Single Neuron PID Electro-hydraulic Proportional Position Control Based On RBF Identification

Proportional valve-controlled hydraulic cylinder system is widely used in various fields due to its simple structure,low price and high cost performance.However,the variable gain,dead zone and other nonlinear characteristics affect its control performance.Domestic and foreign scholars studied the proportional valve control cylinder system using fuzzy neural network,iterative learning,sliding mode variable structure and other intelligent control algorithms,and achieved certain research results.In this thesis,the single-neuron PID algorithm based on RBF network identification is studied on the basis of compensating the dead zone and friction force for a proportional valve-controlled common hydraulic cylinder system with large dead zone and complex friction.Firstly,the research background and significance of the research are outlined,the research status of electro-hydraulic proportional technology and neural network control technology at home and abroad,and the content of this research are also included.And the experimental system is built,taking the STM32F103ZET6 processor as the core,the signal conditioning circuit is designed.Network communication between processor and human-computer interaction interface is developed with Qt Creator.And the control effect is monitored.Secondly,the network communication between the processor and the human-computer interaction interface is realized.A system mathematical model is established based on the system’s composition and working principle.And based on the open-loop transfer function of the system,the dynamic characteristics of the system are analyzed through the Bode diagram.Then,the nonlinear characteristics of the system are analyzed.The self-learning dead zone compensation algorithm and model-based friction compensation algorithm are designed.Simultaneously,a control system simulation model is built in Matlab/Simulink,and a single neuron PID control algorithm based on RBF network identification is studied.The parameters in the single neuron PID are self-adjusted based on the Jacobian information of the RBF online identification and the simulation is performed.The role of each parameter in the algorithm is analyzed through simulation experiments.Finally,the single neuron PID control strategy based on RBF network identification is adopted in the actual system installation.Position control and tracking control are performed on the proportional valve controlled hydraulic cylinder to verify the effectiveness of the control algorithm.