Intelligent energy adaptive control of loader shoveling system

Area of Science:

• * Mechanical Engineering
• * Artificial Intelligence
• * Robotics

Background:

• * Loader shoveling operations face challenges with varying working resistance and unpredictable time-varying factors, leading to high energy consumption.
• * The compacted layer significantly influences working resistance during the shoveling stage.
• * Existing control strategies lack adaptability to diverse working objects, hindering optimal energy utilization.

Purpose of the Study:

• * To analyze the influence of the compacted layer on loader working resistance.
• * To investigate the effect of boom lifting strategies on compacted layer destruction.
• * To develop an intelligent Energy Adaptive Control (EAC) strategy for optimizing loader shoveling efficiency.

Main Methods:

• * Analysis of the loader shoveling process to determine the impact of the compacted layer.
• * Discrete Element Method (DEM) simulations to model the destructive effect of timely boom lifting.
• * Development of a material recognition model using the Back Propagation (BP) neural network algorithm integrated into the EAC strategy.

Main Results:

• * Timely boom lifting demonstrates a destructive effect on the compacted layer.
• * The EAC strategy intelligently adjusts boom lifting range based on material type, reducing working resistance.
• * Significant reductions in peak engine power consumption were observed: 20.6% for sand, 19.1% for gravel, and 10.9% for boulders.

Conclusions:

• * The proposed EAC strategy effectively reduces energy consumption in loader shoveling systems.
• * Intelligent adjustment of boom lifting range based on material type is crucial for energy efficiency.
• * The integration of BP neural networks and DEM simulations provides a robust approach for optimizing loader operations.