Fires onboard ships cause danger to property, persons, and the surrounding environment. Within a ship, there is a large amount of liquid fuel, lubricating material, electrical equipment, and other auxiliary machineries. During navigation, an occurrence of a fire on a ship can create a high-risk situation that may cause harm or even deaths to passengers and considerable damage to the ship, its structure, and equipment. In this research, an integrated model is constructed to assess the fire risk onboard ships. Interpretive structural modelling (ISM), fuzzy analytical network process (FANP), and the evidential reasoning (ER) approach are incorporated in the model. This study lists representative factors and examined the interrelationships among them. Experts from the marine field are requested to provide their expertise and assess the importance of the factors and various features of the risk assessment. An illustrative example of fire risk analysis onboard a ship is shown using the proposed approach.

In this study, we have proposed a model to identify pulmonary diseases such as COVID-19, pneumonia, and tuberculosis from X-ray images. Identifying and differentiating among these diseases is already a hard task for doctors. Recent findings obtained using radiology imaging techniques suggest that X-ray images contain salient information about these diseases. Application of deep convolutional neural networks coupled with radiological imaging can be beneficial for the accurate diagnosis of these diseases, and it can also be assistive to overcome the problem of shortage of healthcare experts in remote villages. We have implemented eight deep convolutional neural network (Deep CNN) models: AlexNet, VGG16, VGG19, DenseNet201, Xception, ResNet50, Sequential, and InceptionV3. Comparative analysis of the implemented models suggests that deep learning with X-ray imaging extracts significant biomarkers related to these diseases, while the best accuracy and least loss is obtained while training our model with VGG16. The training accuracy, precision, and false-positive obtained for VGG16 is 99.66%, 97.56%, and 2.24%, respectively.

Retrieval of information is the presumption of a real-world user declaration based on the concept of storing, representing, and searching information. It also uses the database to easily access the repository of information. Developing a system that can come up with the ambiguous user needs is a complex task, as each of the incoming queries from the user to the system is unique. The expectations from the system are high in order to fulfill the information needs of the user. The fluidity of the human brain is extremely high, because each person has different cognitive abilities, values, thinking, interpretations, situations, intent, instincts, and domains. Such various characteristics lead to a user's query about or need for data; hence, it could be said that individual users have different apprehensions of query. The key requirement is to consider the users' query to provide the individual with effective and appropriate results according to the knowledge required. This paper implemented BB2 hybrid with a fuzzy logic inference. The purpose of the system is to map the IR model using the fuzzy approach with the cognitive abilities of the user. The similarity between the file, depending on the graininess of the user, was calculated using fuzzy logic to evaluate the query results.

Cloud computing data centers are generally driven by services executing over cost-efficient commodity hardware. The hardware is easily scaled horizontally by the addition of more resources. However, the use of commodity hardware also results in a high failure rate of physical machines (PMs) in the cloud environment. This, in turn, causes the failure of virtual machines (VMs) provisioned on the failed physical machine. The paper proposes a connected dominating set (CDS)-based method for the construction of a virtual backbone over the data center network topology. This backbone is then utilized to develop a replication-based fault tolerance scheme to improve the reliability of VM-based services. The CDS construction approach models the network topology as a graph using the PMs as nodes. It then proposes two rules based on temperature and node degree of PMs to obtain a CDS for the topology graph. The PMs at lower CPU temperature and possessing higher node degree are selected to constitute the CDS, and, subsequently, VM replicas are placed on these CDS nodes. The performance of the proposed replication scheme is evaluated using two metric: the total bandwidth and the energy consumed during replica placement and retrieval. It is compared with a random replica placement method and an existing method of CDS construction. The simulation outcomes demonstrate that the proposed method provides better reliability and consumes fewer resources than other methods.

The system distributed denial of service (DDoS) contains high covert attack characteristics and requires real-time defense. In order to solve such two problems for system DDoS, this paper proposes a novel DDoS attack real-time defense mechanism based on deep Q-learning network (DQN). This mechanism regards the terminal adaptive control system as the protection object, periodically extracts the network attack characteristic parameters, and takes such parameters as the input parameters of the deep Q-learning network. Our defense measures are based on dynamic service resource allocation, which dynamically adjusts the service resource according to the current operating state of the system. The current operating state will ensure the response rate of normal service requests. Finally, the attack and defense processes are modeled and simulated using colored Petri network (CPN) combined with DQN. Experimental results show that the proposed mechanism has real-time and high sensitivity defense for the response to DDoS attacks. The proposed mechanism significantly improves the automation degree of system defense. By using such a mechanism in the real-time defense of DDoS attacks, the system will be safer than the state-of-the-art mechanisms.

To improve the performance of the channel estimation in the cascaded scenario, we propose a sparsity-aware LMS algorithm by using a new cost function with a variable step size and l_p norm constraint. The step size is updated according to the square root of the estimated error each iteration, which allows the adaptive filter to track the changes in the channel to produce a small steady-state error. By exploiting the sparsity of the channels, the proposed algorithm integrates the l_p norm penalty, which imposes a zero attraction of the sparse channel coefficients. Next, the convergence performance of the proposed algorithm is analyzed, and the stability condition is derived. Our theoretical analysis shows that the proposed algorithm effectively decreases the amount of mis-adjustment and improves the channel estimation accuracy. Finally, the simulation results demonstrate that the proposed algorithm can converge quickly, while its performance outperforms that of the conventional LMS-based identification algorithm.

To address the instruction order effects of dynamic decision-making for landing signal officers (LSO) of carrier-based aircraft in a comparative semantic context, in this paper, we propose a prediction approach of LSO guidance decision-making based on quantum interference. By constructing the quantum decision-making model of incompatible events, the restraint of law of total probability is shaken. Comparing the decision deviation of alternatives with non-comparative and comparative contexts, we analyze the interference nature of order effects. Then, we represent the similarity of events based on projective similarity index in Hilbert space and design a q-test prediction that satisfies the order effects of quantum decision-making. The experimental results show that the modified approach is accurate in predicting for assimilation effects and contrast effects. The theoretical basis for improving the accuracy of LSO decision-making is confirmed in a less projective similarity comparative semantic context. As such, the obtained results have certain reference value for personnel allocation, work periods, and landing safety of carrier-based aircraft.

Spatial verification for duplicate image retrieval is often time-consuming and not sensitive to similar images. To address this problem, we propose a fast and precise spatial verification strategy for duplicate image retrieval. The motivation of this strategy is to use angle and scale information to filter similar images. This is because the matched descriptors of similar images are not transformed according to consistent angles and log-scales. The angle differences and log-scale differences of matched descriptors can be projected to points in two-dimensional space. Intuitively, the two-dimensional point distribution of non-duplicate images is relatively discrete, and the two-dimensional point distribution of duplicate images is relatively concentrated. Therefore, this paper utilizes the inverse cloud algorithm to calculate the discrete degree of the two-dimensional point distribution to exclude the non-duplicate images that have large fluctuation distributions. Then, the new voting algorithm can be used to re-rank the images to improve the retrieval accuracy. The experimental results showed that, compared with traditional algorithms, the new strategy was able to effectively improve retrieval accuracy without adding extra storage overhead and computational overhead.

Surface electromyography signal (sEMG) plays an important role in gesture recognition and prosthetic control. Aiming at the problems of complex combination of RNN parameters, setting difficulty, and structure dependence of model quality, an EMG pattern recognition method based on particle swarm optimization recurrent neural network (PSO-RNN) is proposed. This method uses the characteristics of particle swarm optimization (PSO), such as high global search efficiency, fast convergence speed, and wide optimization range, and automatically finds the optimal structure of RNN through continuous iterative updating. On the Ninapro EMG database, the classification of 12 types of EMG actions by the PSO-RNN algorithm is tested, and the results are compared with four algorithms applied in the same data set. The results show that the proposed PSO-RNN algorithm model achieves a high accuracy of 94.1667%, and it has certain effectiveness and practicability.

The effective evacuation route decision for metro stations is critical for maintaining operation safety and tackling emergency issues. The paper aims to study the intelligent evacuation route decision problem based on social media. The paradigm intelligent evacuation route decision of metro stations is put forward, including the intelligent preparedness, smart technical platform, professional personnel, and intelligent evacuation route. Then, the intelligent algorithms of evacuation route decisions for metro stations are selected, which includes artificial neural networks, fuzzy systems, and the genetic algorithm. Beside it, the comparative analysis of the intelligent algorithms are presented, which contributes to select the suit method to make an evacuation route decision of the metro station. Meanwhile, an application of Optics Valley Square Station is introduced. The results show that the evacuation routes of Optics Valley Square Station are not scientific, and advice to improve the evacuation routes are given.

To address the shortcomings of traditional Simple Network Management Protocol (SNMP)-based network management systems, including their inflexible data collection mechanism, poor scalability and being difficult to develop and maintain, this paper proposes designing and implementing a universal and portable SNMP data collection and processing framework in which the management information base (MIB) to eXtensible Markup language (XML) module is regarded as the crosscutting concern through which automatic data filtering and data mapping are realized, and a non-relational database (Redis) is used for storing data processed through XML mapping. We used AdventNet_Simulation_Toolkit to simulate a multi-agent environment in which various modules of the above framework (from data acquisition to data storage) were constructed, and the functional and performance testing of the entire framework was conducted. According to the test results, this study presents a new route for the development of web-based SNMP management systems.

In view of the low node accuracy in wireless sensor node positioning, this paper proposes a node positioning algorithm based on the Improved Grey Wolf Optimization and Least Squares Support Vector Machine (IGWO-LSSVM). First, a wireless sensor positioning model in two-dimensional space is established. Then, the least squares support vector machine is used to model and locate unknown nodes. Finally, the least square vector machine parameters are optimized based on chaos mapping, adaptive factors, and the golden sine gray wolf algorithm to obtain node positioning. Simulation experiments show that compared with other algorithms, this algorithm has a better effect on the accuracy of node positioning.

With the advent of the era of big data, a large number of information data are generated at every moment. When people are faced with a large amount of information, they are often overwhelmed in the ocean of data and unable to make a quick and accurate decision. In this context, a personalized cloud computing recommendation system emerges as the times require, which can solve the problem of information overload. The development of the catering industry and the mutual penetration of the catering culture meet people's requirements for the diversity of food. However, with the substantial growth of the number of dishes and the emergence of new products, diners will also encounter difficulties in choosing a large number of dishes, especially sous-vide cooking dishes. In this paper, we choose the recommendation mechanism based on association rules to study the association information between diners and dishes. The Apriori algorithm is selected to mine the potential association information between dishes from a large number of historical orders, and the frequent pattern set of associated dishes is obtained as the association rule set. According to the characteristics and advantages of Sous vide cooking dishes, and then according to the dishes that the diner has ordered, through the cloud computing, the mining association rules recommend dishes that meet the diner's taste.

This paper investigates the influence of resonator radius distributed nonuniformly in the circumferential direction on the accuracy of hemispherical resonator gyro under the parametric excitation and the positional excitation. According to the dynamic equation of ring-shaped hemispherical resonator, the error expression of gyro precession angular rates is derived. Then, the influence of radius nonuniformity on processing is studied through simulations under the parametric excitation. Further, the effect of radius nonuniformity on solving error of input angular rate is investigated through simulations under the open-loop and force-rebalance modes. In the end, tolerance towards radius nonuniformity is determined from the perspective of frequency splitting. Research results indicate that the fourth harmonic of radius' nonuniformity affects the accuracy of gyro.

During the past few years, Internet of things (IoT) has become an unprecedented buzzword in most areas of business and industry. It has also been one of the most revolutionary technologies in recent times, catalyzing a paradigm shift in traditional healthcare methods. This paper focuses on the monitoring and tracking of sleep health and designs a smart mattress system, which mainly includes information collection of mattress sensors, data transmission of communication networks, and distributed data management center of the message queuing telemetry transport (MQTT) protocol. Through the coordination of these parts, the sleep quality monitoring of a specific group of people can be achieved. The diagnosis of corresponding diseases, the tracking of health conditions, and the warning of special situations can be performed via the bed system.

The energy demands for heating have been increasing rapidly in Northern China. To improve the energy efficiency of residential buildings in Changchun, a major city in Northeast China, a residential building model is constructed in the DesignBuilder software and used to assess the effects of geometric parameters including floor height and window-to-wall ratio on the energy performance of residential buildings. The simulation is conducted with the parameters of both traditional plan and alternative plans to calculate the energy consumption of each plan and evaluate the parameters. Based on the simulation results, the most optimal geometric parameters of the residential building model are proposed. The floor height of 2.8 m gives the highest energy-saving efficiency in the tested range. The optimal window-to-wall ratios of the north wall, south wall, east wall, and west wall of the building model are determined to be 0.25, 0.45, 0.30, and 0.30, respectively. Our study identifies an efficient passive design plan upon the geometric parameters of residential buildings in severely cold areas and provides a theoretical reference for decision-making in the early design stage.

Prefabricated buildings are an important part in realizing the industrialization of housing. As our country vigorously promotes prefabricated building, it is particularly important to judge the risk of prefabricated buildings entering the market. According to the grey assessment model of the strategic cost risk of prefabricated buildings, the risk of the prefabricated building market will be analyzed holistically. It is possible to accurately determine whether the risk management of prefabricated buildings in different regions and different enterprise development projects is in a reasonable controllable range when entering the market. In the various aspects of the strategic cost risk management of prefabricated buildings, reasonable improvement measures are proposed for the upstream value chain and the downstream value chain. For the further promotion of prefabricated buildings, the development planning of residential industrialization and the related departments to formulate and improve relevant policies and industry standards have very important guiding significance. When prefabricated buildings enters the market, they can reduce the subjectivity of decision makers and provide a decision-making basis for the enterprise to develop a development strategy related to assembly.

Engine misfire fault diagnosis model based on probabilistic neural network (PNN) is established, in which the vibration acceleration signal of engine cylinder block surface is used as diagnosis parameter. The time domain signal of vibration acceleration is directly used as network input by virtue of the advantage of PNN in rapid processing of large and complex data. Simplify the diagnostic process. In the PNN, the dispersion constant "spread" value determines the sensitivity and stability of the network model. Aiming at this problem, the particle swarm intelligent algorithm, which can achieve efficient parallel search, is used to find the optimal "spread" value to ensure the best diagnosis effect. The experimental results show that the accuracy can reach 100%, and the fault tolerance rate is high in the non-ideal state, which shows that this method is feasible and convenient.