Search

In this study, complex-shaped Ti alloys with near-full density, using three-way pipe as an example, were achieved by the combination of 3D printing and cold isostatic pressing (CIP). A new CIP capsule consisting of porous internal support core and the external shell was prepared by 3D printing, which made it feasible to achieve internal and external co-pressing on Ti-6Al-4V powder during CIP. Therefore, complex-shaped green body was obtained with good surface quality, without obvious defects or cracks. After sintering at 1230 °C, the sintered sample had equivalent mechanical properties (hardness = 382 HV, elongation = 15.5%, tensile strength = 929 MPa and yield strength = 862 MPa) compared to the ASTMB381 standard. This study presents a novel method for fabricating complex-shaped Ti alloys with comprehensive mechanical properties.

Prostate cancer (PCa) is a carcinoma in which fatty acids are abundant. Fatty acid metabolism is rewired during PCa development. Although PCa can be treated with hormone therapy, after prolonged treatment, castration-resistant prostate cancer can develop and can lead to increased mortality. Changes to fatty acid metabolism occur systemically and locally in prostate cancer patients, and understanding these changes may lead to individualized treatments, especially in advanced, castration-resistant prostate cancers. The fatty acid metabolic changes are not merely reflective of oncogenic activity, but in many cases, these represent a critical factor in cancer initiation and development. In this review, we analyzed the literature regarding systemic changes to fatty acid metabolism in PCa patients and how these changes relate to obesity, diet, circulating metabolites, and peri-prostatic adipose tissue. We also analyzed cellular fatty acid metabolism in prostate cancer, including fatty acid uptake, de novo lipogenesis, fatty acid elongation, and oxidation. This review broadens our view of fatty acid switches in PCa and presents potential candidates for PCa treatment and diagnosis.

In order to improve the electrical quality disturbance recognition ability of theneural network, this paper studies a depth learning-based power quality dis-turbance recognition and classification method: constructing a power qualityperturbation model, generating training set; construct depth neural network;profit training set to depth neural network training; verify the performance ofthe depth neural network; the results show that the training set is randomlyadded 20DB-50DB noise, even in the most serious 20dB noise conditions,it can reach more than 99% identification, this is a tradition. The methodis impossible to implement. Conclusion: the deepest learning-based powerquality disturbance identification and classification method overcomes thedisadvantage of the selection steps of artificial characteristics, poor robust-ness, which is beneficial to more accurately and quickly discover the categoryof power quality issues.

OBJECTIVE: The present study provides an overview of studies investigating white matter (WM) integrity in patients with obsessive-compulsive disorder (OCD) using diffusion tensor imaging (DTI). Furthermore, it studies the correlation of fractional anisotropy (FA) in abnormal cerebral WM areas with the course and clinical signs of the disease. METHODS: The study subjects were divided into two groups, the OCD group (n=38) and the control group (n=40), based on the Diagnostic and Statistical Manual of Mental Disorders 5 (DSM-5) diagnostic criteria for OCD. Patients with untreated first-episode OCD were assigned to the OCD group, while healthy volunteers were assigned to the control group. The study group was evaluated in accordance with the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), Self-Rating Depression Scale (SDS) and Self-Rating Anxiety Scale (SAS). Subjects who met the inclusion criteria underwent whole-brain scanning via 3.0 T structural magnetic resonance imaging (sMRI). The WM FA values in different brain areas were compared between the two groups using voxel-based analysis (VBA). Subsequently, the correlations of the patient Y-BOCS score and disorder course with the FA values in significantly improved encephalic areas were analyzed. RESULTS: (I) The FA values of the right precentral gyrus (PreCG.R), left insular lobe, left inferior frontal gyrus and right inferior occipital gyrus (Occipital_Inf_R) WM were significantly lower in the OCD group than in the control group (P

The mixer settler is widely employed in the extraction and recycling of battery materials. The improvement of the separation efficiency is one of the most important issues. In order to investigate the effect of agitating speed on the flow characteristics and settling performance, both computational fluid dynamics coupled population balance model simulations and experiments with particle image velocimetry measurement have been performed in this work to investigate the flow field at different agitating speeds. The predicted data were in good agreement with the experimental data. The results demonstrated that an appropriate agitating speed could effectively promote the collision efficiency of organic phase droplets and accelerate the separation process. When the agitating speed was too large, the shear stress of the fluid would promote the breakage of the organic phase droplets and decrease the separation efficiency. This work will contribute to optimizing and designing large-scale stirring settlers.

Mixer-settlers have been widely employed in the rare earth element separation industry. Presently, reducing the loss of reagents and occupied areas and achieving a highly efficient separation in the settler are challenging issues. In this work, we report numerical simulations of the liquid–liquid flow in a stirring settler and thereafter describe the experimental validation. A computational fluid dynamics coupled population balance model (CFD-PBM) was developed to investigate the liquid–liquid flow characteristics and settling performance. The dispersion band thickness predicted by the turbulent aggregation model was in good agreement with the experimental measurements. The effects of the total liquid flow rate and initial average droplet diameter on the settling characteristics were further investigated. It was found that the dispersion band thickness increased significantly as the droplet diameter decreased. Moreover, this research shows that the CFD-PBM coupled model is promising for designing large-scale stirring settlers.

Medical image segmentation plays a vital role in clinical application of auxiliary diagnosis. Having said that, it is still a challenging task to accurately segment lesion areas from case images, because lesions are always different in size and shape, and the boundaries of lesion area is usually too rough, which greatly limits the accuracy of the current mainstream automatic segmentation algorithms. The lesions in stroke data are mostly small targets, and the boundary area are very tortuous, which greatly increases the difficulty of segmentation. Therefore, there is still a lot of space for segmentation of stroke data sets. In this paper, we propose a novel multiple encoders network for stroke lesion segmentation, called ME-Net, which is mainly composed of the atrous spatial pyramid pooling (ASPP) module and residual encoder. Particularly, ASPP uses a parallel structure of porous convolutions with different sampling rates to obtain context information for multi-scale features. The residual encoding can obtain high-level semantic features at high resolution. In addition, ME-Net adopts an encoding and decoding framework. We divided the pathological images into three levels according to the number of pixels in the masked area and evaluated our proposed method on an open dataset Anatomical Tracings of Lesions After Stroke (ATLAS) with encouraging performance achieved compared to other state-of-the-art approaches.

In order to better sort the massive scattered defect data and ensure the effect of data processing, an adaptive sorting method based on cloud computing is proposed. In order to improve the convergence of data, multi-objective processing is carried out in the cloud computing environment. Combined with the fuzzy adaptive sorting algorithm, the difference value of multi-objective variation is calculated to achieve the balanced adjustment of variation data. In order to speed up the data convergence, balance the local optimal sorting value, obtain the initial value of scattered defect data, and complete the adaptive sorting of scattered defect big data. Finally, experiments show that the adaptive sorting method based on cloud computing can better repair and accurately sort the scattered data, and improve the data convergence effect.

In order to solve the common problems of high interference and poor detection accuracy in the current multi-period power system data anomaly detection process, a cloud computing-based multi-period power system data interruption anomaly detection method is proposed. Combined with feature acquisition, the abnormal data features are collected in and out, and the acquired feature parameters are de-noised and frequency modulated by genetic algorithm. The final time limit is used to detect the abnormal interruption of data in multi-period power system. Finally, the experiment proves that the anti-interference performance of the method is obviously improved compared with the traditional detection method, and the accuracy of the detection result can be effectively guaranteed.

To improve the detection rate and reduce the correction error of abnormal data for water quality, an outlier detection and correction method is proposed based on the improved Variational Mode Decomposition (improved VMD) and Least Square Support Vector Machine (LSSVM) algorithms. The correlation coefficient is introduced, for solving the optimal parameter k of VMD algorithm, and an improved VMD algorithm is obtained. Combined with LSSVM algorithm, the outliers of water quality can be detected and repaired. This method is applied for the detection and correction of water quality monitoring outliers using dissolved oxygen which is retrieved from the water quality monitoring station in Hangzhou, Zhejiang Province, China. The result shows that the improved VMD algorithm is of higher detection rate and lower error rate than those of Empirical Mode Decomposition (EMD) and Ensemble Empirical Mode Decomposition (EEMD). The LSSVM algorithm increases the fitting accuracy and decreases correction error in comparison with SVM and BP neural network, which provides important references for the implementation of environmental protection measures.

The subjectivity contained in personal information in the context of big data makes it theoretically justified in private regulations. However, the changes in social structure and a series of data paradoxes caused by new technologies such as big data make traditional private law principles suffer. To the destruction, the logic of the market operation began to distort, and it was urgent to involve the public law and make up for it. Anonymization technology should be used to eliminate the subjectivity of data and personal information as much as possible, and a series of element indicators such as the subjectivity and publicity of information should be used as the criterion for determining the applicability of public and private law, that is high publicity and low subjectivity mainly based on public laws and regulations, data with low publicity and high subjectivity should be adjusted based on the principle of autonomy of private law to achieve a balanced application of public and private law in the protection of personal information in the context of data.

Under different ambient temperatures, the thermal aberration certainly affects the imaging quality of infrared multi-spectral camera. Therefore, an athermalized model of infrared multi-spectral cameras is established, and in this model the ambient infrared multispectral camera is equivalent to a separated dual-lens optical system. In the case of the fixed focal length, the influence of the back focal length on the change of the focal power of the front lens and back lens is analyzed. Now, the variation range of the front and rear lens interval is assumed to be restricted. When the back focal length is smaller than the focal length, the ratio of the absolute value of the focal power of the front lens to the absolute value of the focal power of the back lens decreases with the back focal length increasing. The material of the front lens and the back lens have a longer interval on the thermogram. When the back focal length is greater than the focal length, the scenario becomes exactly opposite. Combined with the judgment method of the positive value and negative value of the focal power on the thermogram, the selection range of materials is constrained by the positive value, negative value, and absolute value of focal power, thus realizing the rapid selection of the optical materials. This method is used to design an athermalized infrared multispectral camera with a waveband of 8–14 μm, a focal length of 50 mm, and an F number of 1.4 in a range from –40 ℃ to +60 ℃. Through the simulation analysis, the value of the athermalized infrared multispectral camera, at the Nyquist frequency of 30 lp/mm reaches 0.39, which is close to the diffraction limit; the root mean square radius of the diffuse spot is smaller than the Airy spot radius of 19.17 μm; the axial aberration is less than 0.02 mm, and the design results show that this method can make the long back-focus infrared optical system maintain stable imaging quality in a large temperature range. The SF6 gas is detected experimentally, and the experimental results demonstrate the excellent optical performance of the system.

A routing algorithm, based on a dual cluster head redundant mechanism combined with compressive sensing data fusion algorithm, is proposed to improve reliability and reduce data redundancy of the industrial wireless sensor networks. The Dual cluster head alternation mechanism is adopted to balance the energy consumption of cluster head nodes. Through the compressive sensing data fusion technology to eliminate redundancy, effectively improve the network throughput of the sensor network. The simulation results show that the proposed algorithm is able to enhance the networks performance, significantly reduces the number of lost packets and extend the network’s lifetime.

At present, free-to-move node self-deployment algorithms aim at event coverage and cannot improve network coverage under the premise of considering network connectivity, network reliability and network deployment energy consumption. Thus, this study proposes pigeon-based self-deployment algorithm (PSA) for underwater wireless sensor networks to overcome the limitations of these existing algorithms. In PSA, the sink node first finds its one-hop nodes and maximizes the network coverage in its one-hop region. The one-hop nodes subsequently divide the network into layers and cluster in each layer. Each cluster head node constructs a connected path to the sink node to guarantee network connectivity. Finally, the cluster head node regards the ratio of the movement distance of the node to the change in the coverage redundancy ratio as the target function and employs pigeon swarm optimization to determine the positions of the nodes. Simulation results show that PSA improves both network connectivity and network reliability, decreases network deployment energy consumption, and increases network coverage.

In order to study the effect of the fluid-structure interaction (FSI) on the simulation results, the external characteristics and internal flow features of a diffuser pump were analyzed with a two-way flow solid coupling method. And the static and dynamic structure analysis of the blade was also caculated with the FEA method. The steady flow field is based on Reynolds Averaged N-S equations with standard k-e turbulent model, the unsteady flow field is based on the large eddy simulation, and the structure response is based on elastic transient structural dynamic equation. The results showed that the effect of FSI on the head prediction based on CFD really exists. At the same radius, the van mises stress on the nodes closed shroud and hub was larger than other nodes. A large deformation region existed near inlet side at the middle of blades. The strength of impeller satisfied the strength requirement with static stress analysis based on the fourth strength theory. The dynamic stress varied periodically with the impeller rotating. It was also found that the fundamental frequency of the dynamic stress is the rotating frequency and its harmonic frequency. The frequency of maximum stress amplitude at node 1626 was 7 times of the rotating frequency. The frequency of maximum stress amplitude at node 2328 was 14 times of the rotating frequency. No matter strength failure or fatigue failure, the root of blades near shroud is the key region to analyse.

In power information system, the maintenance of middleware should promote from the traditional post inspection to the fault pre-diagnosis and intelligent early-warning. Prognostics and Health Management (PHM) method used for power oriented middleware has become an urgent research topic. Taking the middleware cluster as the research object, according to the data processing flow of PHM technology, researching on Grey-Markov forecasting, combining with time slice management and dynamic confidence threshold, this paper designs and implements a fault early warning model for the middleware cluster management. Taking the application of operations management platform in Hubei electric power company as an example. The method realizes the accurate warning of the middleware fault, and reduces the false alarm rate. The experimental results show that this method is feasible and effective.

Aimed at the limited energy of nodes in underwater wireless sensor networks (UWSNs) and the heavy load of cluster heads in clustering routing algorithms, this paper proposes a dynamic layered dual-cluster routing algorithm based on Krill Herd optimization in UWSNs. Cluster size is first decided by the distance between the cluster head nodes and sink node, and a dynamic layered mechanism is established to avoid the repeated selection of the same cluster head nodes. Using Krill Herd optimization algorithm selects the optimal and second optimal cluster heads, and its Lagrange model directs nodes to a high likelihood area. It ultimately realizes the functions of data collection and data transition. The simulation results show that the proposed algorithm can effectively decrease cluster energy consumption, balance the network energy consumption, and prolong the network lifetime.

Human computer interaction (HCI) is a classic topic, and hand gesture is one of the most natural languages, so the key point in human computer interaction is hand gesture recognition. A skin color model based on Hue is utilized to detect skin color regions like hands. To distinguish the face and hand, a simple and fast hand segmentation algorithm is developed. Moreover, we use the least square method to fit the trajectory of hand gravity motion and the angle and direction of the movement is got by the inverse trigonometric function of the slope of the fitted line. Ten people doing 80 hand gestures in low and high illumination environment are tested. Experimental results show that the accuracy is 94.375% in average and the processing time is 34ms per frame. These demonstrated the good real-time and robustness of the proposed approach.

The silicon nanoparticle have been grown by radio-frequency plasmas-enhanced chemical vapor deposition technique and the photoluminescence(PL)of the silicon suspension have been studied by both steady-state and time-resolved photoluminescence spectra.The steady-state PL spectra of the silicon suspension show that a blue PL band around 440 nm is found for the as-prepared silicon suspension emits.An increase of the intensity for the blue band around 440 nm followed by a red photoluminescence centered at 750 nm after storing the suspension for two months.The steady-state PL spectra excited by different wavelength and time-resolved PL spectra of the samples demonstrate that the blue PL band with a delaytime of nanosecond order is caused by the band-to-band recombination in Si nanocrystals.The oxidation of the silicon nanoparticle causes the decaytime increasing.The red PL band originates from radiation recombination of carries via defect/surface-related states and the delay-time of which is on the order of microsecond.

A high-order curvature-compensated CMOS bandgap reference (BGR) topology with a low temperature coefficient (TC) over a wide temperature range and a high power supply reject ratio (PSRR) is presented. High-order correction is realized by incorporating a nonlinear current INL, which is generated by ∆VGS across resistor into current generated by a conventional first-order current-mode BGR circuit. In order to achieve a high PSRR over a broad frequency range, a voltage pre-regulating technique is applied. The circuit was implemented in CSMC 0.5 μm 600 V BCD process. The experimental results indicate that the proposed topology achieves TC of 0.19 ppm/°C over the temperature range of 165 °C (−40 to 125 °C), PSRR of −123 dB @ DC and −56 dB @ 100 kHz. In addition, it achieves a line regulation performance of 0.017%/V in the supply range of 2.8–20 V.

Background and objectives: Patients with ankle sprain most commonly experience injuries to the lateral collateral ligament (LCL) complex, which affects the stability and function of the ankle. Early diagnosis of LCL complex injuries is extremely important. In clinical practice, magnetic resonance imaging (MRI) displays the soft tissues of the ankle well, but it is expensive and cannot be used for real-time evaluation. Ultrasound (US) has the advantages of ease of operation, low cost, and real-time capability. Some studies have reported on the stratification of injuries to the LCL complex using US. However, there is controversy regarding the injury grading. The purpose of this study was to investigate the grading of LCL complex injuries (sprain, partial tear, and complete tear) using US. Design: A prospective, open-label, diagnostic trial. Methods: Seventy patients with suspected ankle sprain who have an acute ankle sprain history and present with surgical indicators will be scheduled to receive ankle ligament repair surgery in the Guangzhou Orthopedics Hospital, China. On the day after admission, these patients will undergo US and magnetic resonance imaging (MRI) of the injured ankle prior to ankle ligament repair surgery (test group). The contralateral ankle will be included in the control group and will also undergo MRI and US examination prior to surgery. MRI or surgical results will be used as the gold standard for diagnosing LCL complex injuries. The feasibility of stratification of LCL complex injuries (sprain, partial tear, and complete tear) using US will be evaluated. Outcome measures and preliminary results: The primary outcome measure of this study is the sensitivity of US in the diagnosis of complete LCL complex tear. The secondary outcome measures are the sensitivity of US in the diagnosis of sprain and partial tear of the LCL complex; the specificity, positive predictive value, negative predictive value, accuracy, area under the ROC curve (AUC), and cut-off value of US in the diagnosis of sprain, partial tear, and complete tear of the LCL; and US imaging for the anatomy of the LCL complex. The results of 53 patients with acute ankle sprain included in the preliminary study suggest that there is no significant difference in the type of injuries to the talofibular ligament and calcaneofibular ligament between ultrasound diagnosis results and surgical or MRI findings (P > 0.05). Discussion: The results of this study will reveal whether US has a higher diagnostic value than MRI in the diagnosis of injuries to the LCL complex. Ethics and dissemination: This study was approved by Medical Ethics Committee of Guangzhou Orthopedics Hospital, China in May, 2018 (approval No. 2018-05). This study protocol will be performed in strict accordance with the Declaration of Helsinki. The personal information and data of the testing subjects participating in the clinical trials will be confidential and will be protected in accordance with applicable laws and regulations. The study protocol was designed in February 2018. Patient recruitment will begin in August 2018 and end in August 2019. Data collection will begin in August 2018. Data analysis will be performed in December 2019. The study will end in March 2020. Results will be disseminated through presentations at scientific meetings and/or by publication in a peer-reviewed journal. Trial registration: This trial was registered with the Chinese Clinical Trial Registry (registration number: ChiCTR1800016244). Protocol version: 1.0.

The output of bone-conduction aid is the vibration signal. Its performances affect the physical and mental health of the hearing impaired directly. The existing problems of bone-conduction hearing aid inspection system are analyzed and a novel one that can realize automatic inspection is presented in this paper. The system is designed in accordance with the international standards. Meanwhile, it can inspect many parameters of bone-conduction hearing aid such as maximum output force level, frequency response curves, full-on acoustic gain, total harmonic distortion and equivalent input noise level automatically. Through the amplitude scanning method based on SSR(Steady State Response) technology, the system can improve the inspection efficiency greatly and shorten inspection time. Due to inspection data being processed and outputted to inspection report automatically, the system reduces the requirements on the expertise of the inspectors and has the feature of easy-operation. The experiments show that the system can be competent for routine inspection work for bone-conduction hearing aid.

Due to the problem of low accuracy of uneven wear model of work roll, the works have been done as follows. Proceeding from distribution features of contact pressure between roll and strip, wear mechanism of roll is studied and work roll uneven wear model based on rolling pressure along roll barrel between roll and strip is proposed. The performance of the uneven wear model is enriched by adding k2 (unevenness coefficient of local wear), k3 (unevenness coefficient of local wear area) and k4 (local protuberance coefficient of `cat ear' wear). Industrial experiment shows wear prediction accuracy of work roll is higher and could be applied to industrial production.

We use nanosphere lithography and thermal evaporation techniques to fabricate multilayer metal-dielectric cone nanoparticle arrays. Scanning electron microscope measurements reveal that the ordered nanoparticles are of triangular shapes. Optical transmittance measurements show that there is a transmittance dip for each sample, and the transmittance dip is caused by the localized surface plasmon resonance of the metal/dielectric particles. With the increase of the number of the metal/dielectric layers of a sample, the position of the transmittance dip of the sample shows a blue-shift. Moreover, the transmittance spectrum and the electric field distribution of the metal/dielectric nanoparticles are simulated by HFSS, and the electric field distribution is used to explain the blue-shift phenomenon.

After long-term (1–14.5 years) postoperative follow-up observations of 407 cases of portal hypertension, the therapeutic effects of 204 cases which had undergone gastroesophageal devascularization (GEDV) and 203 cases which had undergon eportal systemic shunt (PSS) were compared. The long-term survival rate ot GEDV (89.0%) is higher than that of PSS (61.67%). The incidence of both bleeding and hepatic encephalopathy is much lower in GEDV than in PSS (6.90%, 0%, versus 15.75 %, 12.95% respectively). There is no significant difference in other respects, such as, pre and postoperative changes in varices, liver function, ascites and working capability. The observation also indicates that GEDV has the advantage of broader spectrum on case selection for operation. On the basis of our observation, we prefer GEDV to shunt procedures for patients of Child’s class C or patients with acute massive bleeding.

In the recent years, the mechanical robustness and reliability of three-dimensional (3D) integrated circuit (IC) packaging have become tremendously challenging due to its continuous reduction in size and increase in vertical integration. Thermal problems such as the silicon (Si) substrate cracking and interfacial delamination are increasingly dominant due to the presence of high heat density in the small multilayer IC packaging. Furthermore, as the manufacturing and processing steps of the 3D ICs are significantly different from that of the conventional ICs, new defect and failure mechanisms that are unique to 3D ICs are not well-investigated. The reduction in size has also increased the sensitivity of the 3D ICs to microdefects that had not previously threatened their reliability and integrity. The increased complexity and miniaturization of 3D IC packaging and its consequential problems demand for new failure analysis methods. Therefore, this PhD research is proposed to study the physical failure modes and mechanisms associated with 3D ICs via molecular dynamics (MD) simulation and finite element method (FEM), and to develop new failure analysis methods for packaging characterization, thereby improving the packaging techniques and processes. Si is commonly used as a substrate material in the IC packaging but yet its fracture behavior theoretical framework is still not well-developed. Thus, studies are first carried out on Si to understand its fracture mechanism through MD simulation and experiment. As the mechanical properties of Si are strongly influenced by the crystallographic orientation, defects, grain boundary (GB) and temperature, their effects on the tensile properties and failure mechanisms of Si are investigated. The results show that when Si is subjected to tensile loading, it undergoes brittle fracture and its (111) plane is the most favorable cleavage plane due to its high atomic packing density, thus resulting in the highest stiffness and lowest fracture strength. The presence of structural defects generally weakens the integrity of Si, particularly for the (100) oriented Si subjected to tensile loading in the [010] direction, its integrity is much more sensitive to these structural defects. The three-point bend test is also performed to investigate the crystallographic orientation and defects effects and the experimental results are found to be consistent with the MD results. The investigation of ∑25(710) tilt and ∑25(001)twist GBs in bicrystal Si nanofilms show that their fracture strengths are lower than that of single-crystal Si, which is caused by the change in crystallographic orientation of the grains. It is observed that these nanofilms generally fail through sliding dislocation mechanism. The reduction in size results in the increase of surface effect, and thus less energy is required for dislocation sliding to occur at the surface. The strength is further weakened by the addition of notch at the free edge of the GB. The increase in temperature (up to 600 K) has insignificant influence on the failure mechanism but generally lowers the fracture strength as expected. These combined simulation and experimental findings are then used to substantiate and explain the fracture behavior of bulk Si subjected at the macroscale, thereby improving the accuracy of root cause analysis. Furthermore, understanding and predicting interfacial delamination failure in the multilayered IC packaging are essential for their performance and reliability. The MD study is first performed to investigate the effect of voids and temperature on the interfacial fracture behavior in a-Si3N4/Si bilayer systems. When the bilayer system is subjected to tensile loading, its fracture behavior is found to be temperature-dependent. At 300 K, the interfacial strength of the bilayer system is ~22.5 GPa and it undergoes brittle fracture at the interface. However, failure occurs in the a-Si3N4 layer in a ductile mode when temperature is increased to 600K. With the existence of void in the Si layer in the a-Si3N4/Si system, crack initiates at the void and propagates towards the interface at 300 K. As the temperature increases to 600 K, the same bilayer system undergoes brittle fracture at the interface at a significantly lowered interfacial strength. With the inclusion of an interfacial void, the bilayer system fractures at the interface with a deteriorated strength regardless of the temperature. The simulation results also show that the presence of an interfacial void has a more adverse impact on the interfacial strength than having a void in the Si layer. Under the shear loading, the bilayer system deforms in three stages: elastic deformation, plastic flow in the a-Si3N4 layer and interfacial sliding. The presence of voids and the increase in temperature lower the stress required for interfacial sliding but they do not have significant effect on the shear deformation process. Subsequently, a combined approach based on MD-FEM simulations is adopted to characterize and predict delamination in a-Si3N4/Si bilayer system. The material parameters that are required to be input into the cohesive constitutive relation in FEM are derived from MD simulation. The work of interfacial fracture obtained by the FEM simulation is about 0.11 J/m2 which is in good agreement with the experimentally obtained value of about 0.085 J/m2. The above approach is also extended to characterize the interfacial energies and predict interfacial delamination in the Cu/Ti/SiO2/Si multilayered system. The thermodynamic work of adhesion W derived from MD simulation is used as the failure criterion for the FEM simulation as it is independent of the system geometry and applied strain rate. The interfacial fracture energy of 8.23 and 9.79 J/ m2 obtained from the MD-FEM simulations and the indentation damage experiment respectively, are also in good agreement. Hence, the combined approach is accurate and effective in characterizing and predicting the interfacial delamination in the multilayered IC packaging. This PhD dissertation has investigated the fracture behaviour of bulk Si, bilayer systems and multilayered systems to understand their failure mechanisms and help in the failure analysis process. This study is able to provide useful information for 3D ICs design and applications and helpful in improving their stability and reliability. Doctor of Philosophy (MAE)