Networked Thermal Displacement Correction for Machine Tool Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current thermal displacement correction devices in industrial machines lack sufficient calculation accuracy and require significant computing power and cost for implementing machine learning, making it difficult to update correction formulas and models effectively.
Innovation Solution
A thermal displacement correction system that utilizes a network-connected computer and thermal displacement correction device, where the computer acquires environmental and device data to calculate correction values using machine learning models, and outputs these values for execution by the device, allowing for distributed processing and flexible updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If machine learning is implemented in the thermal displacement correction device to improve calculation accuracy, then manufacturing precision is improved, but device complexity and computing power requirements increase significantly
Solution Approach 1:
The system divides the machine learning functionality into two separate components: a learning device that performs complex model training and computation, and a thermal displacement correction device that executes the trained model with reduced computational requirements. This segmentation allows high-accuracy machine learning to be achieved without overloading the correction device with excessive computing power requirements.
Solution Approach 2:
A network communication system acts as an intermediary between the learning device and the thermal displacement correction device. The learning device generates correction values through machine learning and transmits them via the network to the correction device, which then applies these values to compensate for thermal displacement. This intermediary approach enables accurate correction while keeping the correction device's computational burden manageable.
2Manufacturing precision
If correction formulas and models are updated to improve calculation accuracy, then manufacturing precision is improved, but development time and cost increase
Solution Approach 1:
The learning device performs preliminary machine learning model training and correction value calculation in advance, storing the results for later use. When thermal displacement correction is needed, the pre-computed correction values are directly applied without requiring time-consuming real-time calculations or model retraining, thus improving response speed while maintaining accuracy.
Solution Approach 2:
The system creates and updates correction models on the learning device separately, then copies the trained models and correction values to the thermal displacement correction device. This allows model updates to be performed independently on the learning device without disrupting the correction device's operations, reducing development time and enabling easier model maintenance.
3Manufacturing precision
If sophisticated correction formulas are implemented to improve manufacturing precision, then thermal displacement correction accuracy is improved, but ease of manufacture deteriorates due to high computing power requirements
Solution Approach 1:
The system segments the complex correction calculation process into model training (performed by the learning device) and model execution (performed by the correction device). This allows sophisticated correction formulas to be implemented without requiring the correction device to have high computing power, as the heavy computational work is offloaded to the learning device during the training phase.
Solution Approach 2:
The system replaces real-time mechanical computation in the correction device with pre-computed correction values transmitted via network communication. Instead of requiring the correction device to perform complex calculations locally, it receives ready-made correction values from the learning device, simplifying the correction device's hardware requirements and making implementation easier.
Data Source
AI summary
A thermal displacement correction system performs thermal displacement correction in cooperation with a thermal displacement correction device and a computer connected via a network. The thermal displacement correction system that corrects thermal displacement caused by processing performed by a machine comprises: the thermal displacement correction device connected to the machine; and the computer connected to the thermal displacement correction device via the network. The computer comprises: a data acquisition unit that acquires environmental data on an external environment of the machine via the network; a correction value inference unit that calculates a correction value using the environmental data; and a correction value output unit that outputs the correction value to the network. The thermal displacement correction device comprises: a correction value acquisition unit that acquires the correction value via the network; and a correction execution unit that performs thermal displacement correction using the correction value.