Priority-Based Data Collection for Machine Tools
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data collection methods for machine tools face network congestion and high costs due to large data volumes, particularly when sampling periods are short, leading to transmission delays and data loss, and require large-capacity storage devices for buffering.
Innovation Solution
A data collection device that calculates a relative priority for each machine tool based on its state, allowing for selective data collection and limiting the number of simultaneous targets to prevent network congestion without the need for a large-capacity storage device, focusing on real-time data collection and prioritizing machines with higher estimated errors or precision demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If measured data is collected at a short sampling period, then measurement precision is improved, but network congestion occurs and data transmission reliability deteriorates
Solution Approach 1:
The patent segments the data collection process by dividing machine tools into multiple priority groups. Instead of collecting data from all machines simultaneously at high frequency, the system divides them into first priority machines (collected at normal high frequency) and second priority machines (collected at reduced frequency or skipped), thereby segmenting the data flow to prevent network congestion while maintaining measurement precision for critical machines.
Solution Approach 2:
The system dynamically changes the sampling parameter based on priority classification. For first priority machine tools, the sampling period remains short to maintain high measurement precision. For second priority machine tools, the sampling period is extended or data collection is skipped, effectively changing the parameter to reduce overall data volume and prevent network congestion.
2Reliability
If a storage device with large storage capacity is provided, then data collection reliability is improved, but device cost increases
Solution Approach 1:
The patent extracts only the necessary portion of measured data for collection based on machine tool priority. Instead of buffering all data from all machines (which would require large storage capacity), the system identifies and extracts only the data from first priority machines that needs immediate collection, while reducing or eliminating storage requirements for second priority machines.
Solution Approach 2:
The system applies partial action by collecting data from only a subset of machine tools (first priority) at the normal sampling rate, while applying reduced action or no action to second priority machines. This partial data collection approach maintains reliability for critical machines without requiring storage capacity for all machines.
3Productivity
If data is collected from all machine tools simultaneously, then productivity is improved, but network congestion occurs
Solution Approach 1:
The patent introduces dynamics into the data collection system by making the collection rate adaptive rather than static. The system dynamically adjusts the data collection rate for different machine tools based on their priority classification and current network conditions. First priority machines maintain high collection rates while second priority machines have reduced rates, creating a dynamic balance between productivity and network load.
Solution Approach 2:
The system implements periodic action by collecting data from first priority machines at regular short intervals while collecting data from second priority machines at extended intervals or skipping collections periodically. This periodic differentiation maintains overall productivity for critical machines while reducing total network traffic to prevent congestion.
Data Source
AI summary
To reliably collect measured data without requiring a large-capacity storage device. A data collection device which is connected with a plurality of machine tools via a network includes: a priority calculation means for calculating a degree of relative priority for each of the plurality of machines tools, based on information related to a state of each of the machine tools; and a measured data collection means for deciding, in a case of receiving a collection request of measured data for the machine tool, which machine tool among the machine tools corresponding to the collection request to set as a collection target of the measured data based on the degree of relative priority, and for collecting the measured data from the machine tool decided as the collection target via the network.