Dynamic Weighing Time Window Selection for Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dynamic weighing in belt weighing devices is prone to measurement inaccuracies due to speed dependencies, pack length, and systematic errors like resonances and vibrations, requiring time-consuming teach-in processes to determine static weight values.
Innovation Solution
The method involves recording multiple dynamic weight profiles during teach-in mode, defining time windows based on comparisons to select consistent measurement values and exclude error-prone periods, thereby improving measurement accuracy by focusing on reproducible data segments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple dynamic weight profiles are recorded and compared to define time windows for selecting measurement values, then measurement accuracy is improved, but the teach-in process becomes more complex
Solution Approach 1:
The dynamic weight profile is segmented into multiple time windows, where each window corresponds to a specific phase of the weighing process. By dividing the continuous weight signal into discrete segments (approach, weighing, departure phases), the system can selectively analyze and combine measurements from different segments across multiple profiles, thereby improving measurement accuracy through targeted data selection while managing complexity through structured organization.
Solution Approach 2:
Instead of using all available measurement data from dynamic weight profiles, the method selectively uses only the measurements that fall within the defined time windows. This partial action approach filters out measurements taken during transient phases (when packs are entering or leaving the weighing zone) and retains only those from the stable weighing phase, improving accuracy by excluding erroneous data points without requiring complete re-measurement.
2Measurement precision
If multiple packs are weighed several times dynamically in a teach-in process, then measurement accuracy is improved, but the time required for teaching increases
Solution Approach 1:
The system performs preliminary analysis of dynamic weight profiles during the teach-in process to automatically identify the optimal time windows for each pack weighing event. By pre-defining these windows based on the characteristic shape and timing of weight curves, the system eliminates the need for repeated manual adjustments and multiple weighing cycles, thereby reducing teach-in time while maintaining measurement accuracy through consistent selection of valid measurement periods.
Solution Approach 2:
The teach-in process is designed to be self-adjusting by automatically analyzing the recorded dynamic weight profiles and determining the appropriate time windows without requiring extensive manual intervention. The system uses the actual measurement data from the profiles to identify patterns and set the windows autonomously, reducing both the time required and the complexity of the teaching operation while ensuring accurate measurement selection.
3Productivity
If measurements are taken during dynamic weighing, then throughput is increased, but systematic errors and vibrations affect measurement accuracy
Solution Approach 1:
The method extracts and isolates the valid measurement portions from the dynamic weight profiles by defining specific time windows that correspond to the stable weighing phase. By separating the useful measurement data from the contaminated transient phases (where systematic errors and vibrations occur during pack entry and departure), the system maintains high throughput dynamic weighing while improving accuracy through selective extraction of clean measurement signals.
Solution Approach 2:
The system converts the harmful effect of dynamic weighing (where motion-induced errors contaminate measurements) into a benefit by using the temporal structure of the weight profiles to identify and exploit the brief periods when measurements are accurate. The very act of dynamic weighing, which creates the problem, also creates the characteristic weight curve shape that allows automatic identification of valid measurement windows, turning the transient nature of dynamic weighing into a tool for selective data acquisition.
Data Source
Figure 1~2
AI summary
The invention relates to a method for generating a teach-in data set for determining a dynamic weight value from a dynamic weight profile recorded in a normal operating mode of a belt weighing device, in particular a dynamic checkweigher, during the dynamic weighing of an article, wherein in a teach-in mode several dynamic weight profiles, each consisting of individual measurements, are recorded successively and, based on a comparison of the several weight profiles with each other, at least one time window is defined which either determines which of the individual measurements of a dynamic weight profile recorded in the normal operating mode are used as the basis for determining a dynamic weight value, or conversely determines which of these individual measurements are not used as the basis for this determination, wherein the time window(s) are stored in the teach-in data set.