Research On Real-time Detection System Of Subgrade Compaction Degree

The subgrade serves as the load-bearing layer of the entire road surface,and its quality directly determines the strength and stability of the entire road surface.In reality,the roadbed often has various quality problems such as uneven settlement,collapse,tumbling,and slope sliding.There are many reasons for these problems,but the lack of subgrade compaction is one of the main reasons.In addition,with the expansion of the roadbed and the wide application of various fillers,it is difficult for a single compaction equipment to meet the construction needs of the roadbed.Therefore,in order to improve the compaction efficiency of the subgrade,we can combine various compaction methods to improve work efficiency.The impact vibration composite compaction technology is one.It has the advantages of high production efficiency and wide adaptability.It combines the impact force of large energy with the exciting force to strengthen the compaction process.Domestic roadbed compaction technology has matured,but there is no real-time compaction detection system matching it.The detection of roadbed often uses hysteresis selection detection,which is poorly representative and cannot be timely and complete.Reflecting the compaction of the entire roadbed.Therefore,it is particularly important to develop a continuous detection system that can match the domestic compaction technology.This is also the basis of this paper.This paper theoretically analyzes the basic characteristics of the test soil,analyzes the compaction purpose,compaction method and compaction theory of the subgrade soil,and introduces the working principle,advantages and disadvantages of the impact vibration roller.Through the analysis of the kinematic differential equation of the "impact-vibration-static rolling dynamics model",the acceleration of the roller is closely related to the stiffness and damping of the material to be pressed,which lays a foundation for establishing a real-time detection system for compaction.Through the MATLAB/Simulink simulation of the impact and vibration dynamics model,the linear relationship between the roller acceleration and the compaction of the pressed material is obtained,and the feasibility of the system is also verified.The paper analyzes the basic principle and composition of the real-time detection system of compaction degree,and performs time-frequency domain analysis,noise source analysis and singular point analysis on the acquired acceleration signals,and uses wavelet analysis to denoise and denoise the acceleration signals.Determine the frequency resolution and the number of sampling points,and design the frequency domain filtering algorithm to filter the signal to obtain a real and reliable acceleration signal.Through the impact vibration compaction test,the data is analyzed and processed by the combination of amplitude method and linear neural network,and the acceleration index for evaluating the degree of compaction is obtained.The linear neural network is used to analyze and process the acceleration signal error,and the method is verified.Reliability.Finally,the similar theory is used to predict the maximum theoretical compaction that can be achieved with a 18-ton impact vibratory roller.