Method Of Measuring Source Activity For Alpha Pollution Inside Pipes Using Long Range Alpha Detector

With the rapid development of nuclear science, nuclear waste disposal has attracted growing attention. For the decommissioning of nuclear facilities, most of the radioactive sources are alpha radiation body. While the amount of alpha sources mainly measured via measuring activity of alpha, the focus of decommissioning of nuclear facilities has shifted to alpha measurement research. For alpha measurement, there are many detectors or methods. But traditional ones are restricted by the short range of the alpha particle, so it is difficult to overcome the geometric constraints of the nuclear facility such as irregular surface and cavity. So far, there’s no known reliable method to determine the radioactive intensity of alpha source and its location interior pipes, making it difficult to effectively deal with pollution facilities. To meet the demands of the above, long-range alpha detection (LRAD) technology came next.Early in the beginning of the1920s, the United States Department of Energy’s Los Alamos National Laboratory (LANL) first put forward the LRAD technology, and developed a prototype. Later on, there were a large number of relevant researches on equipment performance optimizing and application improvement. Thus, the LRAD technology has enriched and improved a lot, and the domain of LRAD research expanded. At the end of the20century, our country joined the ranks of LRAD research. So far, we’ve achieved remarkable improvement on equipment performance optimizing, while there are still much work to do. The source activity of the inner surface of pipe could not accurately be measured so far. For the further realization of accurate measurement of alpha source inside pipes, our research group has independently developed the detection system based on LRAD technology. Also, research on key parameters that may cause non-linear errors on the readout has been carried out. According to the result of experiment, influence character of each key parameter was preliminary obtained. And for the purpose of non-linear correction to the key parameters, many mathematical algorithms have been successfully adopted. However, all of the correction results are not source activity which we want to find out, but readout that cannot reflect radioactive intensity.In addition, there’s another point that is as importance as source activity detection. For nuclear waste disposal, the source localization plays an important role. The realization of source localization could find the contamination source destructively. Finding the source localization, on the one hand, can avoid wasting material that source content is low enough. On the other hand, knowing the source position can help to dispose section of enough source content targetedly. It can not only avoid source detection missing, but also can avoid unnecessary radiation contamination. It obviously that choosing the source localization method as a research subject is of significance meaning to decommissioning of nuclear facilities and environment protection.Aiming at measuring activity of radioactivity source, this paper carried out "LRAD applied to alpha activity detection inside pipes" based on research status both at home and abroad, Supported by National Natural Science Foundation of China under Grant No.41025015.The major research contents and results are as follows:1) Research on LRAD applied to alpha activity detection inside pipes.In order to find out system response features of key parameters such as source-to-detector distance, air speed, air flow, pipe length and test pipe diameter, the variable-controlling approach was applied to this system. The result presents a complicated non-linear relation between readout and key parameters. It’s hard to determine radioactive intensity or source activity by the readout. Therefore, for the purpose of finding relationship of readout and source activity, the BP neural network with input of key parameters and readout, and output of source activity was built. The training results indicate that the network prediction has a high degree of accuracy, indicating that this method is reliable enough for site source measure.2) Radioactive source localization inside pipes using LRADWithout periodic characteristics, alpha source position inside pipe cannot be determined easily. Method in this paper used ion drift distance to establish the distance from source to detector. Based on relationship between speed, time, and drift distance, it is important to obtain airspeed distribution in pipe and ion travel time. The ion travel time is established with CAD software with experiment dada, while the airspeed distribution is established by fluent software. Errors between calculation and source-to-detector distance were small enough. That is to say, this method could be used in site distance measure.3) BP neural network applied in site measure of pipe using LRADBased on BP neural network that built on experiment, the source activity inside pipe can be calculated by the network if the key parameters and readout were obtained. Therefore, this paper combined source localization method which was used to calculate distance with BP neural network to calculate source activity in site source measure. Proved through practice, the calculation in this paper is appropriate and effective, despite its slightly larger error over training result.Generally, the innovations of this paper are the following three points:1) Firstly conduct source localization experiment and method, using fluent software to obtain airspeed distribution and CAD software to calculate ion travel time. The localization of alpha source helped to offer proper treatment to different parts of polluted pipe, avoiding unwanted material wasting or environment pollution. In addition, the nondestructive performance of LRAD is improved. It also solved the problem of obtaining key parameters in BP neural network.2) On the basis of key parameter effect analyzing, this article used BP neural network to build relationship between readout and source activity under affecting of key parameters. Combining source localization, all of the input vectors of the BP neural network can be obtained, helping to realize the purpose of obtaining source activity. This method has not only overcome non-linear effect of key parameters, but also realized LRAD’s site measure value.In summary, this paper aims at obtaining alpha source activity. This method applied BP neural network to build relationship between readout and source activity under affecting of key parameters. Based on this, using fluent software and other methods to measure source-to-detector distance and other input vectors of the BP neural network, the site pipe’s source activity can be calculated by the BP neural network. It proves that this method gives a reference to source detection for pipes using LRAD.