Crane Load Estimation via Machine-Specific Statistical Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the load on working machines, especially dynamic machines like cranes, are prone to interference from external influences and inaccuracies due to factors like rope deflection and weight, and often require costly or impractical sensor installations, particularly in harsh environments.
Innovation Solution
A method involving a two-phase process: a calibration phase where a machine-specific load model is developed using operational sequences and statistical analysis of available machine parameters, and an application phase where this model estimates load without prior knowledge of physics, utilizing existing sensors and potentially external sensors to account for environmental influences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct force measurement sensors are installed at the load capture point, then load measurement precision is improved, but device complexity and cost increase significantly
Solution Approach 1:
The patent uses machine parameters (acceleration, velocity, position from sensors already present on the machine) as intermediary variables to indirectly determine load, avoiding direct installation of force measurement sensors at the load capture point. The load model acts as a mediator that translates these intermediate parameters into load information.
Solution Approach 2:
The patent replaces the mechanical force measurement system (direct force sensors) with a computational model-based system that uses acceleration, velocity, and position data processed through a load model to determine load, thereby eliminating the need for complex direct force sensing hardware.
2Productivity
If load measurement is performed in harsh environments with dynamic machines, then productivity is improved, but measurement precision deteriorates due to external influences
Solution Approach 1:
The patent implements a feedback mechanism where the load model continuously receives updated machine parameters (acceleration, velocity, position) and adjusts load determination accordingly. This allows the system to compensate for changing environmental conditions and maintain measurement accuracy during dynamic operation.
Solution Approach 2:
The patent performs preliminary calibration to determine machine-specific parameters and characteristics before actual load measurement begins. This preliminary action creates a customized load model that accounts for the specific machine's behavior, making subsequent measurements more accurate even in harsh environments.
3Measurement precision
If rope length compensation and filtering algorithms are used to improve load measurement accuracy, then measurement precision is improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent changes the approach from correcting measured values through complex filtering to directly calculating load using fundamental physics parameters (acceleration, velocity, position) and a simplified load model, reducing computational complexity while maintaining or improving accuracy.
4Device complexity
If existing sensors are used for load determination, then device complexity is reduced, but measurement precision may deteriorate due to sensor limitations
Solution Approach 1:
The patent performs preliminary calibration using the existing sensors to determine machine-specific parameters and create a customized load model. This preliminary action compensates for sensor limitations and improves the accuracy of load estimation without requiring additional or more precise sensors.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for determining the load absorbed by a machine for dynamic material handling, in particular a crane, wherein the load is determined during operation by means of a load model, the load model being determined during a calibration phase of the machine as follows: - execution of a plurality of standard machine operations with a recorded reference load while available machine parameters are recorded, - performance of a significance analysis of the recorded parameters to identify statistically significant parameters that show a dependence on the load, - execution of a statistical learning and/or optimization algorithm to create a machine-specific load model that describes the relationship between the parameters identified as significant and the load.