Intelligent Inference Method for Machine Tool Machining Accuracy Driven by Iterative Physically Embedded Graph Network

Abstract

The present application discloses an intelligent inference method for machine tool machining accuracy driven by an iterative physical embedding graph network, which relates to the technical field of machining accuracy inference. The method includes: solving the physical fields of each axis of the machine tool by using a physical field solving model; then, based on the physical fields of each axis of the machine tool, determining the thermal-mechanical errors of each axis of the machine tool respectively, and finally, combining with the forward motion chain model of the machine tool, obtaining a forward transfer model of the thermal-mechanical errors of the machine tool to realize intelligent inference of the machine tool machining accuracy; when the solution error of the physical field solving model is greater than a preset error threshold, the iteration unit retains some network parameters based on the weight sharing mechanism, and iteratively independently learns and fine-tunes the model parameters according to real-time data. The solution of the present application introduces thermal and stress mechanism equations to participate in the training process of the model, realizes efficient solution of the physical field with interpretability, and ensures intelligent inference of the machine tool machining accuracy under dynamic and complex working conditions based on the iterative update mechanism.

Description

Technical Field

[0001] The present application relates to the technical field of machining accuracy inference, and particularly to an intelligent inference method for machine tool machining accuracy driven by an iterative physically embedded graph network. Background Art

[0002] With the rapid development of the manufacturing industry, more stringent requirements are put forward for the machining accuracy of numerically controlled machine tools. Factors such as bearing heating, frictional effects, and unilateral or bilateral fixed constraints of each axis of the machine tool lead to thermal-mechanical deformation of each axis of the machine tool. Statistical data shows that the errors caused by thermal-mechanical factors can account for 40%-70% of the total errors of the machine tool, which has a significant impact on the machining accuracy of the machine tool. Therefore, accurately predicting the real-time thermal-mechanical errors of the machine tool under different working conditions and performing accuracy inference are crucial for adjusting machining parameters and optimizing compensation strategies, which can ensure high precision and stability during the machining process of the machine tool.

[0003] Currently, in order to infer the machining accuracy of the machine tool under the action of thermal-mechanical errors, domestic and foreign scholars mainly adopt two types of methods: mechanism-driven and data-driven. The mechanism-driven method usually takes complex boundary conditions such as bearing heat generation, heat conduction, and mechanical constraints as the calculation basis, and simplifies the analysis of each axis of the machine tool through methods such as the lumped mass method and finite element analysis, and then solves its deformation field. This type of method fully integrates thermal and mechanical mechanisms and often has a high error prediction accuracy under specific working conditions. However, since its calculation accuracy mostly depends on the fineness of mesh division, and with the improvement of the calculation accuracy requirements, the calculation time increases exponentially; at the same time, this type of method has poor adaptability to changes in working conditions and is difficult to predict thermal-mechanical errors in real time in a complex and changeable machining environment.

[0004] The data-driven method is mainly based on technologies such as machine learning, using the sensing data such as temperature and stress at finite nodes of each axis of the machine tool as input and thermal-mechanical errors as output to establish an error inference model. Compared with the mechanism-driven method, the data-driven error inference method can make full use of the real-time data of the actual machining process of the machine tool and usually has good real-time performance and accuracy. However, this type of method generally lacks thermal-mechanical mechanism analysis, making the internal working principle between the "input-output" of the data-driven model lack interpretability, and the prediction accuracy depends on the existing training samples. For unknown working conditions that cannot be covered by the training samples, the inference effect is difficult to guarantee. Summary of the Invention

[0005] The purpose of the present application is to provide an intelligent inference method for machine tool machining accuracy driven by an iterative physically embedded graph network, which can realize the intelligent inference of machine tool machining accuracy under dynamic and complex working conditions.

[0006] To achieve the above object, the present application provides the following solutions:

[0007] The present application provides an intelligent inference method for the machining accuracy of a machine tool driven by an iterative physical embedding graph network, including the following steps:

[0008] Obtain the temperature sensing data and stress sensing data of each axis of the machine tool; the axes of the machine tool include the main shaft and the feed axis.

[0009] Using a pre-trained physical field solution model, solve the stress field of each axis of the machine tool according to the temperature sensing data and stress sensing data of each axis of the machine tool; the physical field solution model is a model based on a physical embedding graph network; the physical field solution model uses a multi-scale graph attention network to capture the sensing features of different scales of each axis of the machine tool, and through a physical embedding loss function, uses the real-time temperature sensing data of each axis of the machine tool, the real-time stress sensing data of each axis of the machine tool, and the mechanism equation to standardize the training process of the physical field solution model.

[0010] Based on the stress field of each axis of the machine tool, determine the thermal-mechanical error of each axis of the machine tool respectively.

[0011] According to the thermal-mechanical error of each axis of the machine tool and the forward motion chain model of the machine tool, determine the forward transfer model of the thermal-mechanical error of the machine tool; the forward transfer model of the thermal-mechanical error of the machine tool includes the thermal-mechanical errors of the main shaft and each feed axis from the main shaft of the machine tool to the tool at the end of the machine tool, realizing intelligent inference of the machining accuracy of the machine tool.

[0012] When the solution error of the physical field solution model is greater than the preset error threshold, the iterative unit locally iteratively adjusts the parameters of the physical field solution model based on the real-time sensing data according to the weight sharing mechanism until the solution error of the physical field solution model is less than the preset error threshold; the solution error of the physical field solution model is the root mean square error value between the real-time stress sensing data of any axis of the machine tool and the value at the corresponding node of the stress field of the axis obtained by solving through the physical field solution model.

[0013] Optionally, locally iteratively adjusting the parameters of the physical field solution model by the iterative unit according to the real-time sensing data based on the weight sharing mechanism specifically includes the following steps:

[0014] Based on the weight sharing mechanism, retain some network parameters in the physical field solution model, and reshape the network connection relationship with the physical field solution model through the iterative unit.

[0015] Independently learn and fine-tune the parameters of the physical field solution model based on the real-time temperature sensing data or stress sensing data, and correct the error of the physical field solution model under specific working conditions.

[0016] Optionally, obtain the pre-trained physical field solution model through the following process:

[0017] Obtain the temperature sensing data of each axis of the machine tool and the stress sensing data of each axis of the machine tool.

[0018] Based on the temperature sensing data of each axis of the machine tool, the stress sensing data of each axis of the machine tool, and the mechanism equation, the actual stress field of each axis of the machine tool is solved.

[0019] Taking the temperature sensing data of each axis of the machine tool and the stress sensing data of each axis of the machine tool as the input of the physical field solution model, and taking the actual stress field of each axis of the machine tool as the target output of the physical field solution model, the physical field solution model is trained to obtain a pre-trained physical field solution model.

[0020] Optionally, when training the physical field solution model, a mechanism-data loss dual-rail traction mechanism that can balance the data loss term and the mechanism loss term is used to train the physical field solution model; the data loss term is the mean square error between the real-time sensing data and the sensing data at the corresponding nodes of the solved physical field; the mechanism loss terms are the heat conduction equation, the thermal stress equation, the stress-strain equation, and the stress transfer equation, which are used to constrain the training process of the physical field solution model. Using the mechanism-data loss dual-rail traction mechanism to train the physical field solution model specifically includes:

[0021] The data loss and the mechanism loss are transformed through an embedding layer to obtain a data loss vector and a mechanism loss vector.

[0022] The dot product attention mechanism is used to calculate the correlation between the data loss vector and the mechanism loss vector, and the correlation score is normalized to obtain the weight of the data loss term and the weight of the mechanism loss term.

[0023] Combining the weighted loss of the data loss term, the weighted loss of the mechanism loss term, the residual penalty, and the penalty term for small weights, the total loss is calculated.

[0024] A dynamic factor is used to adjust the weight of the data loss term and the weight of the mechanism loss term according to the change of the total loss, balance the contributions between the mechanism loss and the data loss of the physical field solution model, thereby improving the stability and accuracy during the training of the physical field solution model.

[0025] Optionally, based on the temperature sensing data of each axis of the machine tool, the stress sensing data of each axis of the machine tool, and the mechanism equation, solving the actual stress field of each axis of the machine tool specifically includes the following steps:

[0026] Based on the temperature sensing data of each axis of the machine tool and the heat conduction equation, the temperature field of each axis of the machine tool is solved.

[0027] Based on the temperature field of each axis of the machine tool and the temperature-thermal strain equation, the thermal strain load of each axis of the machine tool is calculated.

[0028] Based on the thermal strain loads, stress sensing data, stress-strain equations, and stress transfer equations of each axis of the machine tool, the stress fields of each axis of the machine tool are solved and obtained.

[0029] Optionally, the multi-scale graph attention network gradually captures the sensing features of different scales of the sensing data of each axis of the machine tool from local to global through multi-layer graph convolution operations that expand layer by layer; after each layer of graph convolution operation, the multi-head attention mechanism is used to weightedly adjust the information propagation intensity between sensing nodes, and further strengthen the fusion of sensing features at different scales.

[0030] Optionally, based on the stress fields of each axis of the machine tool, the thermal-mechanical errors of each axis of the machine tool are determined respectively, which specifically include the following steps:

[0031] Based on the stress fields of each axis of the machine tool, the strain fields of each axis of the machine tool are determined respectively.

[0032] Based on the strain fields of each axis of the machine tool, the thermal-mechanical errors of each axis of the machine tool are determined respectively; the thermal-mechanical errors include the linear displacement of each axis of the machine tool and the angular deviation of each axis of the machine tool.

[0033] Optionally, according to the thermal-mechanical errors of each axis and the forward kinematic chain model of the machine tool, a forward transfer model of the thermal-mechanical errors of the machine tool is determined, which specifically includes the following steps:

[0034] Based on the transfer relationship of each axis of the machine tool, a forward kinematic chain model of the machine tool is established using a homogeneous transformation matrix.

[0035] Based on the differential motion theory, the thermal-mechanical errors of each axis of the machine tool are transformed into corresponding differential expressions.

[0036] The differential expressions of the thermal-mechanical errors of each axis of the machine tool are fused with the forward kinematic chain model of the machine tool to obtain a forward transfer model of the thermal-mechanical errors of the machine tool.

[0037] According to the specific embodiments provided in this application, the following technical effects are disclosed in this application:

[0038] The present application provides an intelligent inference method for the machining accuracy of a machine tool driven by an iterative physically embedded graph network. In this method, according to the acquired temperature sensing data and stress sensing data of each axis of the machine tool, the stress field of each axis of the machine tool is solved by using a pre-trained physical field solution model. The physical field solution model in this step is a model based on a physically embedded graph network, which can use a multi-scale graph attention network to capture the physical field changes at different levels and in different ranges of the sensing data of each axis of the machine tool, and in the training process, through a physical embedding loss function, the real-time temperature sensing data of each axis, the real-time stress sensing data of each axis, and the mechanism equation are used to standardize the training process of the physical field solution model. Subsequently, based on the stress field of each axis of the machine tool, the thermal-mechanical errors of each axis of the machine tool are determined respectively. Finally, combined with the forward motion chain model of the machine tool, the forward transfer model of the thermal-mechanical errors of the machine tool can be determined to realize the intelligent inference of the machining accuracy of the machine tool. When the solution error of the physical field solution model is greater than a preset error threshold, the iterative unit retains some network parameters based on the weight sharing mechanism, and iteratively independently learns and fine-tunes the model parameters according to real-time data. The solution provided by the present application fully captures the physical field change characteristics at different levels and ranges of the sensing data through a multi-scale graph attention network, and introduces thermal and stress mechanism equations to participate in the training process of the model, realizing an efficient solution of the physical field with interpretability. When the solution error is greater than the threshold, through an online iterative update method, the problem that traditional pre-trained models are difficult to respond to working condition changes in real time is effectively overcome, and the intelligent inference of the machining accuracy of the machine tool under dynamic and complex working conditions is realized. Description of the Drawings

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0040] Figure 1 It is a flowchart of an intelligent inference method for the machining accuracy of a machine tool driven by an iterative physically embedded graph network provided by an embodiment of the present application.

[0041] Figure 2 It is a flowchart of obtaining a pre-trained physical field solution model in an intelligent inference method for the machining accuracy of a machine tool driven by an iterative physically embedded graph network provided by an embodiment of the present application.

[0042] Figure 3 It is a flowchart of step B2 in an intelligent inference method for the machining accuracy of a machine tool driven by an iterative physically embedded graph network provided by an embodiment of the present application.

[0043] Figure 4It is a flowchart of step B3 in an iterative physical embedding graph network-driven intelligent inference method for machine tool machining accuracy provided by an embodiment of the present application.

[0044] Figure 5 It is a flowchart of step A3 in an iterative physical embedding graph network-driven intelligent inference method for machine tool machining accuracy provided by an embodiment of the present application.

[0045] Figure 6 It is a flowchart of step A4 in an iterative physical embedding graph network-driven intelligent inference method for machine tool machining accuracy provided by an embodiment of the present application.

[0046] Figure 7 It is a flowchart of step A5 in an iterative physical embedding graph network-driven intelligent inference method for machine tool machining accuracy provided by an embodiment of the present application. Detailed implementation manners

[0047] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0048] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the drawings and specific implementation manners.

[0049] An iterative physical embedding graph network-driven intelligent inference method for machine tool machining accuracy provided by an embodiment of the present application. In an exemplary embodiment, as Figure 1 shown, it includes the following steps:

[0050] A1. Obtain the temperature sensing data and stress sensing data of each axis of the machine tool; the axes of the machine tool include the main axis and the feed axis. Temperature sensors and stress sensors are arranged near each axis of the machine tool to collect the temperature sensing data and stress sensing data during the operation of the machine tool in real time.

[0051] A2. Use a pre-trained physical field solving model to solve the stress field of each axis of the machine tool according to the temperature sensing data and stress sensing data of each axis of the machine tool; the physical field solving model is a model based on a physical embedding graph network.

[0052] In the physical field solving model constructed in this embodiment, a multi-scale graph attention network is used to capture the sensing features of different scales of each axis of the machine tool, and through a physical embedding loss function, the real-time temperature sensing data of each axis of the machine tool, the real-time stress sensing data of each axis of the machine tool, and the mechanism equation are used to standardize the training process of the physical field solving model.

[0053] Specifically, in this physical field solution model, the multi-scale graph attention network gradually captures the sensing features of different scales of the sensing data of each axis of the machine tool from local to global through multi-layer graph convolution operations that expand layer by layer; after each layer of graph convolution operation, the multi-head attention mechanism is used to weightedly adjust the information propagation intensity between sensing nodes, and further strengthen the fusion of sensing features at different scales.

[0054] The physical embedding loss function includes two parts: a data loss term and a mechanism loss term. The data loss is the mean square error between the real-time sensing data and the sensing data at the corresponding nodes of the physical field obtained by the solution; the mechanism loss is mechanism equations such as the heat conduction equation, the thermal stress equation, the stress-strain equation, and the stress transfer equation, which are used to constrain the training process of the network to ensure that the network solution result conforms to the real physical laws.

[0055] As an exemplary embodiment, a pre-trained physical field solution model is obtained through the following process, as Figure 2 shown, including the following steps:

[0056] B1. Obtain the temperature sensing data of each axis of the machine tool and the stress sensing data of each axis of the machine tool.

[0057] B2. Based on the temperature sensing data of each axis of the machine tool, the stress sensing data of each axis of the machine tool, and the mechanism equations, solve to obtain the actual stress field of each axis of the machine tool. In this embodiment, as Figure 3 shown, step B2 specifically includes the following steps:

[0058] B21. Based on the temperature sensing data of each axis of the machine tool and the heat conduction equation, solve to obtain the temperature field of each axis of the machine tool.

[0059] B22. Based on the temperature field of each axis of the machine tool and the temperature-thermal strain equation, calculate the thermal strain load of each axis of the machine tool.

[0060] B23. Based on the thermal strain load of each axis of the machine tool, the stress sensing data, the stress-strain equation, and the stress transfer equation, solve to obtain the stress field of each axis of the machine tool. Subsequently, based on step B2, the stress field training model is solved, thereby realizing the training process of standardizing the physical field solution model through mechanism equations, real-time temperature sensing data, and stress sensing data.

[0061] B3. Use the temperature sensing data of each axis of the machine tool and the stress sensing data of each axis of the machine tool as the input of the physical field solution model, and use the actual stress field of each axis of the machine tool as the target output of the physical field solution model to train the physical field solution model to obtain a pre-trained physical field solution model.

[0062] Specifically, in this embodiment, when training the physical field solution model, a mechanism-data loss dual-rail traction mechanism that can balance the data loss term and the mechanism loss term is used to train the physical field solution model.

[0063] Using the mechanism-data loss dual-rail traction mechanism to train the physical field solution model, as Figure 4 shown, specifically includes the following processes:

[0064] B31. Convert the data loss and the mechanism loss through the embedding layer to obtain a data loss vector and a mechanism loss vector.

[0065] B32. Use the dot product attention mechanism to calculate the correlation between the data loss vector and the mechanism loss vector, and normalize the correlation score to obtain the weight of the data loss term and the weight of the mechanism loss term.

[0066] B33. Combine the weighted loss of the data loss term, the weighted loss of the mechanism loss term, the residual penalty, and the penalty term for small weights to calculate the total loss;

[0067] B34. Use a dynamic factor to adjust the weight of the data loss term and the weight of the mechanism loss term according to the change of the total loss. Balance the contribution between the mechanism loss and the data loss of the physical field solution model, thereby improving the stability and accuracy during the training of the physical field solution model.

[0068] A3. Based on the stress field of each axis of the machine tool, determine the thermal-mechanical error of each axis of the machine tool respectively. In this embodiment, as Figure 5 shown, step A3 specifically includes the following steps:

[0069] A31. Based on the stress field of each axis of the machine tool, determine the strain field of each axis of the machine tool respectively.

[0070] A32. Based on the strain field of each axis of the machine tool, determine the thermal-mechanical error of each axis of the machine tool respectively; the thermal-mechanical error includes the linear displacement of each axis of the machine tool and the angular deviation of each axis of the machine tool.

[0071] A4. According to the thermal-mechanical error of each axis of the machine tool and the forward motion chain model of the machine tool, determine the forward transmission model of the thermal-mechanical error of the machine tool; the forward transmission model of the thermal-mechanical error of the machine tool encompasses the thermal-mechanical errors of the spindle and each feed axis from the spindle of the machine tool to the tool at the end of the machine tool, realizing intelligent inference of the machining accuracy of the machine tool. In this embodiment, as Figure 6 shown, step A4 specifically includes the following steps:

[0072] A41. Based on the transmission relationship of each axis of the machine tool, establish a forward motion chain model of the machine tool using the homogeneous transformation matrix.

[0073] A42. Convert the thermal-mechanical errors of each axis of the machine tool into corresponding differential expressions based on the differential motion theory, such as small displacements, small rotational changes, etc.

[0074] A43. Integrate the differential expressions of the thermal-mechanical errors of each axis of the machine tool with the forward motion chain model of the machine tool to obtain the forward transfer model of the thermal-mechanical errors of the machine tool.

[0075] A5. When the solution error of the physical field solution model is greater than the preset error threshold, the iterative unit locally iteratively adjusts the parameters of the physical field solution model based on the real-time sensing data according to the weight sharing mechanism until the solution error of the physical field solution model is less than the preset error threshold. The solution error of the physical field solution model is the root mean square error value between the real-time stress sensing data of any axis of the machine tool and the values at the corresponding nodes of the stress field of the axis obtained by solving through the physical field solution model.

[0076] In real-time inference, if the solution error of the physical field solution model is greater than the preset error threshold, the iterative unit locally iteratively adjusts the parameters of the physical field solution model based on the real-time sensing data according to the weight sharing mechanism, such as Figure 7 shown, and the process specifically includes the following steps:

[0077] A51. Retain some network parameters in the physical field solution model based on the weight sharing mechanism, and reshape the network connection relationship between the iterative unit and the physical field solution model. Through the weight sharing mechanism, the same parameters are allowed to be used at different positions or different layers between the iterative unit and the original physical field solution model.

[0078] A52. Independently learn and fine-tune the parameters of the physical field solution model based on the real-time temperature sensing data or stress sensing data, and correct the errors of the physical field solution model under specific working conditions.

[0079] After the solution error of the physical field solution model is lower than the preset error threshold, use the iterated physical field solution model to replace the original physical field solution model, ensuring that the physical field solution model can continuously improve the accuracy through local adjustment and feedback under real-time changing working conditions, while maintaining the calculation efficiency and training stability, and continuously improving the machining accuracy of the machine tool.

[0080] In the above solution provided by this application, according to the temperature sensing data and stress sensing data of each axis of the machine tool, the stress field of each axis of the machine tool is solved by using a pre-trained physical field solution model; the physical field solution model in this step can use a multi-scale graph attention network to capture the sensing features of different scales of each axis of the machine tool, and in the training process, through a physical embedding loss function, the real-time temperature sensing data of each axis, the real-time stress sensing data of each axis and the mechanism equation are used to standardize the training process of the physical field solution model; subsequently, based on the stress field of each axis of the machine tool, the thermal-mechanical errors of each axis of the machine tool are determined respectively, and finally, combined with the forward motion chain model of the machine tool, the forward transfer model of the thermal-mechanical errors of the machine tool can be determined to realize the intelligent inference of the machining accuracy of the machine tool; and when the solution error of the physical field solution model is greater than the preset error threshold, the iteration unit retains some network parameters based on the weight sharing mechanism, and iteratively independently learns and fine-tunes the model parameters according to the real-time data.

[0081] In the above solution, this application fully captures the physical field change characteristics of different levels and ranges of the sensing data through a multi-scale graph attention network, and introduces thermal and stress mechanism equations to participate in the training process of the model, realizing the efficient solution of the interpretable physical field. When the solution error is greater than the threshold, through the online iterative update method, the problem that the traditional pre-trained model is difficult to respond to the working condition changes in real time is effectively overcome, and the intelligent inference of the machining accuracy of the machine tool under dynamic complex working conditions is realized.

[0082] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0083] In this article, specific examples are used to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, based on the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. An iterative physical embedded graph network driven intelligent reasoning method for machine tool machining accuracy, characterized in that: include: Acquire temperature sensing data and stress sensing data of each axis of the machine tool; the axes of the machine tool include the main axis and the feed axis; Using a pre-trained physical field solution model, the stress field of each axis of the machine tool is solved according to the temperature sensor data and stress sensor data of each axis of the machine tool; the physical field solution model is a model based on a physical embedding graph network; the physical field solution model uses a multi-scale graph attention network to capture the sensor features of different scales of each axis of the machine tool, and through a physical embedding loss function, the real-time temperature sensor data of each axis of the machine tool, the real-time stress sensor data of each axis of the machine tool and the mechanism equation are used to standardize the training process of the physical field solution model; Based on the stress field of each axis of the machine tool, the strain field of each axis of the machine tool is first determined, and then the thermal-mechanical error of each axis of the machine tool is determined respectively; According to the thermal-mechanical errors of each axis of the machine tool and the forward kinematic chain model of the machine tool, a forward transfer model of the thermal-mechanical errors of the machine tool is determined; the forward transfer model of the thermal-mechanical errors of the machine tool includes the thermal-mechanical errors of the spindle and each feed axis between the spindle of the machine tool and the tool at the end of the machine tool, and realizes the intelligent reasoning of the machining accuracy of the machine tool; according to the thermal-mechanical errors of each axis of the machine tool and the forward kinematic chain model of the machine tool, a forward transfer model of the thermal-mechanical errors of the machine tool is determined, which specifically includes: based on the transfer relationship of each axis of the machine tool, a forward kinematic chain model of the machine tool is established by using a homogeneous transformation matrix; based on the differential kinematic theory, the thermal-mechanical errors of each axis of the machine tool are converted into corresponding differential expressions; the differential expressions of the thermal-mechanical errors of each axis of the machine tool are integrated with the forward kinematic chain model of the machine tool to obtain the forward transfer model of the thermal-mechanical errors of the machine tool; When the solution error of the physical field solution model is greater than a preset error threshold, the parameters of the physical field solution model are locally iteratively adjusted based on a weight sharing mechanism according to real-time sensor data through an iteration unit until the solution error of the physical field solution model is less than the preset error threshold; the solution error of the physical field solution model is the root mean square error value between the real-time stress sensor data of any axis of the machine tool and the value at the corresponding node of the stress field of the axis obtained by solving the physical field solution model; the parameters of the physical field solution model are locally iteratively adjusted based on a weight sharing mechanism according to real-time sensor data through an iteration unit, specifically including: Retaining some network parameters in the physical field solution model based on a weight sharing mechanism, and reshaping the network connection relationship between the physical field solution model and the physical field solution model through an iterative unit; Based on real-time temperature sensing data or stress sensing data, the parameters of the physical field solution model are independently learned and fine-tuned to correct the error of the physical field solution model under specific working conditions.

2. The iterative physical embedded graph network driven intelligent reasoning method for machine tool machining accuracy according to claim 1 is characterized in that: The pre-trained physics field solving model is obtained through the following process: Acquire temperature sensing data of each axis of the machine tool and stress sensing data of each axis of the machine tool; Based on the temperature sensing data of each axis of the machine tool, the stress sensing data of each axis of the machine tool and the mechanism equation, the actual stress field of each axis of the machine tool is solved; The temperature sensor data of each axis of the machine tool and the stress sensor data of each axis of the machine tool are used as the input of the physical field solution model, and the actual stress field of each axis of the machine tool is used as the target output of the physical field solution model. The physical field solution model is trained to obtain a pre-trained physical field solution model.

3. The iterative physical embedded graph network driven intelligent reasoning method for machine tool machining accuracy according to claim 2 is characterized in that: When training the physical field solution model, a mechanism-data loss dual-track traction mechanism capable of balancing data loss terms and mechanism loss terms is used to train the physical field solution model; the data loss term is the mean square error between the real real-time sensor data and the sensor data at the corresponding node of the solved physical field; the mechanism loss term is the heat conduction equation, the thermal stress equation, the stress-strain equation and the stress transfer equation, which are used to constrain the training process of the physical field solution model; the physical field solution model is trained using the mechanism-data loss dual-track traction mechanism, specifically: Converting the data loss and the mechanism loss through an embedding layer to obtain a data loss vector and a mechanism loss vector; The dot product attention mechanism is used to calculate the correlation between the data loss vector and the mechanism loss vector, and the correlation score is normalized to obtain the weight of the data loss term and the weight of the mechanism loss term; The total loss is calculated by combining the weighted loss of the data loss term, the weighted loss of the mechanism loss term, the residual penalty and the penalty term for small weights; A dynamic factor is used to adjust the weight of the data loss term and the weight of the mechanism loss term according to the change of the total loss, so as to balance the contribution between the mechanism loss and the data loss of the physical field solution model, thereby improving the stability and accuracy of the physical field solution model during training.

4. The iterative physical embedded graph network driven intelligent reasoning method for machine tool machining accuracy according to claim 2 is characterized in that: Based on the temperature sensor data of each axis of the machine tool, the stress sensor data of each axis of the machine tool and the mechanism equation, the actual stress field of each axis of the machine tool is solved, including: Based on the temperature sensor data of each axis of the machine tool and the heat conduction equation, the temperature field of each axis of the machine tool is solved; Based on the temperature field and temperature-thermal strain equation of each axis of the machine tool, the thermal strain load of each axis of the machine tool is calculated; Based on the thermal strain load of each axis of the machine tool, stress sensing data, stress-strain equation and stress transfer equation, the stress field of each axis of the machine tool is solved.

5. The iterative physical embedded graph network driven intelligent reasoning method for machine tool machining accuracy according to claim 1 is characterized in that: The multi-scale graph attention network gradually captures the sensor features of different scales of the sensor data of each axis of the machine tool from local to global through a multi-layer graph convolution operation that is expanded layer by layer; After each layer of graph convolution operation, the multi-head attention mechanism is used to weightedly adjust the information propagation intensity between sensor nodes and further strengthen the fusion of sensing features at different scales.

6. The iterative physical embedded graph network driven intelligent reasoning method for machine tool machining accuracy according to claim 1 is characterized in that: Based on the stress field of each axis of the machine tool, the thermal-mechanical error of each axis of the machine tool is determined separately, including: Based on the stress field of each axis of the machine tool, the strain field of each axis of the machine tool is determined respectively; Based on the strain field of each axis of the machine tool, the thermal-mechanical error of each axis of the machine tool is determined respectively; the thermal-mechanical error includes the linear displacement of each axis of the machine tool and the angular deviation of each axis of the machine tool.

Images