Research And Design Of Operation Stage Perceptron And Matching Control Of Excavator

With the rapid development of industrial economy,the total quantity of excavators in our country has been increased continuously.Due to complex working conditions and sudden load mutation of hydraulic excavators,the engine speed fluctuates sharply,and the output power does not match the load demand.Due to low power utilization,the problem of higher fuel consumption cannot be ignored.In this thesis,by collecting the pilot pressure signal of the multiway valve and using hybrid neural network to classify and identify the excavator operation stage,determine the power required at the current stage and the corresponding optimal working point,and build a fuzzy single neuron PID controller to adjust the displacement of the hydraulic pump to stabilize the engine operating point.Through the strategy combined with segment perception and matching control,the engine operating points in different stages are stabilized,the speed fluctuations are reduced,and the energy saving is realized.The main content of this thesis is as follows:(1)The characteristics of the main components of the excavator(electric injection diesel engine,electric proportional variable pump)and the current control strategy and matching mechanism are analyzed.On this basis,the phased matching control strategy of the excavator is designed.(2)The perception model of excavator cycle operation stage is established.By analyzing the change of the pressure at the outlet of the front and rear pump and the pilot pressure of the actuator of the excavator in a working cycle,the working stages of the excavator are divided.The sliding time window was used to extract the pilot pressure of the multi-way valve to construct the feature vector.The convolutional neural network(CNN)was used to extract the feature and reduce the dimension of the pressure data in the time window.The integrated feature sequence was input into the long and short term memory neural network(LSTM)model to identify the stage of the excavator.The experimental results show that,compared with LSTM,the F1 scores of CNN-LSTM in each stage are improved.(3)Aiming at the problem of power matching between engine and hydraulic pump,an improved fuzzy single neuron PID speed controller is designed by combining fuzzy control and single neuron control.The simulation results show that the improved fuzzy single neuron PID controller based on the speed control can adjust the parameters in real time according to the load changes of the excavator,realize the rapid adjustment of the pump displacement to avoid the large fluctuation of the engine speed,so as to maintain the power matching and achieve the purpose of energy saving and consumption reduction.