Search

By violating semantic constraints that the control process impose, the semantic attack leads the Industry Control Systems (ICS) into an undesirable state or critical state. The spread of semantic attack has caused huge economic losses and casualties to critical infrastructure. Therefore, detecting semantic attack is referred to an urgent and critical task. However, few existing detecting techniques can achieve satisfactory effects in detecting semantic attack of ICS, due to the high requirements of complete critical state-based semantic behavior features description, joint detection on multivariate type state variables, and validity of field states datasets under semantic attacks. In an effort to deal with above challenges, We label device states databases with temporal characteristics and divide impacts on states of field devices under semantic attacks into three categories, including absent in states set, confused sequences, irregular frequency. On this basis, we establish a behavioral model based on secondary labeling of states-duration evolution graph (BMSLS), then implement an adaptive secure state-based semantic attack detection framework furtherly. Compared with the traditional Auto Regression (AR) algorithm, the newer time series correlation graph model, and other five deep learning algorithms, our proposed framework demonstrates the superior effect on the detection of semantic attack.

Designing a secure authentication scheme for session initial protocol (SIP) over internet protocol (VoIP) networks remains challenging. In this paper, we revisit the protocol of Zhang, Tang and Zhu (2015) and reveal that the protocol is vulnerable to key-compromise impersonation attacks. We then propose a SIP authenticated key agreement protocol (AKAP) using elliptic curve cryptography (ECC). We demonstrate the correctness of the protocol using Burrows-Abadi-Needham (BAN), and its security using the AVISPA simulation tool. We also evaluate its performance against those of Zhang, Tang and Zhu, and others.

Mobile cloud computing (MCC) is an emerging computing model with the growing demands of mobile devices by services diversity using mobile cloud technology. Authentication plays a fundamental role between the smart portable devices located in different places and the mobile cloud computing service providers in MCC. Several authors presented many authentication schemes using different cryptographic approaches for MCC in the past years. However, those schemes have not solved the impersonation attack well. To conquer this suffering and support more security functionalities, we propose an impersonation resistance for a biometric-based authentication scheme using hashing and symmetric parameter function in an MCC environment. The proposed scheme can provide mutual authentication and anonymity using the software AVISPA. The security of the proposed scheme is formally proved in the random oracle model. Regarding functionality and performance analysis, the comparative studies demonstrate that the design has a trade-off between resource consumption and security strengthens for MCC services. Therefore, the proposed scheme is more practical in MCC application.

Hydraulic flushing drilling technology can not only improve the efficiency of high-pressure and low-permeability geological reservoir coalbed methane drainage but also effectively reduce the probability of coal and gas outburst disasters through pressure relief. The main mechanism of this technology is to expand the borehole diameter through hydraulic flushing measures, increase the strain of the coal around the borehole, and increase the development of cracked pores, to improve the permeability of the coal seam and realize the dual reduction of ground and gas pressure. However, in the actual application process, the interaction mechanism among the stress field, the structure field, and the seepage field is still not clear, and there is no clear method to accurately determine the pressure relief range based on the pressure relief mechanism in order to carry out reasonable drilling arrangements. Therefore, this article comprehensively uses laboratory experiments, numerical simulations, and field practices to fully explain the hydraulic flushing pressure relief mechanism and proposes a method to accurately determine the pressure relief range based on the radial line strain law. The results based on radial line strain showed that the effective relief radius expands to 0.86 m once adopting the Φ579 mm hydraulic flushing borehole compared to Φ160 mm; the borehole’s equivalent diameter of drilling field #11 is 2 to 3 times than that of #10 and 1.2 times the average CBM extraction amount. Therefore, as the borehole diameter increases, the permeability and radial line strain of the coal around the borehole increase significantly, but the tendency of the increase in permeability decreases with increasing vertical stress. The findings of this study can help for a better understanding of the pressure relief and permeability enhancement mechanism of hydraulic flushing, and the method of determining the pressure relief range based on radial strain can also provide a new way for other mines to practice ideas.

Purpose To investigate and compare the CT and MRI features of hepatic sclerosed hemangioma (HSH) and sclerosing cavernous hemangioma (SCH). Materials and methods Twelve HSH cases and 36 SCH cases were included, the imaging findings on CT (9 HSH and 34 SCH) and MRI (8 HSH and 10 SCH) were analyzed. Qualitative image analysis included the location, size, shape, capsular retraction, density, calcification, signal intensity on T1-weighted image (T1WI) and T2-weighted image (T2WI), presence of diffusion restriction, apparent diffusion coefficient (ADC) map, transient hepatic attenuation difference around the lesion, and the dynamic enhancement patterns. Results The presence of liver cirrhosis in patients with HSH (3/12) was higher than SCH (1/36) (P = 0.043). The morphology appearance before enhancement showed no significant difference between HSH and SCH. Moreover, SCH had a stronger trend of centripetal enhancement patterns of cavernous hemangiomas (83.3%) compared to HSH (25%) (P P P = 0.009, P = 0.002); however, there was no significant difference in ADC values between themselves (P = 0.613). Conclusion SCH showed the same trend of centripetal enhancement characteristics as typical hemangioma, while HSH exhibited atypical enhancement features due to complete sclerosis. Higher ADC values might contribute to the identification of atypical HSH and SCH from malignancies.

Nowadays, the valid development and full utilization of renewable energy sources such as wind power and solar energy are considered to be one of the most significant approaches to get rid of the energy dilemma and realize environmental governance. However, although there are more and more renewable power sources connected to power systems, the renewable energy has not been fully used yet. In this context, an optimization model for the regional power system with the pump storage power, wind, hydro, and thermal power is proposed to enhance the economy, environmental protection, energy-saving and stability of the power system. Besides, the fuzzy chance constraint programming and the improved analytic hierarchy process are used to deal with the uncertainty of the wind power and the load demand and the multi-objectives function, respectively. The case study not only verify the effectiveness of the proposed model but also shows that better comprehensive performance of the power system can be achieved through the proposed model when considering the effect of the pump storage power.

Deep learning has recently demonstrated its promising performance for vision-based parking-slot detection. However, very few existing methods explicitly take into account learning the link information of the marking-points, resulting in complex post-processing and erroneous detection. In this paper, we propose an attentional graph neural network based parking-slot detection method, which refers the marking-points in an around-view image as graph-structured data and utilize graph neural network to aggregate the neighboring information between marking-points. Without any manually designed post-processing, the proposed method is end-to-end trainable. Extensive experiments have been conducted on public benchmark dataset, where the proposed method achieves state-of-the-art accuracy. Code is publicly available at \url{https://github.com/Jiaolong/gcn-parking-slot}., Comment: Accepted by RAL

Pulmonary inflammatory myofibroblastic tumors (IMTs) are rarely reported in adult males. Given the low incidence of IMT and the lack of imaging references and pathological guidance, the misdiagnosis rate of IMT is high. In this article, we describe two cases of IMTs in the lungs. Both patients were adult males with lesions in the right lobe, a history of pulmonary tuberculosis, and a long period of refractory intermittent pulmonary inflammation. Our two male patients both experienced intermittent cough symptoms, but pulmonary IMTs were not suspected for a long time. Both patients were diagnosed with pulmonary tuberculosis before IMT was confirmed and treated with isoniazid (H), rifampin (R), pyrazinamide (Z), and ethambutol (E) (HRZE) or isoniazid (H), levofloxacin (L), pyrazinamide (Z), and ethambutol (E) (HLZE) for months. In Case 2, we observed multiple subpleural cord signs in the left lung, soft tissue mass shadows at the apex of the right upper lobe, a thickened interlobular interval, and scattered patches and nodules in the upper right lung. These features are novel in the identification of IMTs. Both of the pathological findings revealed a great deal of myofibroblasts, fibroblasts and collagen fibers in the lower right lung lesion, accompanied by a large number of plasma cells and foam cell infiltration, which were consistent with the features of IMT. The two patients displayed exceedingly different symptoms, computed tomography (CT) imaging features, and pathological results from those reported in traditional records. These findings provide novel references that will extend understandings of this rare disease.

The feedback vertex set (FVS) problem of network aims to find a vertex subset of smallest size that contains at least one vertex of each cycle of the network. The FVS problem has found wide applications, ranging from circuit design, network security and network control and observation etc. However, the majority of existing studies on FVS problem are confined to single networks. In this brief, under the consideration of one kind of control problem of multiplex network, we introduce the concept of FVS of multiplex network. The control problem of multiplex network is naturally mapped into the FVS problem of multiplex network. We then propose two simulated annealing (SA) algorithms to solve the directed FVS problem of multiplex network and undirected FVS problem of multiplex network respectively. The effectiveness of the SA algorithms is finally verified on both artificial multiplex networks and real-world multiplex networks.

The dismantling problem and minimum vertex cover (MVC) problem of network are two fundamental NP-hard problems where the former aims to find a minimal subset of nodes whose removal leaves the network broken in small components of sub-extensive size and the latter is also to find a minimal vertex set which contains at least one incident vertex of every edge. Both of them have a wide range of applications related to circuit design, communication, network security, transportation, epidemic control, molecular biology and economics. In this brief, we first propose a novel belief-propagation equation for the spin glass model of the MVC problem. A belief-propagation-guided decimation (BPD) algorithm is then presented which could construct approximate optimal vertex cover of network. In addition, based on the relationship between the network dismantling problem and the MVC problem, we propose a simple and fast algorithm formed by the MVC and a general node reinsertion strategy, for solving the network dismantling problem. The effectiveness of the proposed algorithms is finally verified by comparing with other well-known algorithms on a number of real world networks.

In this investigation, the welding quality and the dynamic resistance signal in the welding process were correlated using different dimension reduction techniques and regression models. The 0.4-mm-thickness TC2 titanium alloy was used as the welding material, while the welding experiments were carried out by a small pneumatic high-frequency alternating current welding machine. The Rogowski coil and alligator type wire clips were respectively utilized to collect the welding current and voltage in the welding process. The mechanism of dynamic resistance signal changes was studied and explained. In order to extract the features related to the welding quality and reveal the efficiency of the automatic feature extraction method based on the dimension reduction techniques, several dynamic resistance signals with different welding process parameters were processed by the approaches of principal component analysis (PCA), isometric mapping (Isomap), and locally linear embedding (LLE). After that, the redundant information in the dynamic resistance signal which has little to do with the welding quality would be removed and the useful features could also be isolated. And then three regression models quantifying the extracted features and the nugget size were created and their performances were also compared with each other. The results implied that the regression model based on the LLE technique was more robust than those established on the basis of the features automatically extracted from the dynamic resistance signal using PCA and Isomap. Compared with the existing popular manual feature extraction methods in the previous research work, the method proposed in this investigation outperforms the other methods in terms of two aspects. It not only can minimize the prior assumptions about the certain shape of the dynamic resistance curve and remove the subjective factors caused by the manual extraction method, but can assess and monitor the welding quality with a good level of reliability.

Incorporating plasmonic nanostructures is a promising strategy to enhance both the optical and electrical characteristics of photovoltaic devices via more efficient harvesting of incident light. Herein, we report a facile fabrication scheme at low temperature for producing gold nanoparticles embedded in anatase TiO2 films, which can simultaneously improve the efficiency and stability of n-i-p planar heterojunction perovskite solar cells (PSCs). The PSCs based on rigid and flexible substrates with 0.2 wt % Au-TiO2/TiO2 dual electron transport layers (ETLs) achieved power conversion efficiencies up to 20.31 and 15.36%, superior to that of devices with TiO2 as a single ETL. Moreover, 0.2 wt % Au-TiO2/TiO2 devices demonstrated significant stability in light soaking, which is attributed to improved light absorption, low charge recombination loss, and enhanced carrier transport, and extraction with the plasmonic Au-TiO2/TiO2 dual ETL. The present work improves the practicability of high-performance and flexible PSCs by engineering the photogenerated carrier dynamics at the interface.

Due to the complicated coalbed methane (CBM) occurrence conditions and the diverse geological structures in China, the promotion and application of the coal and gas simultaneous extraction technology have been seriously restricted. In view of this, this paper chooses Qingdong Coal Mine protection layer mining and CBM extraction field practice as the research background. Firstly, based on the similar material simulation experiment that simulates coal mining, the dynamic changing pattern of a mining field’s overburdened strata and corresponding stress are obtained, the relationship between gas desorption and stress can then be clarified. Further, with the help of the fractal theory and box counting method, the fracture development characteristics of the overlying strata are quantitatively described on the basis of experimental images. Finally, by building a model for calculating the penetrability coefficient of coal seam based on fractal dimension of mining fissure and analyzing the relationship between fissure development and fractal dimension, the gas migration law and the fissure development areas of #7 and #8 overburden strata where CBM concentrates can be revealed and determined. According to the orientation of the area mentioned above, the location of the CBM pumping field in relation to the coal seam roof and the arrangement of CBM extraction boreholes can be optimized, which make CBM extraction efficient. Meanwhile, the risk of coal and gas outburst is significantly reduced when the CBM concentration is controlled within 0.2% to 0.6% outside the corner of the working face and 0.1% to 0.35% in return flow, which is lower than 0.8%, the threshold of CBM concentration.

Coal seam water injection is a widely used dust reduction technology in Chinese coal mines. During the process of coal seam water injection, a large amount of broken coal particles will accumulate in the pore, which will affect the flow characteristics of water. It's very important for improving the on-site coal seam water injection and dust reduction technology to study the influence of coal particles on the water migration law. In this paper, a spontaneous upward imbibition experiment was used to study the effect of coal particles stacked in front fractures on water migration in coal seam water infusion. Then, a mathematical model was established to express imbibition speed and stable imbibition height. The results show when the largest particle size is selected, theoretical calculation results of stable imbibition height is closest to the experimental data; coal particles with a size of 564–1589 μm have the greatest influence on the wetting phase of coal seam water infusion; the metamorphic grade and particle size of the samples mainly affect the initial stage of water migration, and the initial imbibition speed of each coal sample is quite different. However, with increasing imbibition height, the imbibition speed difference of each coal sample gradually decreases.

Organic-inorganic hybrid perovskite solar cells (PSCs) have achieved power conversion efficiency (PCE) from 3.8% to 25.2% in recent years. However, the charge transport ability of the transporting layers and the quality of crystalline films limit further development of PSCs. To fabricate highly efficient PSCs, effective charge transport to electrodes through transporting layers is a necessary requirement. Suitable interface engineering is very critical because it can not only benefit the charge extraction and transport, but also effectively passivate the traps on the surface and grain boundary of the perovskite. It has been shown that the passivation layer for the perovskite plays a vital role in the performance of PSCs. The close fit between the charge transport layer and the perovskite layer is required to reduce the interfacial recombination and accelerate the carrier extraction and collection. Besides, the appropriate passivation layer has a potential impact on the crystallization and the morphology of the perovskite film. To achieve high-performance devices, it is crucial to design an effective passivation layer with appropriate nanomaterials. In this work, we synthesized poly[(2-methoxy,5-octoxy)1,4-phenylenevinylene] (MOPPV) on few-layer graphene oxide (GO) by in situ polymerization at room temperature. MOPPV-GO can fully combine the advantages of the two materials. The novel nanocomposite MOPPV-GO is used as the ultra-thin passivation layer of PSCs, of which the structure is ITO/TiO2/(FAPbI3) x (MAPbCl3)1– x /MOPPV-GO/spiro-OMeTAD/Ag. Since the surface morphology of the perovskite layer would affect the performance of devices significantly, the perovskite films with and without the passivation layer by atomic force microscopy is found that the MOPPV-GO can facilitate the growth of perovskite crystal with larger grains and enhance the quality of perovskite films. The root mean square roughness values of the perovskite films with and without MOPPV-GO were evaluated to be 19.87 and 25.85 nm, justifying the improvement of the surface flatness of the perovskite film with the decoration of MOPPV-GO, which would enhance the contact and carrier transport between the hole transport layer and the perovskite layer. As a result, the PSCs with MOPPV-GO exhibited considerably low hysteresis indices (0.02) and showed a short circuit current density of 23.54 mA cm–1, an open circuit voltage of 1.09 V, and a fill factor of 76.46%, corresponding to a high power conversion efficiency of 19.73%. By contrast, the control PSCs without MOPPV-GO showed a high hysteresis index of 0.18, with a poor power conversion efficiency of 16.42%. Meanwhile, it is shown that the EQE curve of MOPPV-GO modified PSC is higher than that of unmodified devices, which is consistent with the experimental results of J-V curve. The steady-state power output of various devices in ambient conditions under continuous AM 1.5G illumination was tested, which shows excellent stability of the PSC with MOPPV-GO during the period of 1000 s, indicating the positive contribution of MOPPV-GO interlayer to the stability. This work could provide a new idea for the synthesis and application of graphene-based materials in solar cells.

In recent years, offshore wind power has been vigorously developed. High voltage AC transmission is one of the main grid connection technologies for offshore wind power. Submarine cable has more charging power causing the problem of overvoltage more prominent. To solve the problem of line overvoltage in the AC transmission system of the offshore wind power, this paper proposes a coordinated reactive power control strategy. First, the reactive power generated by wind turbine is considered to reduce the configuration capacity of the reactive power compensation devices. Second, the reactive power compensation is divided into two parts: the land side and the wind power collecting side. Then, with the objective function of minimizing the power loss and the voltage offset, the assignment of the optimal amount of reactive power on both sides is solved. Finally, simulation results prove that the line overvoltage suppression effect of dual-terminal coordinated control strategy proposed in this paper is both economical and effective.

The separator is of great importance for regulating ion transmission, maintaining electrode stability, and device safety in lithium‐ion batteries. Despite their many advantages, glass microfiber separators have shortcomings that often give rise to poor cycle performance and fatal short‐circuit risk when used in flexible batteries. Here, we propose the use of a scalable non‐Newtonian fluidic cellulose permeation‐diffusion strategy to develop a cellulose/glass microfiber composite film with an hourglass‐gradient structure. Due to the unique gradient morphology resulting from the presence of cellulose macromolecules, the as‐prepared composite film showed an improved surface mass density, adjustable pore structure, controllable ion transport, ideal mechanical flexibility, and robustness. When used as the separator in an assembled lithium‐ion battery, the composite film exhibited rapid ion diffusion and Li dendrite inhibition, which endowed the battery with excellent cycle stability (specific capacity of 108.5 mAh g−1 after 1000 cycles) and good rate performance. This composite film shows great potential application in flexible lithium‐ion batteries including but not limited to pouch cells.

Digital microscopes often require repeated manual focusing to obtain a clear image. Nevertheless, manual focusing easily results in artificial errors, making it difficult to evaluate the image definition, and the focusing process is slow and tedious. This paper presents an automatic focusing method based on Quantum Self-organizing Maps (QSOM) neural network and a new focusing evaluation function. The focusing evaluation function consists of Energy of Gradient (EOG) function and Discrete Wavelet Transform (DWT) function to better evaluates the sharpness of microscopic images at different focusing positions. The obtained data are used as training samples, and QSOM neural network is trained by focusing samples. The trained neural network can accurately predict the position achieve the focus position. Experimental results are given to verify the effectiveness of the proposed method.

In multi-label learning, in order to improve the accuracy of classification, many scholars have considered the relationship between features and features, features and labels or labels and labels, but how to combine the correlation among them is rarely studied. Based on this, this paper proposes a multi-label learning algorithm with kernel extreme learning machine autoencoder. Firstly, the label space is reconstructed by using the non-equilibrium labels completion method in the label space. Then, the non-equilibrium labels space information is added to the input node of the kernel extreme learning machine autoencoder network, and the input features are output as the target. Finally, the kernel extreme learning machine is used for classification. Our method implements the information fusion between features and features, between labels and features, and between labels and labels. Compared with the traditional autoencoder network, the extreme learning machine autoencoder has no iterative process, which reduces the network training time and improves the classification accuracy. The experimental results of the proposed algorithm in the opening benchmark multi-label data sets show that the KELM-AE algorithm has some advantages over other comparative multi-label learning algorithms and the statistical hypothesis testing and stability analysis further illustrate the effectiveness of the proposed algorithm.

Efficient security patch distribution is of essential importance for updating anti-virus software to ensure effective and timely virus detection and cleanup. In this paper, we propose a mixed strategy of patch distribution to combine the advantages of the traditional centralized patch distribution strategy and decentralized patch distribution strategy. A novel network model that contains a central node and a multiplex network composed of patch dissemination network layer and virus propagation network layer is presented, and a competing spreading dynamical process on top of the network model that simulates the interplay between virus propagation and patch dissemination is developed. Such a new framework helps in effectively analyzing the impacts of patches distribution on virus propagation, and developing more realizable schemes for restraining virus propagation. Furthermore, considering the constraints of the capacity of the central node and the bandwidth of network links, an optimal allocation approach of patches is proposed, which could simultaneously optimize multiple dynamical parameters to effectively restrain the virus propagation with a given budget. Accepted version

BACKGROUND Pathological manifestations of hepatic tumours are often associated with prognosis. Although surgical specimens (SS) can provide more information, currently, pre-treatment needle core biopsy (NCB) is increasingly showing important value in understanding the nature of liver tumors and even in diagnosis and treatment decisions. However, the concordance of the clinicopathological characteristics and immunohistochemical (IHC) staining between NCB and SS from patients with hepatic tumours were less concerned. AIM To introduce a more accurate method for interpreting the IHC staining results in order to improve the diagnostic value of hepatic malignancy in NCB samples. METHOD A total of 208 patients who underwent both preoperative NCB and surgical resection for hepatocellular carcinoma (HCC) or intrahepatic cholangiocarcinoma (ICC) between 2008 and 2015 were enrolled in this study. The expression of CK19, GPC3, and HepPar1 were detected by IHC staining. Clinicopathological, NCB, and surgical data were collected and analysed using χ2 and kappa statistics. RESULTS Morphologically, the presence of compact tumour nests or a cord-like structure in NCB was considered the primary cause of misdiagnosis of HCC from ICC. The kappa statistic showed a moderate agreement in histomorphology (k = 0.504) and histological grade (k = 0.488) between NCB and SS of the tumours. A 4-tier (+++, ++, +, and -) scoring scheme that emphasized the focal neoplastic cell immunoreactivity of tumour cells revealed perfect concordance of CK19, GPC3 and HepPar1 between NCB and SS (k = 0.717; k = 0.768; k = 0.633). Furthermore, with the aid of a binary classification derived from the 4-tier score, a high concordance was achieved in interpreting the IHC staining of the three markers between NCB and final SS (k = 0.931; k = 0.907; k = 0.803), increasing the accuracy of NCB diagnosis C (k = 0.987; area under the curve = 0.997, 95%CI: 0.990-1.000; P < 0.001). CONCLUSION These findings imply that reasonable interpretation of IHC results in NCB is vital for improving the accuracy of tumour diagnosis. The simplified binary classification provides an easy and applicable approach.

Large-scale wind and solar power generations have got rapid development in recent years in China and abroad. They always connect to load centres through high-voltage direct current (HVDC) transmission or ultra HVDC (UHVDC) transmission lines. Such systems are so large and it is necessary to make appropriate equivalence according to the purpose of different research. At present, the analysis model of this kind of real system is few, and simplified single-machine infinite-bus system or IEEE example system is adopted most of the time, which may affect the accuracy of analysis results. Here, the UHVDC project from a practical renewable energy base to load centres in China is chosen as an example, the actual structure and parameters are used to establish a typical example system for AC/DC hybrid power grid with high-proportion renewable energy. The new-generation synchronous condenser with large capacity and the supporting synchronous generators are all considered. First, the installed power generation, power grid structure, and load configuration of example system are introduced. Then the load flow distribution and transient stability characteristics are analysed by using DIgSILENT PowerFactory software. Finally, the feasibility of the proposed example system is shown by simulation results.

With the rapid development of power system, wind power becomes an important generating way in AC/DC hybrid power grid. Due to the interference between the wind power and the AC network, the risk of DC system continuous commutation failure will be affected. In order to evaluate this effect reasonably, a method for running out of the fault set is presented in this study. Firstly, a distribution factor is proposed to determine the power line transfer after wind power or power line fault, the line with maximum load rate is chosen as the candidate failure lines; Secondly, the η short-circuit ratio and η effective short-circuit ratio indexes of the DC system is evaluated after every fault; they are used to judge whether the cascading fault of AC system induces the transient overvoltage of the converter bus to lead to wind turbines to take off the power grid net. Thus, a cascading fault sequence of the AC system is formed, and a cascading fault set that causes the transient overvoltage of the converter bus is completed. Finally, the simulation results calculated by power system department-bonneville power administration show that when the commutation failure occurs in the DC system, the system runs out immediately, and the validity of the method is verified.

High share of renewable energy will be one of the basic features of power systems in the future around the world. Compared with photovoltaic (PV) and wind power, concentrating solar power (CSP) equipped with heat storage system has similar regulation characteristics with thermal power units, and it can effectively overcome the intermittent and volatility of PV and wind power generation. Meanwhile, the CSP units can strongly support the voltage, frequency and short-circuit capacity of power systems. So, CSP is a very promising power generation technology for the future power system. The CSP unit has better ramp ability than thermal power unit, and can generally reach a speed between 2% and 5% of rated power per minute. Stated thus, the CSP unit has a good potential to participate in the primary frequency regulation of power grid. At present, there are many researches on the performance of each subsystem of the CSP unit in worldwide, and there are also some theoretical researches on the participation of CSP unit in the dispatching operation of power system. However, there is a lack of research reports on CSP plant’s participation in the primary frequency regulation of the power grid, which is an urgently needed feature for the future power systems with high share of renewable energy. In this paper, a model with primary frequency regulation ability of CSP unit is established in DIgSILENT PowerFactory software. A revised two-area four-machine system is used to investigate the CSP unit’s participation in primary frequency regulation of power system. The simulation results show the good frequency regulation performance of CSP unit, which can effectively support the stable operation of power system with high share of renewable energy in the future.

Integrated energy distribution system (IEDS) could improve the efficiency of energy utilization and minimize environmental pollution based on the multi-energy coupling link. Firstly, the model of distributed energy station is established; secondly, the optimization model for optimal dispatching of IEDS considering AC power flow of distribution system, gas flow of natural gas distribution network and the distributed energy station is established; thirdly, an optimization method for optimal dispatching of IEDS based on the second-order cone relaxation method is proposed. Finally, case studies are performed on IEDS including the modified IEEE 33-node distribution system, the modified 11-node natural gas system and 5 distributed energy stations to verify the effectiveness and efficiency of the proposed method.

A flexible load active management method in optimization operation of distribution networks is proposed to discuss the significant role of active management of flexible load in regulating load curve. The models of translatable load, shiftable load and curtailable load are established and they are divided into price incentive flexible load and contract incentive flexible load firstly. Secondly, a multi-objective optimal dispatch model considering electricity purchasing cost, load peak valley difference and users' acceptance is established for different types of flexible load. Then, considering the proportion of different types of flexible load in different user types, the flexible load response model of residential load, industrial load and commercial load are obtained. Finally, the regulation effects of load curve and dispatch strategies of the overall flexible load are obtained by summarizing the dispatch strategies of different user types.

The focus of financial work is risk prevention and control, and this is also an important part of the overall national security concept. China has repeatedly stressed that strengthening financial supervision, preventing and defusing financial risks, especially systemic financial risks are the fundamental tasks of financial work. Strengthening financial supervision requires financial institutions to continuously enhance compliance awareness and take the initiative to improve compliance management. Relying on new technologies such as big data, blockchain, and artificial intelligence (AI), compliance technology (CompTech) strives to provide faster and more effective supervision schemes for financial institutions. Given the complexity of financial transactions and the huge amounts of data involved, AI and machine learning are playing an increasingly important role for financial institutions and regulators. CompTech needs to grasp the trend of AI, realize timely and efficient supervision and compliance, maintain financial security, and prevent system risks.

Increasingly applied to the financial market over recent years, AI is on track to alleviate the pain points of the market, promote the fast, sustainable development of the financial industry and spearhead transformation of the market. Compared with practices abroad, the application of intelligent finance is at a nascent stage in China, where the regulatory system still requires improvement. In this context, the Chinese financial industry should strengthen self-discipline, enhance the quality and efficiency of financial services, and control the risk of technological innovation to maintain the steady development of the industry.

Since the coining of the term “Artificial Intelligence” (AI), its definition scope has evolved from an application program with pre-set rules of logic, to a technology that simulates human cognitive function. The goal of AI is to allow computers to imitate human intelligence, perception, thinking, and action, so as to achieve automation and gain analytical insights. To imitate human intelligence, AI applications adopt two methods: rule-based and non-rule-based algorithms. Rule-based AI uses pre-set rules to “learn”, while non-rule-based AI uses machine learning algorithms to “learn” and a trained AI model to “think”. The key driving force for the development of AI is the huge increase of available data and the great improvement of computing power in traditional computers and mobile devices, as well as the continuous development of machine learning algorithms. Commercial banks conduct large numbers of customer capital transactions, and hold large amounts of account management information. The application of AI in commercial banks will effectively improve the retail service level, enhance customer experience, and alleviate difficulties such as “financial disintermediation” and fierce market competition, or become the next profit growth point of commercial banks.

Since 2013, when Yu’E Bao ushered in an era of Internet wealth management in China, the country’s “Internet + Wealth Management” market has developed rapidly. Online wealth management meets the demand for wealth management by long-tailed investors, which cannot be fulfilled by traditional wealth management modes. It will provide more constructive policy suggestions to the benefit of the better integrated development of wealth management industry and AI in China in the future by searching thoroughly the history and environment of online wealth management, learning from the effective experience of wealth management by AI abroad and combining with the current integrated development of “AI + wealth management” service industry in China.

The change of payment mode from barter to natural currency, to paper money and today’s e-payment, is a progressive process that continuously improves transaction efficiency on the premise of security and reliability. Today, with the rapid development of information science and technology, what will be the next innovation of payment mode? Already, artificial intelligence has penetrated deeply and extensively into all levels of payment, and is continuously evolving towards further intelligence and diversification. It plays a vital role in elevating payment efficiency, improving user experience, and implementing financial regulation.

A robo-advisor is an investment advisor based on core artificial intelligence technology. The technology encompasses perceptual intelligence, related to human sensory awareness such as vision and hearing; and cognitive intelligence, linked to human language, understanding and logic, such as learning, reasoning, thinking, and planning. At the current stage of development, perceptual AI mainly comprises image recognition technology, including iris and facial recognition, and unmanned driving as regards vision; and speech synthesis, voiceprint recognition, and speech evaluation as regards hearing. Cognitive AI, which is applied in games, chess, and most notably in natural language processing (NLP), includes semantic analysis, pragmatic analysis, machine translation, text matching, text classification, and dialogue system. Hence, AI technology must correspond to at least one of “eyes, ears and mouth”. Any non-human advisor in the financial industry that does not involve one of these three is simply an application of traditional network engineering technology, and cannot be included under the term robo-advisor.

With the deep integration of big data, cloud computing, artificial intelligence, Internet of Things, and other information technologies with financial services, financial boundaries have been repeatedly broken and reconstructed, while scientific and technological forces are providing continuous innovation vitality for financial services. FinTech, as defined by the Financial Stability Board (FSB) in 2016, is technology driven financial innovation, aimed at transforming or innovating financial products and services by using existing high-tech achievements, planning modern business models, and optimizing business processes, so as to promote continuous financial transformation and upgrading and improve quality and efficiency.

As the penetration rate of new energy sources such as wind power in the power grid continues to increase, the power grid generally puts forward requirements for its participating in the primary frequency regulation of power grid. At present, the commonly adopted strategy for wind farms to participate in frequency regulation is that each wind farm should participate in primary frequency regulation of power grid, and frequent adjustment will bring a certain degree of life loss to wind power equipment. On account of this reason, this paper presents a coordinated control strategy for the frequency regulation of wind farms based on a multi-layer hierarchical structure. When the grid frequency fluctuates, the active power of some wind farms with superior performance can be adjusted according to specific circumstances to complete frequency regulation target. The simulation results verify the effectiveness of the strategy and achieve a good frequency regulation effect.

Background Hepatocellular carcinoma (HCC) is the most common malignant tumor of the liver, and its morbidity and mortality have been increasing in recent years. The early diagnosis and prompt treatment of small HCC are crucial to improve the prognosis and quality of life of patients. In China, hepatitis B virus infection is the main cause. HCC with a single tumor nodule of ≤ 3 cm in diameter, or HCC with a number of nodules, in which each nodule is ≤ 2 cm in diameter, with a total diameter of ≤ 3 cm, is considered as small HCC. The MRI liver-specific contrast agent can detect small HCC at the early stage. This has important clinical implications for improving the survival rate of patients. Main body Gd-EOB-DTPA-enhanced MRI can significantly improve the sensitivity and specificity of the detection of HBV-related small hepatocellular carcinoma, providing an important basis for the clinical selection of appropriate personalized treatment. Gd-EOB-DTPA-enhanced MRI can reflect the degree of HCC differentiation, and the evaluation of HCC on Gd-EOB-DTPA-enhanced MRI would be helpful for the selection of the treatment and prognosis of HCC patients. The present study reviews the progress of the application of Gd-EOB-DTPA in the early diagnosis of small HCC, its clinical treatment, the prediction of the degree of differentiation, and the assessment of recurrence and prognosis of HCC, including the pharmacoeconomics and application limitations of Gd-EOB-DTPA. The value of the application of HCC with the Gd-EOB-DTPA was summarized to provide information for improving the quality of life and prolonging the survival of patients. Conclusion Gd-EOB-DTPA-enhanced MRI has the diagnostic capability for small HCC with a diameter of ≤ 2 cm. This will have a broader application prospect in the early diagnosis of small liver cancer with a diameter of ≤ 1 cm in the future. The relationship between GD-EOB-DTPA-MRI and the degree of HCC differentiation has a large research space, and Gd-EOB-DTPA is expected to become a potential tool for the preoperative prediction and postoperative evaluation of HCC, which would be beneficial for more appropriate treatments for HCC patients.

There is currently enormous and growing demand for flexible electronics for personalized mobile equipment, human-machine interface units, wearable medical-healthcare systems, and bionic intelligent robots. Cellulose is a well-known natural biopolymer that has multiple advantages including low cost, renewability, easy processability, and biodegradability, as well as appealing mechanical performance, dielectricity, piezoelectricity, and convertibility. Because of its multiple merits, cellulose is frequently used as a substrate, binder, dielectric layer, gel electrolyte, and derived carbon material for flexible electronic devices. Leveraging the advantages of cellulose to design advanced functional materials will have a significant impact on portable intelligent electronics. Herein, the unique molecular structure and nanostructures (nanocrystals, nanofibers, nanosheets, etc.) of cellulose are briefly introduced, the structure-property-application relationships of cellulosic materials summarized, and the processing technologies for fabricating cellulose-based flexible electronics considered. The focus then turns to the recent advances of cellulose-based functional materials toward emerging intelligent electronic devices including flexible sensors, optoelectronic devices, field-effect transistors, nanogenerators, electrochemical energy storage devices, biomimetic electronic skins, and biological detection devices. Finally, an outlook of the potential challenges and future prospects for developing cellulose-based wearable devices and bioelectronic systems is presented.

With the increasing complexity of power systems and the widespread penetration of renewable energy sources (RES), real-time situational awareness for power systems is of great significance for operation scheduling. With the impact of RES on power system operation considered, a situational awareness key performance index (KPI) system for power systems with high proportion of RES is proposed in this work, which consists of reserve capacity abundance, ramp resource abundance, center of inertia (COI) frequency deviation, interface power flow margin, synthesized voltage stability, and angle stability margin. Then, the KPIs are synthesized and visualized by decision tree method and radar chart method, respectively, for monitoring the operation states (i.e., normal, alert, and emergency states) of power systems with high proportion of RES. Numerical simulations are conducted in the revised New England 16-machine 68-bus power system and the actual CEPRI-RE power system in the northwest region of China with high proportion of RES. The results show that the proposed KPI-based situational awareness method is able to monitor the real-time state of power systems with high proportion of RES accurately, and can assist power dispatchers to make effective decisions.

Clinically significant portal hypertension (CSPH) is associated with an incremental risk of esophageal varices and overt clinical decompensations. However, hepatic venous pressure gradient (HVPG) measurement, the gold standard for defining CSPH (HVPG≥10 mm Hg) is invasive and therefore not suitable for routine clinical practice. This study aims to develop and validate a radiomics-based model as a noninvasive method for accurate detection of CSPH in cirrhosis. The prospective multicenter diagnostic trial (CHESS1701, ClinicalTrials.gov identifier: NCT03138915 ) involved 385 patients with cirrhosis from five liver centers in China between August 2016 and September 2017. Patients who had both HVPG measurement and contrast-enhanced CT within 14 days prior to the catheterization were collected. The noninvasive radiomics model, termed rHVPG for CSPH was developed based on CT images in a training cohort consisted of 222 consecutive patients and the diagnostic performance was prospectively assessed in 163 consecutive patients in four external validation cohorts. rHVPG showed a good performance in detection of CSPH with a C-index of 0·849 (95%CI: 0·786–0·911). Application of rHVPG in four external prospective validation cohorts still gave excellent performance with the C-index of 0·889 (95%CI: 0·752–1·000, 0·800 (95%CI: 0·614–0·986), 0·917 (95%CI: 0·772–1·000), and 0·827 (95%CI: 0·618–1·000), respectively. Intraclass correlation coefficients for inter- and intra-observer agreement were 0·92–0·99 and 0·97–0·99, respectively. A radiomics signature was developed and prospectively validated as an accurate method for noninvasive detection of CSPH in cirrhosis. The tool of rHVPG assessment can facilitate the identification of CSPH rapidly when invasive transjugular procedure is not available.

Sigmoid sinus wall anomalies (SSWA) are a common pathophysiology of pulsatile tinnitus (PT) and usually treated by sigmoid sinus wall dehiscence (SSWD) resurfacing surgery. However, symptoms of tinnitus remain unrelieved after surgery in some patients with PT, and even new tinnitus appears. The cause of the difference in therapeutic effects is unclear. In this study, eight patient-specific SSWA geometric models were reconstructed on the basis of computed tomography angiography, including four cases of postoperative rehabilitation (group 1, 1–4 cases) and four cases of non-rehabilitation (group 2, 5–8 cases). Transient-state computational fluid dynamics (CFD) was performed to clarify the SS blood flow pattern and hemodynamic states. The wall pressure distribution on SSWA area, pressure difference, and flow pattern in SS were calculated to evaluate the hemodynamic changes of rehabilitation and non-rehabilitation patients before and after surgery. The difference of hemodynamics between these patients was statistically analyzed. The accuracy of CFD simulation was evaluated by cross validating the numerical and particle image velocimetry experimental results. Results showed that the SSWA area in patients with PT was loaded with high pressure. No difference was found in the hemodynamic characteristics between the two groups pre- and postoperation. When the average pressure (Pavg) and time-average Pavg (TAPavg) on the SSWA area were studied, the TAPavg difference pre- and postoperation between the two groups was found significant (p = 0.0021). The TAPavg difference had a negative change in postoperative rehabilitation patients (case 1, -44.49 Pa vs. case 2, -15.85 Pa vs. case 3, -25.88 Pa vs. case 4, -16.58 Pa). The postoperative TAPavg of non-rehabilitation patients was higher than the preoperative one (case 5, 24.70 Pa vs. case 6, 28.56 Pa vs. case 7, 5.81 Pa vs. case 8, 13.04 Pa). The velocity streamlines in the SS with rehabilitation became smoother and more regular than that without rehabilitation. By contrast, the velocity streamlines in SS without rehabilitation showed increased twisting and curling. No difference was found in time-average volume-averaged vorticity (TAVavgV) between the two groups. Therefore, the high pressure of the vessel wall on SSWA area was one of the causes of PT. The variation of SSWA wall pressure difference before and after PT was the cause of the difference in therapeutic effects after SSWD resurfacing surgery. In patients with SSWA, disordered blood flow in SS was another cause of PT. SSWD repair may relieve tinnitus to some extent, but blood flow disorders may still arise.

Seeking nodes whose removal can effectively dismantle networks is closely related to the robustness of networks under targeted attacks or malicious software. Modern systems, such as critical infrastructure networks and cyber-physical systems, however, become increasingly dependent on others, which can be encapsulated into the framework of interdependent networks. In this brief, we focus on the optimal dismantling of interdependent networks, and propose a novel algorithm, named inverse explosive percolation (IEP), to find the optimal removal nodes. The IEP proceeds by first identifying the nodes which make the least contribution to the giant mutual connected cluster of interdependent networks. When nodes have the same contribution, the one with the smallest overlapping degree is selected. We apply the IEP algorithm to interdependent networks composed of real-world power grids, Internet networks and artificial networks, and find IEP performs much better than the existing methods in various kinds of networks. Based on its efficiency, IEP can be applied to large-scale systems.

To improve the ability of water to wet coal in the process of coal seam water injection, different ionic liquids and surfactant compound solutions were added to water. Analyses of surface tension, reverse spontaneous imbibition, and contact angle were used to optimize the different compound solutions. The results showed that when the mixture ratio of 1-butyl-3-methylimidazolium tetrafluoroborate ([BMIm][BF4]) and sodium dodecyl sulfate (SDS) was 3:1, and the concentration of the composite solution reached 0.1 wt%, the liquid surface tension reached a minimum of 28.946 mN/m, and the contact angle between the solution and the coal sample was 41.3°, which was lower than that of any concentrations of the other composite solutions. This solution could effectively wet the coal. In addition, infrared spectroscopy showed that different concentrations of the [BMIm][BF4] and SDS composite solution had different effects on the content of functional groups in coal. Further, based on the relationship between the content of the functional groups and contact angle, the relationship between the content of ring hydroxyl and OH⋯O in the hydroxyl structure and the size of the contact angle was found to be the most significant. Based on the above findings, a new type of composite wetting agent was prepared, and the wetting mechanism between coal and the wetting agent was clarified from the perspective of functional groups. The research results have important guidance and practical significance for improving the effect of increasing moisture and reducing dust in the process of coal seam water injection.

At present, community detection methods are mostly focused on the investigation at unweighted networks. However, real-world networks are always complex, and unweighted networks are not sufficient to reflect the connections among real-world objects. Hence, this paper proposes a community detection algorithm based on a deep sparse autoencoder. First, the second-order neighbors of the nodes are identified, and we can obtain the path weight matrix for the second-order neighbors of the node. We combine the path weight matrix with the weighted adjacent paths of the node to obtain the similarity matrix, which can not only reflect the similarity relationships among connected nodes in the network topology but also the similarity relationships among nodes and second-order neighbors. Then, based on the unsupervised deep learning method, the feature matrix which has a stronger ability to express the features of the network can be obtained by constructing a deep sparse autoencoder. Finally, the K-means algorithm is adopted to cluster the low-dimensional feature matrix and obtain the community structure. The experimental results indicate that compared with 4 typical community detection algorithms, the algorithm proposed here can more accurately identify community structures. Additionally, the results of parameter experiments show that compared with the community structure found by the K-means algorithm, which directly uses the high-dimensional adjacency matrix, the community structure detected by the WCD algorithm in this paper is more accurate.

Our study aims at developing an effective quality monitoring system in small scale resistance spot welding of titanium alloy. The measured electrical signals were interpreted in combination with the nugget development. Features were extracted from the dynamic resistance and electrode voltage curve. A higher welding current generally indicated a lower overall dynamic resistance level. A larger electrode voltage peak and higher change rate of electrode voltage could be detected under a smaller electrode force or higher welding current condition. Variation of the extracted features and weld quality was found more sensitive to the change of welding current than electrode force. Different neural network model were proposed for weld quality prediction. The back propagation neural network was more proper in failure load estimation. The probabilistic neural network model was more appropriate to be applied in quality level classification. A real-time and on-line weld quality monitoring system may be developed by taking advantages of both methods.

Approximately 50% hepatocellular carcinoma patients meeting the Milan criteria utilized to develop an improved prognostic model for predicting the recurrence in these patients. Using univariate and multivariate analysis, cytokeratin-19 and glypican-3 expression patterns, tumor number and histological grading from eight putative prognostic factors comprised the risk factor scoring model to predict the tumor recurrence. In the training cohort, the area under roc curve (AUC) value of the model was 0.715 [95% confidence interval (CI) = 0.645−0.786, P

Uncontrolled hemorrhage after trauma to the liver can lead to death. The present study compared the effects of non-focused microbubble-enhanced ultrasound and high-intensity focused ultrasound on hepatic hemostasis in the injured liver. Blood perfusion level, serum liver enzyme levels and the aspartate transaminase/alanine transaminase ratio differed between the two types of treatment (all p values < 0.05). Hepatic cells in the microbubble-enhanced ultrasound group exhibited edema and compressed the hepatic sinus and blood vessels in the portal area. Coagulation and necrosis, inflammatory cell infiltration, and fibrous tissue encapsulation were observed in the high-intensity focused ultrasound group at later stages. The groups also differed in degree of ultrastructural damage and recovery time. Thus, microbubble-enhanced ultrasound has less of an impact on blood reperfusion and surrounding normal tissue than high-intensity focused ultrasound and is a better choice for the treatment of liver trauma.

Our study aims at developing an efficient quality monitoring system in small scale resistance spot welding based on dynamic resistance. The dynamic resistance variation was related to weld nugget formation process. An initial resistance peak caused by asperity heating was detected. The second peak in dynamic resistance and single peak in dynamic voltage could be attributed to bulk material heating. An obvious interrelationship could be found between end resistance and weld quality. The features extracted from dynamic resistance curve were mainly influenced by welding current. The overall resistance level was dropped as welding current enlarged. The multiple linear regression analysis and back propagation neural network were then used to estimate the weld quality in combination with extracted features. Result of the regression analysis for quality prediction was basically satisfactory. The proposed neural network model showed a better performance than regression analysis regarding the maximum estimation error and root mean square error. Accuracy of the neural network based quality estimation could be further improved combining quality level classification strategy. Combination of the dynamic resistance measurement with neural network model was supposed effective to achieve the quality monitoring purpose in small scale resistance spot welding.

Alpha-helical transmembrane proteins ( $\alpha \text{TMPs}$ ) belong to an important category of integral membranes. Their structures are highly valuable in relevant research, but costly to solve experimentally. Sequence-based topology prediction provides a practical computational approach to characterize the structure features. Although much progress had been made in the past decade, there is significant room for improvement in predicting the topology structure. Deep learning brings a great opportunity for its capability of mining new features from data. In this work, we propose a novel $\alpha \text{TMP}$ topology prediction method DMCTOP using a Deep Multi-Scale Convolutional Neural Network (DMCNN), which composes of two deep convolutional blocks to extract local and global contextual features. Consequently, the inputs of DMCTOP is simplified to a protein sequence feature and an evolutionary conservation feature. DMCTOP can efficiently and accurately identify all topological types and the N-terminal orientation for an $\alpha \text{TMP}$ sequence. In the testing experiments, the prediction accuracy was calculated according to the whole sequence, the transmembrane segments and the traditional criterion. Our state-of-the-art method achieved the highest prediction accuracy compared to all the previous methods. The standalone tool is available at https://github.com/NENUBioCompute/DMCTOP.

Hepatocellular carcinomas(HCC) consisted of heterogeneous subtypes with different recurrence probabilities after liver transplantation(LT). Our study aimed to develop an improved model for predicting the recurrence of solitary HCC after LT. In this retrospective study, 151 solitary HCC patients who received orthotopic LT over a period of 10 consecutive years were included. All recipients received graft from deceased donors. The first eligible 50 patients were used as validation cohort and others were utilized to construct the model. A two-tailed P