Crane Activity Identification Using Hydraulic Pressure Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to reliably identify crane activities due to the complex and nonlinear interactions between kinematic patterns, load variations, and operational sequences, making it difficult to determine whether a load has been lifted or what activity was performed using a crane.
Innovation Solution
A computer-implemented method using a learned model, such as a neural network, integrates data from various sensors to recognize complex patterns and relationships between kinematic parameters, load interactions, and operational conditions, enabling accurate detection of crane activities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If simple kinematic pattern comparison is used to identify crane activities, then the method is simple and fast, but it cannot reliably determine whether a load was lifted due to diverse movement possibilities and varying loads
Solution Approach 1:
The patent transforms the identification approach from analyzing kinematic parameters (angles, lengths, speeds) to analyzing pressure parameters in hydraulic cylinders. This parameter change enables reliable load detection because pressure directly reflects load presence regardless of movement sequence or kinematic pattern variations.
Solution Approach 2:
The patent introduces pressure sensors in hydraulic cylinders as intermediary elements that indirectly detect load presence. Instead of directly measuring load or analyzing complex kinematic patterns, the system uses hydraulic pressure as a mediator that reliably indicates whether a load is attached, resolving the contradiction between simple detection and reliable identification.
2Reliability
If multiple sensors and complex analysis are used to reliably identify crane activities, then identification accuracy improves, but system complexity and computational requirements increase
Solution Approach 1:
The patent extracts the essential information needed for load detection from the complex system of multiple sensors and kinematic analyses. By focusing solely on pressure sensor data from hydraulic cylinders, it removes unnecessary sensors and computational complexity while retaining reliable load identification capability.
Solution Approach 2:
The patent replaces complex mechanical analysis systems (multiple position sensors, tilt sensors, force sensors, and kinematic pattern recognition algorithms) with a simpler pressure-based detection system. This substitution maintains reliability while significantly reducing device and computational complexity.
3Adaptability or versatility
If different movement sequences are allowed to achieve the same goal, then crane versatility is improved, but pattern recognition becomes unreliable due to varying kinematic parameters
Solution Approach 1:
The patent changes the measurement parameter from kinematic variables (angles, lengths, speeds) to hydraulic pressure. This parameter change makes measurement consistent and reliable regardless of movement sequence flexibility, as pressure directly indicates load presence independent of how the load was positioned.
Data Source
Figure 1~2
Figure 3
Figure 4a~4c
AI summary
A computer-implemented method (100) for identifying activities of a crane is proposed, comprising receiving a set of sensor readings (110) that represent a time course of operating parameters of the crane; and determining an activity performed by the crane (120) based on the set of sensor readings and a machine-learned model of the crane.