Impeller Clutch Predictive Control for Variable Material Loads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated systems for controlling the impeller clutch in earth-moving machinery are inefficient due to reliance on heuristic rule-based controllers and linear algorithms, which fail to adapt to varying material densities and conditions, leading to suboptimal torque and wheel speed control.
Innovation Solution
A system utilizing a time-step predictive analytical model, trained with a recurrent neural network, that receives fluidic pressure data to determine impeller clutch engagement values, allowing for adaptive and efficient control of the impeller clutch based on real-time working conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If heuristic rule-based controllers are used for impeller clutch control, then the system is simple to implement, but the control precision and adaptability to varying material densities are insufficient
Solution Approach 1:
The patent replaces heuristic rule-based controllers with a neural network-based automated control system. The neural network model processes sensor inputs (hydraulic pressure, engine speed, wheel speed) to dynamically determine optimal impeller clutch engagement, substituting simple rule-based logic with an adaptive intelligent system that achieves superior control precision while maintaining implementation feasibility through standardized computational approaches.
Solution Approach 2:
The system dynamically adjusts impeller clutch engagement parameters based on real-time sensor data including hydraulic pressure, engine speed, and wheel speed. The neural network processes these varying parameters to determine optimal clutch engagement values, enabling adaptive control that responds to changing material densities and operating conditions rather than relying on fixed heuristic rules.
2Device complexity
If linear algorithms are used for tractive effort control, then the control system is straightforward, but the system cannot adapt to varying work pile conditions and material densities
Solution Approach 1:
The patent implements dynamic control by replacing static linear algorithms with a neural network-based system that continuously adapts to changing conditions. The neural network processes real-time sensor inputs (hydraulic pressure, engine speed, wheel speed) to dynamically adjust impeller clutch engagement, enabling the system to respond to varying material densities and work pile conditions rather than following predetermined linear control paths.
Solution Approach 2:
The system incorporates feedback mechanisms by continuously monitoring sensor data (hydraulic pressure, engine speed, wheel speed) and using this information to adjust impeller clutch engagement. The neural network processes this feedback loop of real-time data to optimize control decisions, enabling adaptation to varying operating conditions while maintaining system simplicity through integrated sensor-controller-actuator architecture.
3Ease of operation
If manual clutch control is used, then the operator can respond to conditions, but human operators use the clutch inefficiently leading to suboptimal productivity
Solution Approach 1:
The patent implements self-service control by replacing manual operator control with an automated neural network-based system. The system autonomously processes sensor data and determines optimal impeller clutch engagement without human intervention, eliminating the inefficiencies of manual operation while maintaining the ability to respond to changing conditions through intelligent algorithms that continuously optimize productivity.
4Productivity
If automated control systems are implemented, then productivity can be improved, but the systems fail to account for un-modelable disturbances and varying material densities
Solution Approach 1:
The patent replaces traditional automated control systems with a neural network-based intelligent system that can handle un-modelable disturbances. The neural network's adaptive learning capability allows it to process complex, non-linear relationships between sensor inputs and optimal clutch engagement, reliably responding to varying material densities and unanticipated conditions that would challenge conventional control algorithms.
Data Source
AI summary
A system for controlling a machine impeller clutch includes a power source, a transmission unit, an impeller clutch operatively coupling the power source to the transmission unit, and an input device configured to generate a force data signal indicative of a working fluid pressure. An electronic control module in communication with the input device is configured to execute a time-step predictive analytical model for impeller clutch engagement. The electronic control module configured to receive the force data from the input device, determine an impeller clutch engagement value based at least in part on the force data signal and utilizing the time-step predictive analytical model, and cause engagement of the machine impeller clutch according to the impeller clutch engagement value.