Machining Time Prediction Using Machine-Specific Delay Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machining time prediction technologies fail to accurately account for machine delays due to servo control and movement, leading to inaccuracies in predicting machining time, especially at the initial stages of machining, and are not adaptable to different machine characteristics.
Innovation Solution
A machine learning device that uses supervised learning to construct a model based on data from test machining, including movement amount, program commands, machining speed, and workpiece weight, to predict delay times due to servo control and machine movement, which are then used to correct conventional machining time predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional machining time prediction methods are used, then the prediction process is simple, but the prediction accuracy is low due to not accounting for machine delays
Solution Approach 1:
The patent replaces conventional mechanical calculation methods for machining time prediction with a machine learning-based prediction system. The learning unit uses supervised learning to train a prediction model that automatically accounts for machine delays, servo control characteristics, and movement properties, substituting complex manual calculations and simulations with an intelligent system that achieves higher accuracy without requiring users to understand the underlying complexity.
Solution Approach 2:
The patent introduces a learning unit as an intermediary component between the machining system and the prediction process. This learning unit accumulates actual machining data, learns machine-specific delay characteristics, and provides corrected predictions to the machining time prediction device, acting as a mediator that bridges the gap between conventional prediction methods and accurate delay-compensated predictions.
2Adaptability or versatility
If general prediction models are used, then the model is universally applicable, but the prediction accuracy varies across different machine types
Solution Approach 1:
The patent implements preliminary learning through the learning unit that accumulates actual machining data from each specific machine before making accurate predictions. The system performs preliminary data collection and model training for each machine type, storing learned characteristics in advance. This preliminary action enables the prediction device to adapt to different machine types while maintaining high accuracy, as each machine's unique delay characteristics are learned and stored beforehand.
Solution Approach 2:
The patent creates a dynamic prediction system where the learning unit continuously accumulates data and updates the prediction model based on actual machining results from different machine types. The system adapts dynamically to each machine's characteristics through supervised learning, allowing the same prediction device to accurately predict machining times across various machine types by adjusting to their specific delay patterns.
Data Source
AI summary
A machine learning device acquires from a numerical controller information relating to machining when the machining is performed, and further acquires an actual delay time due to servo control and due to machine movement which are caused in the machining when the machining is performed. Then, the device performs supervised learning using the acquired machining-related information as input data, and using the acquired actual delay time due to servo control and due to machine movement as supervised data, and constructs a learning model, thereby predicting the machine delay time caused in a machine with high precision.