Fuzzy logic control for a hydraulic hybrid excavator based on torque prediction and genetic algorithm optimization

Abstract

Optimization of the control strategy, whose primary mission is to solve the problem associated with energy management, is an effective way to minimize the fuel consumption of the hydraulic hybrid excavator. As a widely used control strategy, fuzzy logic control can be adopted to realize suboptimal power split with robustness and adaptation, which is one of the most logical approaches for multidomain, nonlinear and time-varying plant. However, the membership functions are difficult to determine according to manual experiences; meanwhile, the optimization-based membership functions are difficult to utilize in real time control. This paper aims to improve the fuel consumption of a hydraulic hybrid excavator by proposing a fuzzy control strategy whose membership functions are optimized by the genetic algorithm, which considers predicted torque of the internal combustion engine (ICE) as a known quantity to realize real time control. The needed torque of the ICE is predicted by superposition of the previous torque. A fuzzy logic control strategy is then designed with membership functions optimized by the genetic algorithm according to the predicted needed torque to achieve better performance. Finally, a numerical experiment is carried out to verify the proposed control strategy.

Review on control systems and control strategies for excavators

“…Moreover, vehicle experiments were not performed to verify the effectiveness of the strategy. Zhou et al [23] designed a fuzzy controller based on membership function optimization and torque prediction to solve the problem mentioned. They used a genetic algorithm to optimize and adjust the membership function and BP neural network to realize the demand torque prediction of the internal combustion engine.…”

Novel control strategy for the energy recovery system of a hydraulic excavator

“…It is divided into deterministic rulebased control strategy and fuzzy rule-based control strategy [21] . The optimization-based control strategy is divided into global optimization control strategy, such as dynamic optimization [22] , particle swarm optimization [23] , genetic algorithm optimization [24] , and real-time optimization control strategy, such as equivalent consumption minimization strategy [25] , reinforcement learning control strategy [26] , model predictive control strategy [27] . Chen et al [28] proposed a rule-based control strategy based on the operating efficiency interval of the engine and the SOC of the ultra-capacitor.…”

Model predictive control based on time-varying efficiency for hydraulic hub-motor driving vehicle

“…11,12 A fuzzy logic control strategy based on torque prediction and genetic algorithm optimization has been proposed to reduce the energy and fuel consumption of hydraulic hybrid excavators. 13 The PI parameters were modified online by using fuzzy control rules to ensure that the vehicle would reach the target speed and self-balance performance. 14 The main powerful tool for establishing robust control theories consists of H ∞ optimization and μ synthesis.…”