Research On Adaptive Cutting Control Strategy Of Roadheader Based On Coal Rock Hardness

The cutting part is regarded as an important part of the roadheader.In the process of crushing coal and rock,it will generate huge energy consumption,accounting for about 75% of the power of the whole machine.At the same time,it is also subjected to huge loads and shock loads from coal and rock.Due to the complex and changeable physical properties of coal and rock,the cutting load is constantly changing,so that the cutting motor is often overloaded or underloaded,which seriously affects the cutting efficiency of the roadheader.Therefore,the cutting part should be able to dynamically adjust the cutting state according to the hardness of the coal and rock to ensure that the cutting part is always in an efficient cutting state.The control system of the roadheader still has problems such as inaccurate dynamic perception of coal and rock and low cutting efficiency.Aiming at the above problems,this paper proposes an adaptive cutting control strategy of roadheader based on coal rock hardness.The main work contents are as follows:Firstly,the dynamic analysis of the cutting part under horizontal,vertical and drilling conditions is carried out on the basis of understanding the structure composition and working principle of the cutting part of roadheader.The influence of different parameters on cutting head load was studied.On the basis of the appeal research,the cutting performance under the swing condition is optimized to further improve the cutting efficiency of the roadheader.The swing speed and rotational speed of the cutting head are the optimization variables,and the cutting specific energy consumption and load fluctuation are the optimization objectives.The optimal rotational speed and swing speed under different coal rock hardness are obtained by optimization,which provides the basis for efficient cutting of roadheader.Secondly,Secondly,theoretical analysis shows that load torque fluctuation will produce amplitude modulation and frequency modulation on motor current signal,and simulation verifies the correctness of this conclusion.A signal processing method of complete ensemble empirical mode decomposition and wavelet threshold denoising is proposed to solve the problem that the motor current signal could not reflect the change of load torque due to the large disturbance of downhole environment.The CEEMDAN method is used to decompose the current signal to obtain a series of intrinsic mode functions,and then the noise-containing intrinsic mode components are filtered through the correlation coefficient.Then the wavelet threshold algorithm is used to denoise the components with more noise,and the denoised IMF component and the undenoised IMF component are reconstructed to remove noise and retain useful features in the signal.Then,in order to solve the problem of inaccurate identification of coal and rock,a rock hardness estimation algorithm based on Adaboost improved BP neural network is designed with BP neural network as a weak learner,multi-scale permutation entropy of stator current signal and coal and rock hardness as training samples.The number of hidden layer nodes,the learning rate in BP neural network and the number of BP weak learners in Adaboost strong learner model are determined by trial and error method.The relative root mean square error evaluation model is used to quantitatively analyze the estimation results of coal rock hardness of BP model and Adaboost-BP model.Finally,the cutting method based on coal and rock hardness is applied to the cutting control of roadheader,and the adaptive cutting control strategy of roadheader based on coal and rock hardness is designed.The adaptive cutting of coal and rock hardness change was simulated on MATLAB/Simulink platform,and the feasibility of the control strategy was verified.