An investigation into accidents in laboratories in universities in China caused by human error: A study based on improved CREAM and SPAR-H.

Although considerable research has been devoted to improving safety in university laboratories, accidents, in that environment, have still occurred frequently at the cost of serious injury or even death of laboratory personnel. Currently, few Human Reliability Analyses (HRA) have been conducted with respect to a university laboratory. The aim of the research was to conduct a reliability study relating to human behaviour in a university laboratory to explore quantitatively the causes and influencing factors relating to the frequency of laboratory accidents. Improved Cognitive Reliability and Error Analysis Method (CREAM) and improved Standardized Plant Analysis Risk HRA (SPAR-H) were employed to assess Human Error Probability (HEP) of 23 subjects. The HEP calculated through improved CREAM proved more accurate than results obtained through improved SPAR-H. Unexpectedly, the results demonstrated that under similar environmental conditions, the HEP of subjects did not decrease with an increase in educational background, including additional experimental time and experience. Moreover, environmental conditions exerted greater impact on personnel reliability than Human Inherent Factors (HIFs) in laboratories. It is anticipated that the study would provide valuable insights, in respect of research methods, and to serve as a practical basis for lowering the accident rate in university laboratories.

Remaining useful life prediction techniques for electric valves based on convolution auto encoder and long short term memory.

To optimize the operation and maintenance of nuclear power systems, this study presents a remaining useful life (RUL) prediction method for electric valves by combining convolutional auto-encoder (CAE) and long short term memory (LSTM). CAE can extract deeper features and LSTM is efficient in dealing with time-series data. Moreover, by designing a parallel structure between the outputs of CAE and the original data, features fed into the LSTM are enriched. Also, network structure and corresponding hyper-parameters are compared to obtain a more suitable model. Moreover, the accuracy of the proposed method is tested and compared with other machine learning algorithms. This work also serves as a critical innovation to enhance the safety and economic operation of nuclear plants and other complex systems.

A hybrid fault diagnosis methodology with support vector machine and improved particle swarm optimization for nuclear power plants.

The safety and public health during nuclear power plant operation can be enhanced by accurately recognizing and diagnosing potential problems when a malfunction occurs. However, there are still obvious technological gaps in fault diagnosis applications, mainly because adopting a single fault diagnosis method may reduce fault diagnosis accuracy. In addition, some of the proposed solutions rely heavily on fault examples, which cannot fully cover future possible fault modes in nuclear plant operation. This paper presents the results of a research in hybrid fault diagnosis techniques that utilizes support vector machine (SVM) and improved particle swarm optimization (PSO) to perform further diagnosis on the basis of qualitative reasoning by knowledge-based preliminary diagnosis and sample data provided by an on-line simulation model. Further, SVM has relatively good classification ability with small samples compared to other machine learning methodologies. However, there are some challenges in the selection of hyper-parameters in SVM that warrants the adoption of intelligent optimization algorithms. Hence, the major contribution of this paper is to propose a hybrid fault diagnosis method with a comprehensive and reasonable design. Also, improved PSO combined with a variety of search strategies are achieved and compared with other current optimization algorithms. Simulation tests are used to verify the accuracy and interpretability of research findings presented in this paper, which would be beneficial for intelligent execution of nuclear power plant operation.

Virtual reality-based adaptive dose assessment method for arbitrary geometries in nuclear facility decommissioning.

This paper presents an improved and efficient virtual reality-based adaptive dose assessment method (VRBAM) applicable to the cutting and dismantling tasks in nuclear facility decommissioning. The method combines the modeling strength of virtual reality with the flexibility of adaptive technology. The initial geometry is designed using three-dimensional computer-aided design tools, and a hybrid model composed of cuboids and a point-cloud is generated automatically according to the virtual model of the object. In order to improve the efficiency of dose calculation while retaining accuracy, the hybrid model is converted to a weighted point-cloud model, and the point kernels are generated by adaptively simplifying the weighted point-cloud model according to the detector position, an approach that is suitable for arbitrary geometries. The dose rates are calculated using the point kernel method. To account for radiation scattering effects, buildup factors are calculated using the geometric progression formula in the fitting function. The geometric modeling capability of VRBAM was verified by simulating basic geometries, which included a convex surface, a concave surface, a flat surface and their combination. The simulation results show that the VRBAM is more flexible and superior to other approaches in modeling complex geometries. In this paper, the computation time and dose rate results obtained from the proposed method were also compared with those obtained using the MCNP code and an earlier virtual reality-based method developed by the same authors.

A fast simulation method for radiation maps using interpolation in a virtual environment.

In nuclear decommissioning, virtual simulation technology is a useful tool to achieve an effective work process by using virtual environments to represent the physical and logical scheme of a real decommissioning project. This technology is cost-saving and time-saving, with the capacity to develop various decommissioning scenarios and reduce the risk of retrofitting. The method utilises a radiation map in a virtual simulation as the basis for the assessment of exposure to a virtual human. In this paper, we propose a fast simulation method using a known radiation source. The method has a unique advantage over point kernel and Monte Carlo methods because it generates the radiation map using interpolation in a virtual environment. The simulation of the radiation map including the calculation and the visualisation were realised using UNITY and MATLAB. The feasibility of the proposed method was tested on a hypothetical case and the results obtained are discussed in this paper.

Two decoupling methods for non-isothermal DSC results of AIBN decomposition.

During thermal decomposition of azobisisobutyronitrile (AIBN), the endothermic process of phase transition disturbed exothermic decomposition, which brought deformation in its thermal graphs. Therefore, exact kinetic parameters of the decomposition could not be obtained by the existing kinetics analytic models, and the accurate enthalpy data of the decomposition and phase transition were not available. Two methods, i.e., a solvent method and a mathematical method, were introduced in this paper to resolve the coupling phenomenon. In the former method, AIBN was dissolved into aniline to eliminate the endothermic process and obtain curves of the liquid-state decomposition. In the latter method, MATLAB software was employed to get the "pure" exothermic decomposition curve without the influence of phase transition by fitting coupling curves within the section after the transition point and extrapolating to the initial stage of decomposition. Moreover, the kinetic parameters of the "pure" exothermic decomposition of AIBN obtained by the mathematical fitting agreed with the results from the solvent method, verifying the accuracy of the decoupling. The research is of great significance for comprehending the exact characteristics of thermal behaviors and safety parameters of AIBN. It also provides a great help to determine the safe operating temperature and alarm temperature for processes in industry.