Data-driven modeling method for machine tool thermal error caused by environment temperature change
By combining data-driven modeling methods based on the analysis of thermal error formation mechanisms, and utilizing multi-layer neural networks and sensor data optimization, the problem of insufficient interpretability and predictive ability of thermal error models for precision CNC machine tools is solved, and efficient thermal error compensation is achieved.
Patent Information
- Application Number
- CN202411404587.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-10
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-10-10
AI Technical Summary
Existing technologies lack data-driven modeling methods that incorporate thermal error formation mechanism analysis in precision CNC machine tools, resulting in insufficient model interpretability and predictive ability, which affects the thermal error compensation effect.
By analyzing the formation mechanism of machine tool thermal errors caused by changes in ambient temperature, a data-driven modeling method is designed. Using changes in ambient temperature, machine tool surface temperature, and internal temperature, a multi-layer LSTM and GRU neural network combined with a fully connected layer and a multi-head attention mechanism is used to construct a data-driven sub-model. Data is collected by temperature sensors and displacement sensors to optimize the model.
It improves the interpretability and generalization ability of data-driven models, simplifies model input requirements (requiring only an ambient temperature sensor), and enhances prediction accuracy with prediction residuals around ±1μm, significantly outperforming traditional methods.
Smart Images

Figure CN118917219A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of error compensation of precision numerically controlled machine tools, and in particular relates to a data-driven modeling method for thermal errors of machine tools caused by changes in ambient temperature. Background Art
[0002] Precision CNC machine tools are high-end manufacturing equipment widely used in aviation, aerospace, shipbuilding, new energy vehicles and other industries. Changes in ambient temperature cause thermal deformation of machine tools, which in turn causes deviations from the ideal position between the tool tip and the workpiece, forming thermal errors. Thermal error modeling is the core of thermal error compensation. At present, the main data-driven modeling method used is to directly establish the mapping relationship between ambient temperature and machine tool thermal error. There is a lack of data-driven modeling methods combined with thermal error mechanism analysis, which easily leads to insufficient model interpretability and predictive ability, affecting the subsequent thermal error compensation effect. Therefore, it is of great significance to invent a data-driven modeling method combined with thermal error formation mechanism analysis. Summary of the invention
[0003] The purpose of the present invention is to provide a data-driven modeling method for machine tool thermal errors caused by ambient temperature changes. The method analyzes the formation mechanism of machine tool thermal errors to design a modeling algorithm, and uses multi-physical quantity data such as ambient temperature, machine tool surface temperature, and machine tool thermal error to identify model parameters, thereby improving the interpretability and generalization ability of the data-driven model and laying the foundation for subsequent thermal error compensation applications.
[0004] To achieve the above purpose, the present invention adopts the following technical solution:
[0005] A data-driven modeling method for machine tool thermal errors caused by ambient temperature changes comprises the following steps:
[0006] Step 1: Analyze the formation mechanism of thermal error of precision CNC machine tools caused by ambient temperature changes; According to the physical principle that the naturally flowing air transfers heat to the machine tool surface by convection heat transfer, the functional relationship between the machine tool surface temperature and the ambient temperature is analyzed; Based on the physical principle that heat from the machine tool surface is transferred to the inside of the machine tool through heat conduction, the functional relationship between the temperature at any position inside the machine tool and the surface temperature is analyzed by considering the influence of internal thermal resistance; Divide the machine tool into n parts and analyze the functional relationship between the thermal deformation of the machine tool and the internal temperature changes of different parts;
[0007] Step 2: Establish a data-driven sub-model of ambient temperature change-machine tool surface temperature change; According to the analysis results of the functional relationship between the machine tool surface temperature and the ambient temperature, a neural network with memory capability is designed to construct a data-driven sub-model of ambient temperature variation-machine tool surface temperature variation; The data-driven sub-model of the ambient temperature variation-machine tool surface temperature variation includes multiple LSTM layers and fully connected layers connected in sequence;
[0008] Step 3: Establish a data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change, and the machine tool thermal deformation; According to the analysis results of the functional relationship between the temperature at any position inside the machine tool and the surface temperature and the functional relationship between the thermal deformation of the machine tool and the internal temperature change of different parts, a data-driven sub-model between the surface temperature change of the machine tool, the internal temperature change of the machine tool and the thermal deformation of the machine tool is designed; The data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change and the machine tool thermal deformation includes a multi-layer GRU layer, a multi-head attention mechanism layer and a fully connected layer connected in sequence;
[0009] Step 4: Machine tool thermal characteristic data collection; When the machine tool is in a stationary state, the temperature sensor and displacement sensor are used to collect the surface temperature data of the stationary machine tool, the temperature data of the natural flow environment near the machine tool, and the thermal displacement data during the working period at different ambient temperatures;
[0010] Step 5: Optimize the hyperparameters of the data-driven sub-model; The hyperparameters of the data-driven sub-model of ambient temperature variation-machine tool surface temperature variation are optimized using the machine tool surface temperature data and the natural flow environment temperature data. The machine tool surface temperature data and thermal displacement data are used to optimize the hyperparameters of the data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change, and the machine tool thermal deformation;
[0011] Step 6: Establish an ambient temperature-machine tool thermal error data-driven model with prior knowledge; The data-driven sub-model of ambient temperature change-machine tool surface temperature change after the hyperparameter optimization in step 5 and the data-driven sub-model of machine tool surface temperature change-machine tool internal temperature change-machine tool thermal deformation are connected in series to form an ambient temperature-machine tool thermal error data-driven model with prior knowledge.
[0012] Furthermore, the functional relationship between the surface temperature of the machine tool and the ambient temperature in step 1 is established using the lumped heat capacity method, and the functional relationship is: , In the formula, is the surface temperature of the machine tool; It’s time; is the thermal time constant of the machine tool; is the maximum change amplitude of the ambient temperature; is the frequency of ambient temperature change; is the lagging phase; is the initial value of the ambient temperature.
[0013] Furthermore, the functional relationship between the temperature at any position inside the machine tool and the surface temperature in step 1 is: , In the formula, Any position of the machine tool The internal temperature of the location; It’s time; is the maximum variation of the machine tool surface temperature; is the frequency of temperature change on the machine tool surface; is the thermal diffusivity of the machine tool; It is the initial value of the machine surface temperature.
[0014] Furthermore, the functional relationship between the thermal deformation of the machine tool in step 1 and the internal temperature change of different parts is: , In the formula, is the total thermal deformation of the machine tool, i.e. thermal error; is the weight coefficient, which indicates the contribution of temperature changes of different parts of the machine tool to the total thermal deformation; is the comprehensive thermal deformation coefficient; It is the first The change in the internal temperature of the part compared to the initial moment reflects the The average temperature change information of the part.
[0015] Furthermore, the surface temperature data of the machine tool in step 4 is collected and obtained by temperature sensors installed near the top of the machine tool spindle box, the bottom of the spindle box, the top of the column, and the center of the bottom of the column; The natural flow environment temperature data is collected and obtained by a temperature sensor installed on the top of the machine tool protective cover; The thermal displacement data is collected by an eddy current sensor installed along the radial / axial direction on the machine tool spindle.
[0016] Furthermore, in step 5, a GA genetic optimization algorithm is used to optimize the number of LSTM layers and the number of hidden units in the data-driven sub-model of ambient temperature change-machine tool surface temperature change, and the number of GRU layers, the number of hidden units in the GRU layer, and the number of heads in the multi-head attention mechanism layer in the data-driven sub-model between machine tool surface temperature change-machine tool internal temperature change-machine tool thermal deformation.
[0017] Furthermore, the ambient temperature-machine tool thermal error data driven model described in step 6 adopts fine-tuning technology, takes the ambient temperature change value as input, and the machine tool radial / axial thermal displacement data as output, iterates, optimizes the model parameters, and finally forms the ambient temperature-machine tool thermal error data driven model.
[0018] The present invention has the following beneficial effects:
[0019] (1) The data-driven model is designed by analyzing the formation mechanism of machine tool thermal errors. The data-driven model includes a data-driven sub-model of ambient temperature change-machine tool surface temperature change and a data-driven sub-model of machine tool surface temperature change-machine tool internal temperature change-machine tool thermal deformation. This improves the interpretability and generalization ability of the data model. A data-driven model of ambient temperature-machine tool thermal error is established, which only requires data from one ambient temperature sensor as input. The model is simple and efficient.
[0020] (2) Based on the analysis results of the functional relationship between the machine tool surface temperature and the ambient temperature, a neural network with memory capability is designed to construct a data-driven sub-model of the ambient temperature change-machine tool surface temperature change to reveal the convective heat transfer process of the machine tool.
[0021] (3) Based on the analysis results of the functional relationship between the temperature at any position inside the machine tool and the surface temperature and the functional relationship between the thermal deformation of the machine tool and the internal temperature change of different parts, a data-driven sub-model of the machine tool surface temperature change-machine tool internal temperature change-machine tool thermal deformation is designed. The temperature changes of multiple machine tool surfaces are used as input, and a GRU neuron with memory capability is used to control the information flow through reset gates and update gates. A multi-head attention mechanism is used to extract the internal temperature features of the machine tool, and a fully connected layer is added to integrate the extracted temperature features. Finally, the thermal deformation of the machine tool is output by multiplying them by different coefficients.
[0022] (4) Compared with the traditional data-driven modeling algorithm, the present invention utilizes ambient temperature data, machine tool surface temperature data and machine tool thermal error data, and considers the knowledge of the mechanism of machine tool thermal error caused by ambient temperature changes, so as to improve the interpretability and generalization ability of the data-driven model and lay the foundation for the subsequent thermal error compensation application. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 It is a schematic diagram of the method flow of the present invention.
[0024] Figure 2 It is a schematic diagram of the data driven sub-model of the ambient temperature variation-machine tool surface temperature variation of the present invention.
[0025] Figure 3 It is a schematic diagram of the data-driven sub-model among the machine tool surface temperature variation, the machine tool internal temperature variation and the machine tool thermal deformation according to the present invention.
[0026] Figure 4 It is a schematic diagram of the prediction results of the present invention and the comparative model.
[0027] Figure 5 Schematic diagram of prediction residuals of the present invention and the comparison model.
[0028] Figure 6 Schematic diagram of MAE and RMSE of the present invention and the comparison model. DETAILED DESCRIPTION
[0029] like Figure 1 As shown, this embodiment provides a data-driven modeling method for thermal errors of precision CNC machine tools caused by ambient temperature changes, comprising the following steps:
[0030] Step 1: Analyze the mechanism of thermal error of precision CNC machine tools caused by ambient temperature changes; According to the physical principle that the naturally flowing air transfers heat to the machine tool surface by convection heat transfer, the functional relationship between the machine tool surface temperature and the ambient temperature is analyzed based on the lumped heat capacity method. , In the formula, is the surface temperature of the machine tool; It’s time; is the thermal time constant of the machine tool; is the maximum change amplitude of the ambient temperature; is the frequency of ambient temperature change; is the lagging phase; is the initial value of the ambient temperature.
[0031] According to the physical principle that heat from the machine tool surface is transferred to the inside of the machine tool through heat conduction, the influence of internal thermal resistance is considered and the heat at any position inside the machine tool is analyzed. The functional relationship between the temperature at the surface and the temperature at the , In the formula, Any position of the machine tool The internal temperature of the location; It’s time; is the maximum variation of the machine tool surface temperature; is the frequency of temperature change on the machine tool surface; is the thermal diffusivity of the machine tool; It is the initial value of the machine surface temperature.
[0032] Divide machine tools into Part and analyze the functional relationship between the thermal deformation of the machine tool and the internal temperature changes of different parts; , In the formula, is the total thermal deformation of the machine tool, i.e. thermal error; is the weight coefficient, which indicates the contribution of temperature changes of different parts of the machine tool to the total thermal deformation; is the comprehensive thermal deformation coefficient; It is the first The change in the internal temperature of the part compared to the initial moment reflects the The average temperature change information of the part.
[0033] Step 2: Establish a data-driven sub-model of ambient temperature change-machine tool surface temperature change; The temperature field information of the machine tool is reflected by measuring the surface temperature of the machine tool; based on the analysis results of the functional relationship between the surface temperature of the machine tool and the ambient temperature, a neural network with memory ability is designed to construct a data-driven sub-model of the ambient temperature change-machine tool surface temperature change, such as Figure 2 As shown, it represents the physical process of convective heat transfer in machine tools.
[0034] This embodiment uses LSTM neurons with input gate, forget gate, output gate, candidate cell state, and cell state mechanism to build a neural network; Input Gate ; Forget Gate ; Output Gate ; Candidate cell states ; Cell state ; Hidden State ; in, , , , , , They are the current input gate, forget gate, output gate, candidate cell state, cell state, and hidden state; , , , , , , , are all weight parameter matrices; 、b f , , is the bias parameter vector; It is the hidden state at the previous moment; represents the sigmoid activation function; tanh is the hyperbolic tangent function; Represents the Hadamard product operation, which is an operation method for multiplying the elements at corresponding positions of two matrices of the same dimension to form a new matrix.
[0035] A fully connected layer is added to the LSTM layer to integrate the features, improve the stability and accuracy of the prediction, and establish a data-driven sub-model of ambient temperature change-machine tool surface temperature change to reveal the convection heat transfer process of the machine tool.
[0036] Step 3: Establish a data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change, and the machine tool thermal deformation; The machine tool has a large size in space, and the surface temperature change of the machine tool cannot fully reflect the temperature field change information of the whole machine tool. Based on the analysis results of the functional relationship between the temperature at any position x inside the machine tool and the surface temperature and the functional relationship between the thermal deformation of the machine tool and the internal temperature change of different parts, a data-driven sub-model between the surface temperature change of the machine tool, the internal temperature change of the machine tool, and the thermal deformation of the machine tool is designed, such as Figure 3 shown.
[0037] The temperature changes of multiple surfaces of the machine tool are used as input, and the GRU neurons with memory capability are used. The GRU neurons have a reset gate, an update gate, a hidden state, and a candidate hidden state. The reset gate and the update gate are used to control the flow of information and calculate the output of the neural network. Reset Gate , Update Gate , Hidden State , Candidate hidden states , in, , , , They are the reset gate, update gate, hidden state, and candidate hidden state at the current moment; , , , , , is the weight matrix; , , is the bias vector; is the hidden state value at the previous moment, represents the sigmoid activation function; tanh is the hyperbolic tangent function; Represents the Hadamard product operation, which is an operation method for multiplying the elements at corresponding positions of two matrices of the same dimension to form a new matrix.
[0038] The temperature field changes in different parts of the machine tool contribute differently to the thermal deformation of the whole machine. The multi-head attention mechanism is used to extract the key internal temperature features. , Where Q, K, V are query, key, and value matrices; is the output of the ith attention head, ; Indicates the feature concatenation and fusion of the output results of different self-attention heads; It is a transformation matrix that transforms the fusion output of multi-head attention to be consistent with the dimension of the initial value matrix.
[0039] A fully connected layer is added after the multi-head attention mechanism to integrate the extracted key temperature features; then, the features are multiplied by different coefficients for output, the thermal deformation of each part is calculated and summed up, and finally a sub-model driven by machine tool surface temperature change, machine tool internal temperature change and machine tool thermal deformation data is constructed.
[0040] Step 4: Machine tool thermal characteristic data collection; When the machine tool is in a stationary state, four pt100 temperature sensors are installed near the top of the machine tool spindle box, the bottom of the spindle box, the top of the column, and the center of the bottom of the column to collect the surface temperature data of the machine tool.
[0041] A temperature sensor is installed on the top of the machine tool protective cover to collect natural flow environment temperature data.
[0042] An eddy current sensor for collecting radial / axial thermal displacement data of the machine tool is installed on the machine tool spindle in the radial / axial direction.
[0043] The thermal characteristics experiment of machine tools was carried out during the daytime working hours with different ambient temperature rise trends. The surface temperature data of machine tools, natural flow environment temperature data and thermal displacement data were obtained for data-driven model parameter training and performance testing.
[0044] Step 5: Optimize the key hyperparameters in the data-driven sub-model established in steps 2 and 3; Using the ambient temperature data and machine tool surface temperature data collected by the temperature sensor in step 4, the GA genetic optimization algorithm is used to optimize the hyperparameters such as the number of LSTM layers and the number of hidden units in the data-driven sub-model of ambient temperature change-machine tool surface temperature change; Using the machine tool surface temperature data collected by the temperature sensor in step 4 and the thermal displacement data collected by the eddy current sensor, the GA genetic optimization algorithm is used to optimize the hyperparameters such as the number of GRU layers, the number of hidden units in the GRU layer, and the number of attention mechanism heads in the data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change, and the machine tool thermal deformation.
[0045] Step 6: Establish an ambient temperature-machine tool thermal error data-driven model with prior knowledge; The data-driven sub-model of ambient temperature change-machine tool surface temperature change after the hyperparameter optimization in step 5 and the data-driven sub-model of machine tool surface temperature change-machine tool internal temperature change-machine tool thermal deformation are connected in series to form an ambient temperature-machine tool thermal error data-driven model with prior knowledge.
[0046] The fine-tuning technology is adopted, with the ambient temperature change value as input and the radial / axial thermal displacement data of the machine tool as output. Ten small iterations are performed to optimize the entire model parameters, and finally the ambient temperature-machine tool thermal error data driven model, referred to as NDDTEM, is formed.
[0047] The machine tool thermal characteristics experimental method in step 4 is used to re-collect different ambient temperatures and machine tool radial thermal displacement data, and the NDDTEM modeled in this embodiment is used to perform predictions with the LSTM, GRU, BPNN, and SVR constructed by the traditional modeling method. The prediction results are as follows: Figure 4 As shown, the NDDTEM prediction effect proposed in this embodiment is the best, and the prediction residual is about ±1μm. Figure 5 As shown; the mean absolute error MAE and root mean square error RMSE of each model, as shown Figure 6 As shown, the results show that the MAE and RMSE of the NDDTEM data-driven model are only 0.76μm and 0.91μm, which are reduced by 55%, 32%, 121%, and 55% and 78%, 29%, 131%, and 74% of the MAE and 78%, 29%, 131%, and 74% of the RMSE, respectively, compared with the LSTM, GRU, BP, and SVR models built based on traditional data modeling methods, verifying the effectiveness and superiority of the data-driven modeling method proposed in this embodiment.
[0048] The above description is only a preferred implementation manner of the present invention, but the protection scope of the present invention is not limited thereto, and any modification and replacement based on the technical solution and inventive concept provided by the present invention should be included in the protection scope of the present invention.
Claims
1. A data-driven modeling method for machine tool thermal errors caused by ambient temperature changes. It is characterized in that The steps include: Step 1: Analyze the formation mechanism of thermal error of precision CNC machine tools caused by ambient temperature changes; According to the physical principle that the naturally flowing air transfers heat to the machine tool surface by convection heat transfer, the functional relationship between the machine tool surface temperature and the ambient temperature is analyzed; Based on the physical principle that heat from the machine tool surface is transferred to the inside of the machine tool through heat conduction, the functional relationship between the temperature at any position inside the machine tool and the surface temperature is analyzed by considering the influence of internal thermal resistance; Divide the machine tool into n parts and analyze the functional relationship between the thermal deformation of the machine tool and the internal temperature changes of different parts; Step 2: Establish a data-driven sub-model of ambient temperature change-machine tool surface temperature change; According to the analysis results of the functional relationship between the machine tool surface temperature and the ambient temperature, a neural network with memory capability is designed to construct a data-driven sub-model of ambient temperature variation-machine tool surface temperature variation; The data-driven sub-model of the ambient temperature variation-machine tool surface temperature variation includes multiple LSTM layers and fully connected layers connected in sequence; Step 3: Establish a data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change, and the machine tool thermal deformation; According to the analysis results of the functional relationship between the temperature at any position inside the machine tool and the surface temperature and the functional relationship between the thermal deformation of the machine tool and the internal temperature change of different parts, a data-driven sub-model between the surface temperature change of the machine tool, the internal temperature change of the machine tool and the thermal deformation of the machine tool is designed; The data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change and the machine tool thermal deformation includes a multi-layer GRU layer, a multi-head attention mechanism layer and a fully connected layer connected in sequence; Step 4: Machine tool thermal characteristic data collection; When the machine tool is in a stationary state, the temperature sensor and displacement sensor are used to collect the surface temperature data of the stationary machine tool, the temperature data of the natural flow environment near the machine tool, and the thermal displacement data during the working period at different ambient temperatures; Step 5: Optimize the hyperparameters of the data-driven sub-model; The hyperparameters of the data-driven sub-model of ambient temperature variation-machine tool surface temperature variation are optimized using the machine tool surface temperature data and the natural flow environment temperature data. The machine tool surface temperature data and thermal displacement data are used to optimize the hyperparameters of the data-driven sub-model between the machine tool surface temperature change, the machine tool internal temperature change, and the machine tool thermal deformation; Step 6: Establish an ambient temperature-machine tool thermal error data-driven model with prior knowledge; The data-driven sub-model of ambient temperature change-machine tool surface temperature change after the hyperparameter optimization in step 5 and the data-driven sub-model of machine tool surface temperature change-machine tool internal temperature change-machine tool thermal deformation are connected in series to form an ambient temperature-machine tool thermal error data-driven model with prior knowledge.
2. According to claim 1, a data-driven modeling method for machine tool thermal errors caused by ambient temperature changes, It is characterized in that The functional relationship between the machine tool surface temperature and the ambient temperature described in step 1 is established using the lumped heat capacity method, and the functional relationship is: , In the formula, is the surface temperature of the machine tool; It’s time; is the thermal time constant of the machine tool; is the maximum change amplitude of the ambient temperature; is the frequency of ambient temperature change; is the lagging phase; is the initial value of the ambient temperature.
3. According to claim 1, a data-driven modeling method for machine tool thermal errors caused by ambient temperature changes, It is characterized in that The functional relationship between the temperature at any position inside the machine tool and the surface temperature in step 1 is: , In the formula, Any position of the machine tool The internal temperature of the location; It’s time; is the maximum variation of the machine tool surface temperature; is the frequency of temperature change on the machine tool surface; is the thermal diffusivity of the machine tool; It is the initial value of the machine surface temperature.
4. The method for data-driven modeling of machine tool thermal errors caused by ambient temperature changes according to claim 3, It is characterized in that The functional relationship between the thermal deformation of the machine tool and the internal temperature change of different parts in step 1 is: , In the formula, is the total thermal deformation of the machine tool, i.e. thermal error; is the weight coefficient, which indicates the contribution of temperature changes of different parts of the machine tool to the total thermal deformation; is the comprehensive thermal deformation coefficient; It is the first The change in the internal temperature of the part compared to the initial moment reflects the The average temperature change information of the part.
5. The data-driven modeling method for machine tool thermal errors caused by ambient temperature changes according to claim 1, It is characterized in that The surface temperature data of the machine tool in step 4 is collected and obtained by temperature sensors installed near the top of the machine tool spindle box, the bottom of the spindle box, the top of the column, and the center of the bottom of the column; The natural flow environment temperature data is collected and obtained by a temperature sensor installed on the top of the machine tool protective cover; The thermal displacement data is collected by an eddy current sensor installed along the radial / axial direction on the machine tool spindle.
6. The method for data-driven modeling of thermal errors of machine tools caused by changes in ambient temperature according to claim 1, It is characterized in that In step 5, the GA genetic optimization algorithm is used to optimize the number of LSTM layers and the number of hidden units in the data-driven sub-model of ambient temperature change-machine tool surface temperature change, as well as the number of GRU layers, the number of hidden units in the GRU layer, and the number of heads in the multi-head attention mechanism layer in the data-driven sub-model between machine tool surface temperature change-machine tool internal temperature change-machine tool thermal deformation.
7. The method for data-driven modeling of machine tool thermal errors caused by ambient temperature changes according to claim 1, It is characterized in that The ambient temperature-machine tool thermal error data driven model described in step 6 adopts fine-tuning technology, takes the ambient temperature change value as input, and the machine tool radial / axial thermal displacement data as output, iterates, optimizes the model parameters, and finally forms the ambient temperature-machine tool thermal error data driven model.
8. An ambient temperature-machine tool thermal error data-driven model established according to a method for data-driven modeling of machine tool thermal errors caused by ambient temperature changes as described in any one of claims 1 to 7.
Citation Information
Patent Citations
Selection optimization method of numerically-controlled machine tool thermal error compensation modeling temperature measuring point combination
CN103926874A
Thermal error modeling method of numerical control machine tool based on two-step method
CN109240204A
Numerical control machine tool spindle axial thermal error physical modeling method
CN113126566A
CNN-GRU-based combined neural network machine tool feeding system thermal error prediction method
CN116540632A
Cited By
Numerical control machine tool spindle thermal error dynamic compensation model and system
CN118963241A
A dynamic compensation model and system for thermal error of CNC machine tool spindle
CN118963241B
Longitudinal cutting automatic lathe thermal error modeling method based on single temperature sensor
CN120630874A
A single-temperature-sensor-based thermal error modeling method for longitudinal cutting automatic lathe
CN120630874B
