Abstract:There are more and more complex control objects with large inertia, time-varying and time-delaying in modern industrial production. The traditional control algorithms fail to meet their higher and higher control requirements. Inspired by the physiological two-way network regulation mechanism of blood glucose concentration in human body, an intelligent coordinated controller is proposed. The intelligent coordinated controller includes four parts: a main control unit（MCU）, an auxiliary control unit（ACU）, a monitoring adaptation unit（MAU） and a cooperative control unit（CCU）. Under the supervisory control of the MAU and the coordinated control of the CCU, the MCU and the ACU work synergistically to ensure that the intelligent coordinated control system reaches the target value of the control system with a relatively smaller rise time and adjustment time on the premise of no overshoot. In order to verify the control performance of the intelligent coordinated controller, an industrial ethanol fermentation bioreactor is selected as the controlled object, and the temperature of the bioreactor is controlled and simulated. The experimental results show that the intelligent coordinated controller has better dynamic performance, steady-state performance and anti-interference ability when compared with the common BP neural network controller and fuzzy controller.

Abstract:Aiming at the random-frequency-hopping synthetic wideband radar system as well as the application background of hypersonic vehicle platform, this paper analyzed the large scale Doppler broadening effect and random/non-linear range-Doppler coupling effect, as well as their influence on target range profile. Meanwhile，a method of real-time recursive and mematic imaging algorithm based on Doppler pre-processing was proposed. By this method, the multi-speed-channel range profiles were obtained by using the echo data of current frame and past frames, and Doppler-beam-sharpening was advanced to be performed at the speed channel of target and its adjacent channels to suppress the above special effects caused by high dynamic motion of platform and random frequency hopping in bandwidth synthesis. The recursion was used so that the range profiles of each frame can be obtained from those of the previous frame through adding a few computations. The memory factor was adopted to further reduce the storage and computation costs. Theoretical analysis and simulation results show that the proposed imaging method can effectively improve the imaging quality, and can more easily meet the high-data-rate real-time imaging requirements in the hypersonic platform-borne application.

Abstract:Diabetic retinopathy （DR） is one of the major causes of blindness, and the appearance of retinal neovascularization （RN） is an important sign of DR deterioration. In order to detect RN more accurately, a method based on color fundus photograph for retinal neovascularization detection is proposed. First, an improved U-shaped convolutional neural network is used to segment the blood vessels. Then, a sliding window is used to extract the morphological characteristics of blood vessels in the specific area. A support vector machine （SVM） is used to classify the blood vessels into normal vessels and retinal neovascularization in the window. The experiments use color fundus photographs with retinal neovascularization from the MESSIDOR dataset and the Kaggle dataset for training and testing. The result shows that the accuracy of this method for the RN detection is 95.96%; This method has potential application prospects in the computer-aided diagnosis of diabetic retinopathy.

Abstract:Accurate modeling of software reliability and effective measurement and prediction of reliability trends are critical to software development. The closer to the real process of software testing the model gets, the more specific factors should be considered and integrated into the imperfect debugging model, and the software reliability growth model （SRGM） with more accurate factors should be built. Considering the intrinsic relationships among the three sub-processes including fault detection, repair and introduction, a unified and flexible imperfect debugging framework model TCM-ID is established to study the relationships among cumulative detection, repair and introduced faults. The overall efficiency of the software test is measured from the perspective of fault detection rate, fault repair rate and fault introduction rate. Further, from the perspective of test coverage, a reliability model TCM-ID （Testing Coverage Software Reliability Model under Imperfect Debugging） is established to discuss its perturbation effect on the model and to evaluate the performance of the model. Finally, the validity and rationality of the proposed model are verified in real application scenarios. The model has better fitting and prediction performance, and it is better than other models overall. The model proposed in this paper is of great significance for selecting the appropriate SRGM for the test coverage under the imperfect debugging conditions and improving the test efficiency and software reliability.

Abstract:Building an accurate energy-consumption model of servers can assist resource providers in predicting and optimizing energy consumption of data center. To address the problem of low accuracy of energy consumption model caused by the failure to consider "load characteristics" of servers in data center, a new energy consumption model and quantitative calculation method are proposed in this paper. The main ideas are summarized as follows: Firstly, we divide the tasks into three classes: CPU intensive task, transactional web task, and I/O intensive task. Then, energy consumption contributions of all components in a server are analyzed. After that, the dominant component parameters of server energy consumption are chosen by using the Principal Component Analysis （PCA）, to build a power model through the multiple linear regression method and non-linear regression method. Experimental results show that the prediction accuracy of the proposed energy consumption model can achieve more than 95%. Compared with other energy consumption models, the accuracy can be improved by around 3%.

Abstract:Named entity is the main carrier of relevant medical knowledge in Electronic Medical Records （EMRs），so clinical named entity recognition（CNER） has become one of the basic and crucial tasks of clinical text analysis and processing. Due to the particularity of medical text structure and Chinese language，the recognition of clinical named entities for Chinese EMRs still faces great challenges. In this paper, a Chinese clinical named entity recognition method based on multi-head self-attention neural network is proposed . In this method, a character-level feature representation method combined with a domain dictionary is presented. Moreover, based on the BiLSTM-CRF model, a multi-head self-attention mechanism is incorporated to accurately capture the multiple features from different aspects, such as dependency weights between characters and contextual semantic relationships, thereby effectively improving the ability of Chinese clinical named entity recognition. Experimental results demonstrate that the proposed method outperforms other existing methods and has the best recognition performance.

Abstract:In order to effectively solve the problem of grey state in the process of oil and gas pipeline safety management, the evolution process of oil and gas pipeline safety risk is analyzed based on the risk source theory, and a safety management model for oil and gas pipeline system is constructed. The model is composed of three workflows in series: security risk identification, security risk early warning and security risk control. Based on the factual attribute of hazard source, the reasonable or approximate reasonable result of risk identification is deduced by using the uncertainty transfer algorithm and the extended generation rule in combination with the risk database. By using the extension theory, the warning level is divided into four levels: no alarm （N1）, low alarm （N2）, medium warning （N3） and high warning （N4）, and qualitative analysis and calculation of oil and gas storage and transportation risk warning are performed. Using Java programming language and Oracle database technology, a safety management platform for oil and gas pipeline system is designed, including safety risk identification, safety risk warning, safety information management and other functional modules. The research findings show that the introduction of uncertainty theory can effectively solve the subjective and fuzzy problems of risk identification in the process of oil and gas storage and transportation. According to the calculation of the extension risk warning model, the characteristic value of the level variable of the tank is 2.0404, which belongs to the low warning range. The obtained results are in good agreement with the actual work area. The method of information management makes up for the defects of empirical management and provides good technical support for the integrity of risk data and safety emergency decision.

Abstract:This paper proposes a joint relay selection and power allocation scheme for the cognitive radio network based on energy harvesting. Under the condition that the primary user performance is guaranteed, this scheme can significantly improve the energy efficiency of system. This paper considers that the transmitters of primary user and secondary user both conduct energy harvesting. In addition, we consider the imperfect channel status information. First，the secondary user is selected as a relay to assist the primary user to transmit data, while the secondary user collects the energy of the radio frequency. Then， we present the relay selection and power allocation tactics to maximize energy efficiency. Since the convex optimization problem is still a fractional programming, we apply fractional deformation and Lagrange method to solve the problem of optimizing the maximum throughput of the primary network. Based on the proposed scheme, an off-line optimization problem is proposed to optimize the throughput of the primary user, and we apply generalized the Bender decomposition to solve this problem. Numerical results show that the proposed scheme has an obvious advantage in energy efficiency.

Abstract:A surge in the workload of the hospital for testing potential patients and an increased risk of cross-infection among patients due to emergent epidemic, poses a critical social public safety problem. Through the methods including literature review, epidemiological theory, expert consultation，subjective weighting （involving experts from the Departments of Infection Control, Respiratory Medicine, Critical Care, and the Center for Disease Control）, human-computer interactive design，user test，usability evaluation and etc，the content of online affected risk self-check application, available options, risk weighting and threshold, and human-computer interactions of high usability are investigate. An early prevention and control strategy is proposed for suspected cases screening and reliable suggestions to self-monitor at home, self-isolate, or seek immediate medical attention. The human-computer interaction application and screening model for suspected symptom risk assessment are designed and promoted online. The results indicate that in the early phase of emergent epidemic, the online emergent epidemic affected risk self-check application can provide a simple and feasible scientific basis for clinical judgement, alleviate the social panic, resolve the shortage of manual screening, and reduce the risk of cross-infection among patients by providing a straightforward, easy-to-understand scientific protocol. It provides references for future new epidemics prevention and control through online classification and offline triage of patients of suspected symptoms.

Abstract:For marine vessel target detection in SAR images，deep learning technology alone usually has difficulty in satisfying the detection requirements in accuracy and timeliness. Vessel targets in SAR image are of small sizes and resolutions，which are easily interfered by noise and spot interruption. It is challenging to extract subtle features and eliminate background interference under complex conditions. To overcome the above problems，we propose an end-to-end new detection model based on YOLOv3 framework. The residual module structure is used to avoid network degradation. Combined with deep and shallow feature detection of different target sizes，we extract network parameters for basic features to avoid training from scratch. At the same time，according to the characteristics of small vessel targets in SAR image，the neural network structure is further optimized to achieve fast target detection and categorization in wide-area SAR images，and the detection model is compressed and light-weighted. We construct and utilize a SAR image dataset with different vessel targets for target detection and classification test. The experimental results show that the proposed detection method shows significant anti-jamming ability and detection performance in complex scenes.

Abstract:In order to address the problem that traditional medical image error prediction algorithm can not select image features well，there are some problems such as low fitting degree of image error prediction value，low actual value and long prediction time，a medical image error prediction algorithm based on convolution neural network and feature selection was proposed. Firstly，five integrated rules were selected to construct adaptive multi-classifiers to classify medical image regions. Secondly，the training convolution neural network was used to extract different types of medical image area features by using the training neural network. Then，multiple evaluation criteria were combined to generate special features. The optimal feature subset was searched to complete the feature selection of suspicious region image. Finally，the multiple linear regression matrix between the prediction sample and the training sample was established to realize the error prediction by taking the pixel points of the feature region as the training sample. The experimental results show that the proposed algorithm has high fitness of integration rules and good classification performance，the accuracy of region distance calculation is about 95%，the AUC value of feature selection is high，and the fitting degree and prediction time of the prediction results are better than those of the traditional algorithm.

Abstract:When lifting the wire rope connection counterweight, synchronous control has high requirements on speed, response, precision and stability, and the fixed parameters PID （FPP） is hard to meet the control requirements under wide load conditions. Firstly, a master-slave variable benchmark strategy （VBS） is designed as the bottom strategy to achieve the maximum speed and accuracy of system design. Secondly, it proposes a variable parameters PID self-adaption approach （VPPSA）. In this approach, the parameters Kp，Ki，Kd following tuning rules are introduced into deviation objective function as cost items for the fitness calculation of meta-heuristic algorithm. For the searched group（Kp，Ki，Kd），when the cost items are small, the objective value is small and the fitness is large. Therefore, the parameters can be tuned adaptively to obtain the minimum deviation value. Genetic algorithm （GA） and particle swarm optimization algorithm （PSO） are used for real-time parameters tuning, deviation optimizing and synchronization tracking，respectively, and the synchronization effects are compared with that of FPP. Simulation shows that the VPPSA can adapt to wide range of load conditions, and has better performances in terms of accuracy, response and stability than the FPP. The VPPSA based on VBS can realize the rapid and stable synchronous lifting of large inertia system.

Abstract:Motivated by the difficulty of the electromechanical product （system） performance reliability analysis and modeling, an general analysis method is developed for meta action unit performance reliability based on hybrid copula. The characteristics of performance degradation process are analyzed to establish the relationship between the product performance degradation and the realized specific functions. Using the functional motion decomposition method of "Function-Motion-Action", the electromechanical products are decomposed into a series of meta action units via linking functions with their performance. The components performance reliability analysis model is constructed that we consider it obeys the nonlinear gamma random degradation process is used to describe the degradation process of key parts in the unit. Considering the coupling relation of meta action unit internal parts and structure, a performance reliability model of the meta action unit is constructed based on the hybrid copula function. In order to more accurately evaluate the performance reliability of meta-action unit, the optimal Copula Functions having is selected to develop the performance reliability model of meta-action unit. The unknown parameter estimation of hybrid copula function is realized by improved genetic algorithm, which adds penalty term to fitness function. Combining the hybrid copula mathematical expression and the concept of a series system, the performance reliability expression of meta action unit considering the coupling dependence is obtained. The feasibility and effectiveness of the method are verified by an example.

Abstract:To test and verify the reliability of relay valve （with output pressure of 12.0 MPa and response time of 0.2 s）, and to study the influence of dynamic characteristics of the relay valve on braking performance of the full hydraulic braking system, taking a type off-road vehicles fully hydraulic braking system as the research object, a theoretical analysis model of the relay valve was established. A simulation model of the full hydraulic brake system was established by applying the software AMESim, and the influence of the spool friction, initial cover, return spring the initial amount of compression and spring stiffness on braking performance was analyzed. The accuracy of the simulation model was verified by experimental results. The comparison shows that the relay valve applied to the hydraulic braking system can meet the braking requirements（output pressure of 12.0 MPa and response time of 0.2s）. Excessive friction of the spool increases the opening pressure of the relay valve, which leads to the increase of proportional hysteresis of the relay valve and affects the reset performance of the spool. The greater the initial cover of the relay valve orifice is, the greater the friction to be overcome by opening the orifice becomes, and the longer the response time of the braking system is required. By adjusting the initial compression of the reset spring of the relay valve and the stiffness of the spring, the fine tuning of the brake pressure can be realized. The theoretical model and simulation model provide a reliable basis for further optimization of the full hydraulic braking system.

Abstract:Under turning conditions，due to the combined action of the concrete flow in the mixer drum of the mixer truck and the turning centrifugal force，the mixer truck is easy to roll over. In response to this problem, the dynamic mass center of the mixer drum and the stability of the mixer truck rollover were studied. On the basis of using EDEM software to simulate the movement of concrete during transportation, a mathematical model of the position change of the center of mass is obtained through calculation and fitting, and the relationship between the position change of the center of mass and the centrifugal force and the supporting force of the mixing drum is applied to the multi-body dynamics simulation. The influence of the dynamic mass center position change on the rollover stability of the mixer is analyzed. Then, based on the differential braking theory, a self-tuning PID anti-rollover control algorithm optimized by improved particle swarm optimization is proposed. The research results show that when considering the influence of the dynamic center of mass, the mixer truck is more likely to roll over; the self-tuning PID algorithm iterative optimization times is reduced by 33.3%, and the optimal control parameters can be found more quickly. Finally, the effectiveness of the proposed anti-rollover control algorithm is verified through angle step conditions and Fishhook conditions. The results show that the self-tuning PID control method can more effectively prevent the rollover of the mixer truck and improve the stability of the mixer truck.

Abstract:To study the influence of the disturbance-flow fan on the passengers' thermal comfort on the subway train, the new disturbance-flow fan type B subway car occupants was taken as the research object. The human body's physiological temperature regulation model with numerical simulation loading Stolwijk was used to study the cabin crew's human thermal comfort by combining the airflow discomfort index and Berkeley thermal comfort evaluation model. Experiments verified the accuracy of the simulation model. The influence of the loading disturbance-flow fan on the human body microenvironment and various indexes were analyzed when the air supply temperature of the air conditioner in the compartment was 20 ℃. The difference of occupant index in different frequency perturbation field was analyzed. The results show that the disturbance-flow fan can improve the flow velocity and uniformity and the air distribution in the compartment, and it can optimize the human microenvironment's heat flow field. After loading the disturbance-flow fan, the occupant's overall thermal sensation is decreased by 7.3%, and the thermal comfort is increased by 0.76%. Within a certain range, the human body's thermal comfort decreases as the disturbance frequency increases, and the optimal disturbance field function frequency is 2.75 times /min.

Abstract:The paper addresses the microstructure evolution of dissimilar resistance spot welding between martensitic steel and high-strength low-alloy steel. Resistance spot welding tests were carried out on MS1400 and HSLA420 with a thickness of 1.2 mm. Shear samples were obtained by changing the welding time, current and pressure. Shear tests were carried out by a tensile testing machine. The better welding process parameters were determined by generating the pull-out failure mode. By analyzing the microstructure of fusion zone and heat-affected zone on both sides under better parameters, it is found that the fusion zone is completely martensitized. The heat-affected zone is mainly composed of mixed ferrite, martensite and granular bainite structure. The microhardness test of the joint shows that the microstructure changes in each region are basically consistent with the microhardness distribution curve.

Abstract:This paper proposes a centralized coordinated ramp merging control method for intelligent connected vehicles. First，a model of centralized coordinated ramp merging control is established. Then, the model is converted to a non-linear optimization problem through discretization，which can be solved by the NOMAD algorithm. Simulations with different randomly initialized merging conditions are performed to verify the effectiveness of the proposed method and to discuss the impact on fuel consumption. Besides，the proposed method is compared with the methods in existing literature，and the results show that the proposed method is effective under different initial merging conditions. Compared with the benchmark method，the proposed method reduces the average fuel consumption by 42.38%.