threshold, the optimization process is triggered.

[0022] Specifically, according to the error type, the relevant printing parameters are adjusted, including: Temperature adjustment: Adjust the temperature of the print head and the printing bed to reduce the error caused by temperature changes; Printing speed adjustment: Adjust the printing speed, especially in cases where equipment errors or dynamic instability are likely to occur; Interlayer deviation correction: Fine-tune the printing path of each layer based on the error model to avoid cumulative errors.

[0023] Refer to Figure 4As shown, input the real-time data and error metrics into the pre-trained artificial intelligence optimization model to obtain a parameter adjustment plan, dynamically update the current printing instructions, and correct the printing path of subsequent wafers. Specifically, it includes: Based on the historical printing error data, input it into the parameter optimization model for training to obtain the parameter optimization model; Based on the situation of triggering the optimization process, input the real-time collected printing data and error metric data into the parameter optimization model to obtain the output parameter adjustment plan; Based on the obtained parameter adjustment plan, automatically update the current printing instructions; Through local correction and overall correction, correct the subsequent printing path to ensure the geometric accuracy of each layer of wafer.

[0024] Specifically, input the real-time collected printing parameters and error metrics into the trained artificial intelligence optimization model. The model outputs a parameter adjustment plan, and based on the obtained parameter adjustment plan, automatically update the current printing instructions; The update of the printing path may require correction of the subsequent printing layers to ensure the geometric accuracy between layers. During this process, local or overall correction is performed; Local correction corrects the error of the current layer. For example, increase or decrease the compensation of the printing path in certain areas. The formula is: , where, is the corrected printing path, is the current printing path, is the error correction amount; Overall correction adjusts the parameters of the path during the entire printing process and corrects the geometric shape of the printing path; During the subsequent printing process, based on the error data of the current layer, the printing path of the next layer can be adjusted. To avoid discontinuity introduced by error compensation, smooth the subsequent printing path and use a smoothing function to ensure the smoothness of the path and printing stability.

[0025] Refer to Figure 5 As shown, based on after each layer of printing is completed, obtain the actual forming data through a 3D scanner, calculate the cumulative error, and update the weight parameters of the artificial intelligence optimization model. Specifically, it includes: Based on the printed object that has been printed, scan the whole through a 3D scanner to obtain the overall forming data; Compare the corrected printing data of each layer of wafer with the data of each layer of wafer in the digital model to obtain the parameter optimization error; Accumulate the parameter optimization errors of each layer of wafer to obtain the overall cumulative error data; Based on the overall cumulative error data, input it into the input parameter optimization model, and update the weight parameters of the model through the gradient descent method; Update the updated weight parameters to the parameter optimization model in real time.

[0026] Specifically, after each layer is printed, use a 3D scanner to scan the overall printed object to obtain the three-dimensional forming data of the entire printed object. Compare the corrected printing data of each layer of thin slices with the data of each layer of thin slices in the digital model to obtain the parameter optimization error, which is calculated by the Euclidean distance. The formula is: , where, is the parameter optimization error, and the ideal coordinate of each printing point is , and the coordinate after actual printing is ; Accumulate the errors of each layer to obtain the overall cumulative error, and then input it into the parameter optimization model. Optimize the weight parameters through the gradient descent method to minimize the loss function. The update rule of the gradient descent method is: , where, is the updated weight parameter, is the weight parameter before update, is the learning rate, is the gradient of the loss function with respect to the weight parameter ; By calculating the gradient of each parameter, the gradient descent method will gradually adjust the weight parameter , so that the prediction error output by the model is close to the actual error. At the same time, the updated optimization model will immediately affect parameters such as path planning, printing speed, and printing temperature in the subsequent printing process, ensuring that the geometric accuracy of each layer of printing is improved.

[0027] Furthermore, the method according to the embodiment of the present application can also be implemented by means of the Figure 6 shown architecture of the electronic device. As Figure 6 shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to the network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store a 3D printing parameter optimization method and system provided by the present application. The electronic device 500 may further include a terminal interface 508. Of course, Figure 6 the architecture shown Figure 6 is only exemplary. When implementing different devices, one or more components shown in the electronic device may be omitted according to actual needs.

[0028] Figure 7 This is a schematic structural diagram of a computer-readable storage medium provided by an embodiment of the present application. As Figure 7 shown, it is a computer-readable storage medium 600 according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are run by a processor, a 3D printing parameter optimization method and system based on artificial intelligence according to the embodiment of the present application described with reference to the above drawings can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0029] It should be noted that: the above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of this specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0030] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key point of each embodiment is to illustrate the differences from other embodiments.

[0031] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A 3D printing parameter optimization method based on artificial intelligence, characterized in that: include: Scan the printed object to determine its size and shape, and obtain a digital model of the printed object; The three-dimensional digital model of the printing object is divided into a plurality of layered slices according to a preset rule, and a single-layer printing path corresponding to each slice in the printing process is obtained; The printing data of each sheet is collected in real time based on the sensor, and a dynamic error analysis model is constructed to calculate the geometric error index of the current sheet; Compare the geometric error index with the preset tolerance threshold. When any index exceeds the threshold, the parameter optimization process is triggered. Input the real-time data and error indicators into the pre-trained AI optimization model to obtain parameter adjustment solutions, dynamically update the current printing instructions, and correct the printing path of subsequent sheets; After each layer is printed, the actual molding data is obtained through a 3D scanner, the cumulative error is calculated, and the weight parameters of the artificial intelligence optimization model are updated.

2. The method for optimizing 3D printing parameters based on artificial intelligence according to claim 1, characterized in that: The method of dividing the three-dimensional digital model of the printing object into a plurality of layered sheets according to a preset rule and obtaining a single-layer printing path corresponding to each sheet in the printing process specifically includes: The preset rules specifically determine the layer thickness based on the printing accuracy and material properties; Cut along the Z axis from the bottom to the top of the model based on the determined layer thickness; The outer boundary and inner filling path of each layer of the printed object are obtained by the convex hull algorithm and the inner filling algorithm, and all contour functions of each layer are calculated; Move the print head from the end point of the previous layer to the start point of the current layer, and generate a printing path based on the contour function of each layer through a linear scanning algorithm.

3. The method for optimizing 3D printing parameters based on artificial intelligence according to claim 1, characterized in that: The real-time collection of printing data of each sheet based on the sensor, and the construction of a dynamic error analysis model to calculate the geometric error index of the current sheet specifically include: Execute the printing program based on the printing path and collect the printing data of each layer in real time through sensors; A dynamic error analysis model is built based on possible influencing sources, and the real-time printing data of each layer is compared with the ideal printing data of each layer. The geometric error of each layer is obtained through finite element analysis. The possible sources of influence include: temperature change, material deformation, equipment error, and dynamic characteristics during printing; The geometric errors include: height error, radial error and profile error.

4. The method for optimizing 3D printing parameters based on artificial intelligence according to claim 1, characterized in that: The geometric error index is compared with the preset tolerance threshold. When any index exceeds the threshold, the parameter optimization process is triggered, which specifically includes: The height error, radial error and contour error obtained based on the dynamic error analysis model are compared one by one with the preset tolerance threshold. If they are within the threshold, the next layer is printed. If they exceed the threshold, the optimization process is triggered.

5. The method for optimizing 3D printing parameters based on artificial intelligence according to claim 1, characterized in that: The method of inputting the real-time data and error index into the pre-trained artificial intelligence optimization model, obtaining the parameter adjustment scheme, dynamically updating the current printing instruction, and correcting the printing path of the subsequent sheets specifically includes: Based on the historical printing error data, the data is input into the parameter optimization model for training to obtain the parameter optimization model; Based on the situation of triggering the optimization process, the real-time collected printing data and error index data are input into the parameter optimization model to obtain the output parameter adjustment plan; Automatically update the current printing instructions based on the acquired parameter adjustment plan; Through local correction and overall correction, the subsequent printing path is corrected to ensure the geometric accuracy of each layer of the sheet.

6. The method for optimizing 3D printing parameters based on artificial intelligence according to claim 1, characterized in that: After each layer is printed, the actual molding data is obtained through a 3D scanner, the cumulative error is calculated, and the weight parameters of the artificial intelligence optimization model are updated, which specifically include: Based on the printed object, the whole is scanned by a 3D scanner to obtain the overall molding data; The corrected printed data of each layer of the slice is compared with the data of each layer of the slice in the digital model to obtain the parameter optimization error; Accumulate the parameter optimization errors of each layer of thin slices to obtain the overall cumulative error data; Based on the overall accumulated error data, it is input into the parameter optimization model, and the weight parameters of the model are updated by the gradient descent method; The updated weight parameters are updated to the parameter optimization model in real time.

7. An artificial intelligence-based 3D printing parameter optimization system, used to implement an artificial intelligence-based 3D printing parameter optimization method as described in any one of claims 1 to 6, characterized in that: include: Scanning and modeling module: The scanning and modeling module is mainly used to generate an accurate digital model by scanning the printing object and determining its size and shape for subsequent printing path planning; Layering and path generation module: The layering and path generation module is mainly used to cut the digital model into multiple layered sheets according to preset rules, generate the printing path for each sheet, and ensure the accuracy and stability of the printing process; Dynamic error analysis module: The dynamic error analysis module is mainly used to collect the printing data of each layer in real time using sensors, build a dynamic error analysis model, and evaluate the geometric errors in the printing process; Error threshold comparison module: The error threshold comparison module is mainly used to compare the geometric error index collected in real time with the preset tolerance threshold. When the error exceeds the threshold, the printing parameter optimization process is triggered; Artificial intelligence optimization module: The artificial intelligence optimization module is mainly used to input real-time data and error indicators into the pre-trained artificial intelligence optimization model, generate parameter adjustment plans and dynamically update printing instructions; Cumulative error calculation module: The cumulative error calculation module is mainly used to calculate the overall cumulative error of the printed object scanned by the three-dimensional scanner and update the weight parameters of the optimization model; Processor: The processor is mainly used for the calculation process of each formula and the construction calculation process of each model.

8. An electronic device, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the artificial intelligence-based 3D printing parameter optimization method as described in any one of claims 1-6.

9. A computer-readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by a processor, an artificial intelligence-based 3D printing parameter optimization method according to any one of claims 1 to 6 is implemented.

Images