| With the increase of terrorist attacks worldwide and the use of high-tech and criminal methods by criminals,it has brought great threats to the security of peoples around the world.Therefore,the detection of explosives has become a problem that countries attach great importance to,and the timely and effective detection of hidden explosives has become an urgent task for the international community.In recent years,fast neutron-based detection systems have been used to detect contraband in large containers and detect hidden explosives in small luggage.Among them,the tagged neutron method has been widely used at home and abroad as an effective detection method.By analyzing the characteristic ? rays generated by the interaction between 14 Me V tagged neutrons and the elements in the test object,it can effectively distinguish whether the test object is an explosive and can achieve the spatial positioning of suspicious objects.Compared with the traditional neutron detection method,the tagged neutron method reduces the measurement time,improves the detection accuracy,effectively reduces the background,and greatly improves the signal-to-noise ratio of the system.In this paper,experimental measurements and simulation calculations are carried out for the portable explosive package detection system based on the marker neutron method and the hidden explosive detection system in the wall,respectively.The MCNP5 program based on the Monte Carlo method is used to simulate different calculations.The ? spectrum of different explosives(TNT and Ammonium nitrate)is hidden behind the wall thickness(10 cm,15 cm and 20 cm),and the different explosives(TNT and Ammonium nitrate)are hidden in the luggage(Milk powder and Detergent)After the gamma spectrum.The experimental data are compared with the simulation results.For the hidden explosives detection system in the package,the main difference exists in the area below 3.50 Me V,partly because of the background caused by the surrounding environment,which is not calculated in the simulation.Another small reason is that the differences in actual experimental instruments(such as photomultiplier tubes)are not taken into account in the simulation model.For the explosives detection system in the wall,due to the complex structure of the detector,the emission peaks of carbon and oxygen in the simulated spectrum are slightly higher than the experimental spectrum.Because the exact hydrogen content in the wall cannot be determined,the 2.2 Me V hydrogen peak count in the simulated spectrum is higher than the experimental value.The C/O and C/N values of the experimental and simulated spectra were calculated,and their relative errors did not exceed ±20%.Individual differences between experimental results and simulated data may come from the uncertainty of the system,such as the accuracy of the nuclear data used in MCNP5.In addition,uncertainty in material composition during the modeling process and count losses in the data acquisition system may be another source of uncertainty.The differences between the simulations and experiments of the two systems are at an acceptable level,which verifies the feasibility of the model.The research results provide a reference for the establishment and optimization of the actual explosive detection device,and also for the data collection and analysis using the MCNP model.For different environments,qualitative and semi-quantitative analysis can be performed by modifying the model.In this paper,the principal component analysis(PCA),which is a commonly used data analysis method in statistics,is used to distinguish between explosives and non-explosive materials through principal component analysis. |
PDF Full Text Request