Research And Application On Nonlinear Prediction Methods Of Bench Blasting Vibration Effect

Bench blast is most widely used in blasting engineering such as in opencast mining, road engineering, railway engineering, hydraulic and hydro-power engineering, subway construction, urban construction, national defense construction. In China, earthwork of blasting excavation is nearly 8 billion every year, in which 70 percent is excavated by bench blast, and the harmful effects of blasting continue to affect the normal life of the surrounding residents. With the increasing awareness of environmental protection and the request to improve the living environment, controlling harmful effects of blasting and protecting the surrounding environment is an important aspect of blasting operations. So prevention and control of blast induced harmful effects have become a hot spot in blasting engineering. The harmful effects induced by blast include flying rock, harmful gas, blasting dust, blasting shock wave, blasting vibrations and noises etc. The harmful effects except blasting vibration can be well controlled by optimizing explosive mixture ratio, water mist dust removal, safety protection, and controlling charge weight at simultaneous initiation etc. However, it is difficult to control the blasting vibration only by optimizing blasting design, and the accidents induced by blasting frequently occurred in blasting engineering. Therefore, it is urgent requirement to develop new prediction method of blasting vibration in order to prevent and control blasting vibration in blasting engineering. This work is systematically based on the domestic and international research, combine with the combination of “the technology research of bench blasting in fractured rock(science and technology comprehensive talent team of GuiZhou [2014] No. 4003)”,“The technology research of reducing blasting vibration in moderate long hole bench blasting based on energy principle(science and technology comprehensive talent team of GuiZhou [2013] No. 30)”,“The research and application of deep hole blasting numerical simulation in karst areas(science and technology comprehensive talent team of GuiZhou GY[2013] No. 3054)”and other research projects, targeting bench blasting vibration characteristics of the nonlinear prediction. A research route, which combines theoretical model construction, code development, numerical simulation and field investigation, has been employed in this work. A systemic research has been proceeded to obtain a new method of blasting vibration prediction. The main content includes: 1. Field investigation of blasting vibration induced by bench blastThe blasting vibrations are measured at multi measured points for different hole depth, charge weight, charge structure and initiation time, analyzing the blasting vibration attenuation and dissipation law through the vibration waveform in different situations. Conclusions:The maximum velocity and dominant frequency amplitude decreases as the charge decreases, radial and vertical maximum velocities are larger than the tangential, these indicate that the blasting stress wave at radial and vertical are dominant; the maximum velocity and dominant frequency amplitude by continuous charge are larger than interval charge, while the tangential amplitude is largest, then the radial, and the vertical is smallest; The main vibration frequency by the continuous charge is smaller than the interval charge, and the vibration velocity amplitude by continuous charge is larger than the interval charge; the peak velocity of particle vibration spectrum by the interval charge are larger, and the amplitude is smaller than the continuous charge. obtain the value of K?a by regression analysis based on Sodev's vibration prediction empirical formula as the objective function, as the value of K?a by the interval charge are much larger than the continuous charge, these indicate that the interval charge for reducing the blasting vibration is obviously useful. the shorter the delay time, the larger the peak velocity; blasting vibration frequency are mainly in 10 Hz to 200 Hz, the dominant frequency are in 20 Hz to 60 Hz. The low frequency occupies more than high frequency in continuous charge blasting, and the amplitude are larger; but in the interval charge explosion, the high frequency occupies more and the amplitude are smaller; Attenuation speed of the peak velocity of particle vibration is faster in close distance from blasting source than far away distance from blasting source. Attenuation speed of the peak velocity of particle vibration is A typical nonlinear relationship. The peak velocity of particle vibration is almost at the last arrived wave crest or trough. The results show that the blasting vibration velocity and frequency are affected by explosive load, charge structure, the distance from blasting source, hole depth, delay time and so on various factors. The peak particle vibration velocity attenuation decreases with the increase of distance, which is a typical nonlinear characteristics. 2. Modeling Research on nonlinear system of forecasting model.1) A nonlinear prediction method based on Volterra Functional Series of bench blasting vibration effect is proposed.Research of solving nonlinear problem is based on estabilishing a mathematical model to describe the nonlinear system. Therefore, this study is based on significant nonlinear characteristics of rock and soil media between explosion source and measuring points. Introduce Volterra Series method into prediction of bench blasting vibration effect. Volterra functional series is used to describe nonlinear characteristics of geotechnical medium between explosion source and measuring points. Identifying kernel function of Volterra functional series through single hole blasting vibration experiment, According to blasting parameters for the design, constructing pulse sequence function and forecasting blasting vibration effect of the given program is completed.2) Anderson forecasting model based on time-domain waveform superposition is improved.In order to overcome the lack of Anderson model this is unabled to reflect nonlinear characteristics of the geotechnical medium and waveform linear superposition. In order to fully reflect the geotechnical medium of discontinuity, inhomogeneity and anisotropy, identify seismic proportional coefficient of Anderson model. According to the characteristics of bench blasting, the improved Anderson model based on the seismic proportional coefficient of bench blasting is proposed.3) Volterra nonlinear system model, the improved Anderson model and the calculation program of structural response prediction are developed respectively.Efficient combined of STL with OpenGL database resources based on C++ platform, Volterra nonlinear model and the calculation program of improved Anderson model are compiled. According to the improved Anderson model and forecasting vibration waveform of nonlinear Volterra Functional Series, use C++ language to develop calculation program of mode-superposition response spectrum method, time history analysis method and transfer function method using in predicting law of vibration effect with time change of internal force, shear stress and displacement of building structure, etc.4) Using three methods to analyze the dynamic response of the structure of ground vibrationUsing three methods(the mode decomposition response spectrum method, the transfer function method and the time history analysis method) to analyze the dynamic response of the structure of ground vibration, and establish a comprehensive prediction model which combines the field vibration test and the theoretical model. Firstly, the vibration waveform of ground is predicted, thenthe vibration response of the structure is predicted by the mode decomposition response spectrum method, the transfer function method and the time history analysis method. According to this, it can be judged that whether the blasting vibration is harmful to the building in a given case of blasting scheme. If any hidden dangers, modify the blasting design according to the predicting results, reducing the damage on structure caused by blasting vibration effect to a minimum. 3. Numerical simulation about bench blasting vibration actionIn order to verify the reliability and validity of Anderson model and Volterra nonlinear prediction model, by forecasting results of the two models compared with the experimental results and using numerical simulation analysis. Application of ANSYS / LS-DYNA respectively holes, three holes and four holes at different charge structures and conditions for extension of time to simulate synchronization, simulation data and measured data at the same measuring point error can be controlled within 60% most simulation results close to the measured values, the data closer to the actual value of the radial than vertical data, vertical data deviate slightly; from the waveform chart, analog waveforms close to the measured waveform. But the whole rock as homogeneous continuous media, and live rock media is defective joints crack sand other geological body, so the simulation data and real data there is always a certain bias. Meanwhile, the numerical simulation of the process without considering geotechnical characteristics of the propagation medium has a high-frequency filter, so the numerical simulation of the process of showing a high frequency vibration wave. Anyway numerical simulation is used as a theoretical reference blasting design has important practical significance. It's main drawback is difficult to determine the numerical simulation parameters, long simulation time, and requires killed technical personnel must have experience in numerical simulation, the environment is more complex and difficult to operate, it allows numerical simulation methods are not easy to spot applications. It confirmed that the spread of the explosive stress wave in rock mass is nonlinear characteristics by researching on the numerical simulation of rock blasting under different parameter condition, and can provide a sufficient theoretical basis for nonlinear prediction 4. Verification and applications of the prediction methods in bench blastsTo verify the validities of the prediction methods, the nonlinear model proposed in this work was verified by tests and used in bench blasts of a open pit limestone mine. The validity of the nonlinear model in practical engineering was verified by comparison of the prediction results and the measured results. The results of comparison show that the prediction results of the nonlinear prediction method which is based on the Volterra functional series have high accuracy. The inaccuracy of the PPV predicted by the nonlinear prediction method is less than 16%; The ZC frequency deviation is less than 31%; The inaccuracy of spectrum value corresponding with ZC frequency is less than 10%.The inaccuracy of the PPV predicted by improved Anderson model method is less than 24%; The ZC frequency deviation is less than 50%; The inaccuracy of spectrum value corresponding with ZC frequency is less than 17%. At the same time, The predicted results of forecasting model are closer to the actual vibration test data than the simulation results through the comparison between results of the numerical simulation results and the nonlinear model predictive, the predict the results of nonlinear prediction model fully confirm to reflect geotechnical medium nonlinear characteristics, the predicted results can provide theoretical basis for the evaluation of rock and soil blasting vibration effect.In the experimental verification of building response subjected to the blasting vibration, firstly, we apply the improved Anderson model to predict the vibration waveform in the foundation of building and then take the vibration waveform in the foundation of building as importation to predict the response of building subjected to vibration in the foundation. We have applied modal decomposition response spectrum method to predict the internal forces and moments of the building subjected to impacting vibration of demolition block. Using time history analysis method to predict the response of structure subjected to blasting vibration is truly reflecting the effects of the duration of vibration on the response of the building subjected to vibration. Using the transfer function method to predict the fulltime history of vibration, the prediction results approximately approach to the measured waveform and the spectrum are basically the same. It indicates that the improved Anderson model prediction method is proper and can be used to predict the response of building subjected the blasting vibration.