Đang tải... (xem toàn văn)
Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.Nghiên cứu ứng dụng điều khiển hiện đại trong nổ mìn tại các mỏ lộ thiên ở Việt Nam.
MINISTRY OF EDUCATION AND TRAINING HA NOI UNIVERSITY OR MINING AND GEOLOGY DAO HIEU RESEARCH TO APPLY MODERN CONTROL IN BLASTING AT OPEN-PIT MINES IN VIETNAM Major: Control engineering and automation Major code: 9520216 THESIS SUMMARY Ha noi – 2022 The thesis was completed at: Automation Department, Faculty of Electro-Mechanics, University of Mining and Geology Supervisors: Dr Dang Van Chi Associate Professor Dr Pham Van Hoa Reviewer 1: Associate Professor Dr Hoang Sy Hong Hanoi University of Science and Technology Reviewer 2: Associate Professor Dr Nguyen Van Tiem University of Transport and Communications Reviewer 3: Associate Professor Dr Dam Trong Thang Military Technology Academy The thesis will be defended at Ha Noi University of Mining and Geology in day month year 2022 The thesis can be found in the Ha Noi National Library or the Library of Ha Noi University of Mining and Geology INTRODUCTION Urgency of the topic In mining blasts in Vietnam, there are many factors that are limiting the effectiveness of solutions applied to control explosive energy In which, the ground coefficient of rock and the error and the fixation of the delay detonator are the two main reasons Science and technology with modern control methods are bringing great achievements, forming a common development trend in most fields From that, blasting in the world has been very developed In that, the requirement to research and apply technology for mine blasting in Vietnam becomes extremely urgent Therefore, the thesis topic “RESEARCH TO APPLY MODERN CONTROL IN BLASTING AT OPEN-PIT MINES IN VIETNAM” is topical, scientifically significant, and meets the requirements of practice in the mining industry Research objectives of the thesis Application of modern control techniques to determine reasonable delay time value and propose solutions to build automatic delay time correction system for blasting at open-pit mines in Vietnam Object, research scope of the thesis Research on modern methods to determine and control delay time for blasting at open pit mines in Vietnam The scope of the study is blasting on open-pit mines in Vietnam Research Methodology The research methods used in the thesis are: theoretical research, simulation research, synthesis of practical data, experiment Scientific and practical significance of the thesis Scientific: Using statistical methods, Kalman filtering techniques and expectation maximization (EM) to analyze blasting shock wave data to determine wave propagation velocity Application of artificial neural network (ANN) to build a model to recognize the relationship between differential time and shock wave propagation velocity Practical: Combine recognition model and algorithm to propose suitable delay time and predict tremor level for the study area The research results are the initial step to improve the technology for blasting in open-pit mines in Vietnam The proposed system solution is suitable for domestic engineering and technology, with high practical applicability Defensive arguments The post-explosion shock wave propagation velocity indirectly describes the state of rock in the explosion area The propagation velocity of the shock wave can be determined by analyzing the explosion diagram in conjunction with the shock waveform obtained at the measurement point The algorithm uses an artificial neural network that can determine a reasonable delay time and predict the magnitude of the tremor New point of the thesis Algorithm for determining shock wave propagation velocity and differential time correction based on neural network application Dissertation layout In addition to the introduction and conclusion, the content of the thesis includes 129 typed pages excluding appendices, chapters, 12 tables, and 67 illustrations and 100 domestic and foreign references Chapter OVERVIEW OF EXPLOSION AND SHOCK WAVES 1.1 Mine blasting overview 1.1.1 Introduce Blasting in mining is done by drilling rows of drill holes into the rock The amounts of explosives in the detonated boreholes will disrupt rock structures 1.1.2 The effects of delay blasting Controlling the delay time so that the interference of waves between explosions occurs will increase the total stress acting on the rock mass, the breaking strength is increased The more free surfaces around the explosion point, the higher the breaking rate The broken rocks have different speeds and directions of movement When they collide, a secondary break occurs 1.1.3 Some basic types of delay time diagrams Delay time diagram depicting the order of control of explosive quantities over time Common are the through-row delay time and through-hole delay time diagrams 1.2 Shock waves 1.2.1 Basics of shock waves Shock waves are vibrational waves that propagate across the earth's surface Shock waves are divided into three components: Compressional waves - abbreviated as P), shear waves abbreviated as S and Rayleigh waves - abbreviated as R V Blast point L T Figure 1.4 Coordinate axes used to describe blasting shock waves 1.2.2 Blasting shock waves When the stress wave caused by the explosion propagates over a certain distance, its energy is no longer capable of breaking, but can only create rock vibrations, at this time it becomes a shock wave The longer the distance and the smaller the amount of potion per blast, the less concussion will be The propagation velocity represents the speed at which the shock wave travels from the point of explosion to the measured point on the earth's crust In a certain area, this value is almost constant The propagation velocity of the shock wave is inversely proportional to the moisture of the rock; is directly proportional to the density of rock and the compressive resistance (compression resistance), inversely proportional to the porosity of the rock, and many other relationships Because the ground is a heterogeneous medium, different directions will have different levels and frequencies of vibrations Wavelength and frequency are parameters that determine the ability to create interference effects between waves 1.2.3 Control shock waves level Using delay time reduces the explosive charge instantaneously at a time Or adjust the distance to the explosion point 1.3 Some experimental studies on the relationship between delay time and shock waves and smashing efficiency Experimental studies on various structures in the laboratory show a close relationship between the stress wave, the delay time interval, and the smashing efficiency 1.4 Research situation, application of modern control techniques for blasting in the country and in the world 1.4.1 Research situation and applications in the world The research in the world is comprehensive, updating modern techniques, latest technology The research results have been thoroughly applied to become very modern equipment systems that very well serve the requirements when performing explosions 1.4.2 Research and application situation in the country The studies are local, so the results are limited and the applicability is low 1.4.3 Comment Due to the specificity of the region and mining methods and technology, it is not possible to apply the world's development achievements to mine blasting in Vietnam 1.5 Overall conclusion The experience of the world shows that the research and application of modern technologies and techniques is a development trend Therefore, "Research and application of modern control techniques in differential blasting" is an urgent need for blasting in Vietnam Two of the important factors affecting the quality of blasting are the physical and mechanical condition of the rock and the capacity and quality of the delay time being used Focusing on research and limiting the disadvantages of these factors for blasting in Vietnam is the goal The shock wave propagates through the rock Therefore, the shock wave propagation velocity indirectly describes the state of rock in the explosion area The level of technological development in the country completely allows to improve the accuracy and flexibility of delay time in blasting Chapter RESEARCH FOR DETERMINATION OF THE PROPAGATION VELOCITY OF SHOCK WAVES 2.1 Introduce The shock wave is closely related to the stress wave, and depends on the rock in the explosion area Therefore, the propagation velocity of the shock wave is the basis for indirect identification of the rock condition - an important factor affecting the quality of the explosion 2.2 Explosion shock wave data collection 2.2.1 Principles of data collection Complete, continuous data collection over time The measuring device is the standard one The measurement process is done by an expertise, and experience person 2.2.2 Research area and measuring - recording data solutions Data logging option with two opencast mines with different rock structures: coal mines and limestone mines 2.2.3 Record data at Nui Beo coal mine (a) (b) Figure 2.2 Some results of recording seismic data at Nui Beo mine Table 2-3 Some parameters of mine blasting data are recorded at Nui Beo coal mine No Parameter (unit) Total amount of explosives (Kg) Amount of explosives in a borehole (Kg/lỗ) Total number of holes drilled Row spacing – holes (m) Delay times (mili second) Measuring distance (m) Data recording time (giây) (a) 784 56 (b) 972 54 14 5-6 42, 100 396 18 5-6 42, 100 335 2.2.4 Measure and record data at Hong Son limestone quarry (a) (b) Figure 2.5 Some measurement results recorded at Hong Son limestone quarry Table 2-5 Some parameters of mine blasts have data recorded at Hong Son limestone quarry No Parameter (unit) Total amount of explosives (Kg) Amount of explosives in a borehole (Kg/lỗ) (a) 1136 47,3 (b) 2040 51 2.4.2.2 Analysis of the second explosion Figure 2.19 Schematic diagram of the differential timing of the analyzed HS2 explosion X T: 0.06055 T: 0.05566 A: 0.339 A: 0.323 T: 0.1064 A: 0.236 T: 0.00488 A: 0.268 T: 0.1094 A: 0.252 T: 0.1182 A: 0.221 T: 0.1465 A: 0.205 T: 0.1275 A: 0.102 T: 0.223 A: 0.221 T: 0.1605 A: 0.15 T: 0.2305 A: 0.158 Figure 2.21 The results of determining the group of wave peaks T: 0.1367 A: 0.0867 T: 0.2412 A: 0.11 A: 0.0236 along the L axis and synthesizing the PPV of the HS2 explosion Table 2-9 Calculation results of the propagation velocity of the HS2 explosion (m/s) Results calculated according to the data of the L axis Calculation value range 590 - 3200 Average velocity in group analysis 1511 Average velocity in group analysis 2680 Overall average speed 2096 Results calculated according to the data of the PPV Calculation value range 540 - 5050 Average velocity in group analysis 1435 Average velocity in group analysis 2637 Overall average speed 2036 2.5 Conclusion of chapter The relative velocity of the shock wave propagation can be determined by analyzing the wave characteristics obtained after each explosion The relationship between the propagation velocity of the shock wave and the differential time indirectly describes the relationship between the current state and the physical and mechanical properties of the rock and the differential time value Chapter BUILDING A MODEL TO IDENTIFY THE RELATIONSHIP BETWEEN DELAY TIME AND PROPAGATION VELOCITY OF SHOCK WAVE 3.1 Introduce Execution Unit Borehole network Control Objects (Earth rock explosion area) Amount of potion time explosion Explosive Differential Diagramresults Request CPU Measure Figure 3.1 Equivalence diagram depicting differential blasting according to the principle of controlled system Reference PPV Propagation velocity Explosion point - measuring point distance Amount of explosion one time Previous Propagation velociy Tvi sai CPU SYSTEM MODEL Propagation velocity and PPV Data processing (EKF,EM) M U X PPV Previous Delay time Figure 3.3 Principle of delay time correction and shock level prediction for delay blasting on open pit mines in Vietnam 3.2 Research and develop data processing methods 3.2.1 Analyze and identify sources of interference The sources of noises: rock structure, topography; due to the environment; due to the sensor device; due to other factors 3.2.2 Data processing solutions The noise processing methods are: Kalman Filter _ KF; Extent Kalman Filter _ EKF; Expectation Maximization _ EM 3.2.3 Building data processing solutions The noise processing solution chosen is a combination of EKF and EM 3.3 Build an identity model Figure 3.12 The process and results of training ANN artificial neural network with a hidden layer Figure 3.13 The process and results of training artificial neural network ANN with two hidden layers in case Figure 3.14 The process and results of training artificial neural network ANN with two hidden layers in case Figure 3.15 The process and results of training artificial neural network ANN with two hidden layers in case Figure 3.16 The process and results of training artificial neural network ANN with three hidden layers 3.4 Selecting and verifying the recognition model Table 3-1 Comparison table of test results of ANN network model structures Network structure ANN Best Training Performance ANN hidden layer 0.051099 ANN hidden layers (case 1) 0.025369 ANN hidden layers (case 2) 0.074087 ANN hidden layers (case 3) 0.0037392 ANN hidden layers 0.091401 Table 3-1 combined with the comparison of the Training graph, the results of selecting the hidden layer network structure according to Figure 3.15 To check the quality of the recognition model, 30% of the previously left data corresponding to 100 datasets were used These data are fed to the Input Data input for the Custom Neural Network model Figure 3.17 Hình 3.17 Kết kiểm tra mơ hình 3.5 Model testing (a) (b) Figure 3.18 Structural diagram depicting test cases The best delay time value is The best pair of delay time values 15.8ms, the predicted shock is 15ms and 9.65ms, the predicted level is 1,439mm/s vibration level is 1.6mm/s The best delay time value is 11.9ms, the predicted shock The best pair of delay time values is 15ms and 10.7ms, the predicted level is 1.418mm/s vibration level is 1.366mm/s The best delay time value is The best pair of delay time values 12.2ms, the predicted shock is 25ms and 18.4ms, the predicted level is 1.423mm/s shock level is 0.482mm/s The best pair of delay time values is 25ms and 17.8ms, the predicted shock is 1,834mm/s Figure 3.19 Test results with 1st case: The two delay time Figure 3.20 Test results with the values are the same and vary 2nd case between 8-22 ms 3.6 Comments and conclusions chapter Neural network application successfully trained the model to recognize the relationship between the shock wave propagation velocity and the delay time value The identification model is the basis for correcting the delay time with the study area and predicting the magnitude of the tremor To further develop this solution, it is necessary to build an automatic, continuous data collection and storage system over time Increase the accuracy of wave propagation velocity data Build automated analysis software on a database that is regularly updated Chapter RESEARCH SOLUTIONS TO CONSTRUCTION SYSTEM AUTOMATICALLY ADJUSTING DIFFERENT TIME AND FORECASTING SHOCKET LEVELS FOR EXPLOSIONS ON THE LO THIEN GENERAL IN VIETNAM 4.1 Introduce and conditions for applying the research system From the chapter conclusion, the analysis software algorithm is described: Begin Tvs0=Tvs1, Vxt, Txt Determine the trend of Vrd change of the last 10 explosions Y N Up Vxt=1 Y N Down Vxt=-1 Vxt=0 Determine the trend of Tvs change of the last 10 explosions Y N Up Txt=-1 Txt=1 N Vxt=1&Txt=-1 or Vxt=-1&Txt=1 or Vxt=0&Txt=-1 Vxt=1&Txt=1 or Vxt=-1&Txt=-1 or Vxt=0&Txt=1 Y down Tvs0 Y Up Tvs0 N Run ANN model Y Vrd0 -> And Tvs1-10