Search

Sea cucumber Apostichopus japonicas (Selenka) is one of the important aquaculture species distributed in northern China. In recent years, global warming caused frequent high temperature weather in summer in northern China, resulting in dramatic losses of the sea cucumber aquaculture industry. In the present study, we focused on the effect of oxidative stress in Apostichopus japonicus (Selenka) subjected to high temperature stress. Sea cumbers were exposed to the control (16 °C), and high temperature treatments (20 °C, 24 °C, and 28 °C) for 7 days. Then, reactive oxygen species (ROS) level, superoxide dismutase (SOD) activity, catalase (CAT) activity, peroxidase (POD) activity, reduced glutathione (GSH) content, malondialdehyde (MDA) content and 8-hydroxy-2'-deoxyguanosine (8-OHdG) level in the respiratory tree and body wall were detected, respectively. Results showed that 24 °C and 28 °C acute exposure induced the elevation of ROS level, SOD, CAT, POD activities, GSH content, MDA content and 8-OHdG level in the respiratory tree of sea cucumber. In contrast, no significant changes were observed for ROS and 8-OHdG levels in the body wall of sea cucumber, while the antioxidants including SOD, CAT, POD, and GSH decreased to some extent. Moreover, MDA content exhibited a noticeable increase in the body wall, similarly to that in the respiratory tree, indicating that high temperature could induce severe lipid peroxidation in two tissues. Considering the differences in various biomarkers measured in two tissues, respiratory tree might be more susceptible to the high temperature changes compared to the body wall. Our findings may help understand the oxidative stress response to high temperature in the respiratory tree and the body wall in A. japonicus.

Compared with space-vector PWM (SV-PWM) schemes, carrier-based PWM (CB-PWM) schemes feature less digital calculation burden. Unfortunately, since different CB-PWM schemes were not represented by a unified expression and the relationship between the injected zero-sequence components (ZSCs) and the switching-state sequences is not revealed, the CB-PWM has poor flexibility in designing the switching-state sequences. In this article, a unified carrier-based PWM (UCB-PWM) is proposed to unify the expressions of different CB-PWM schemes. The ZSC of the proposed UCB-PWM is classified into a basic ZSC (B-ZSC) and a redundant ZSC (R-ZSC). The B-ZSC is a shared component of all CB-PWM schemes, and the dwelling time of redundant small vectors can be distributed by injecting different R-ZSCs. Also, the neutral-point (NP) voltage balance control based on either continuous PWM (CPWM) or discontinuous PWM (DPWM) can be achieved by injecting R-ZSCs. Besides, not only the existing carrier-based DPWM schemes but also new carrier-based DPWM schemes are derived by the proposed UCB-PWM. Finally, experimental results verify the correctness and advantages of the proposed UCB-PWM.

Neutral-point (NP) voltage imbalance is an inherent issue of three-level converters. The alternating input current distortion is generated by using the conventional discontinuous pulsewidth modulation (DPWM) scheme without NP voltage balance control. In this article, the deviation vector between the synthesis vector and the reference vector, which is the fundamental cause of alternating input current distortion under NP voltage unbalance conditions, is revealed. Then, by optimizing the subsector division and the switching state sequence, a dynamic-space-vector DPWM (DSV-DPWM) is proposed to eliminate the deviation vector. Further, the comparisons of the proposed DSV-DPWM, the conventional DPWM, and an existing triangle carrier-based DPWM (TCB-DPWM), are addressed. Finally, a 10-kW prototype was built to verify the effectiveness of the proposed DSV-DPWM scheme. Experimental results show that, compared with the conventional DPWM scheme and the TCB-DPWM scheme, both the current THD performance and efficiencies can be significantly improved by the proposed DSV-DPWM scheme under unbalanced NP voltage conditions.

Compared with continuous pulsewidth modulation (PWM), discontinuous PWM (DPWM) is a preferred solution in high-frequency Vienna-type rectifiers due to its inherent characteristics of less commutation number and higher efficiency. In this article, redundant clamping modes at the same subsector with opposite effects on the neutral point (NP) voltage are presented. Then, a modified DPWM (MDPWM) scheme is proposed to regulate the NP voltage by using redundant clamping modes. Thus, the NP voltage can be controlled in each switching period to lower the NP voltage fluctuation. The implementation of the proposed MDPWM is given with variable power factors analysis. Further, the switching loss comparison of conventional space vector PWM (SVPWM), hybrid DPWM (HDPWM), and proposed MDPWM is presented as well. A simulation model was built to verify the proposed NP voltage balance with different modulation indices and variable power factors. Finally, a 10-kW prototype is built to evaluate the proposed MDPWM scheme at conversion efficiency and current total harmonic distortion (THD). Experimental results and analysis show that comparing conventional SVPWM and HDPWM, the efficiency of MDPWM is the highest. The THD of DPWM is slightly higher than that of SVPWM, and it is only 2.5% at rated output power.

The load of industrial users in a load-intensive urban smart park is taken as the research object, and a device-based model of energy use for smart park users is established. Considering various load constraints, the comprehensive energy user load model is constructed, the model is analyzed and the solution algorithm is determined to be the genetic algorithm. It can be seen from the calculation example that the load modeling method can realize the load optimization operation of industrial users in the smart park. The load modeling method can provide a basis for further operation scheduling, thereby reducing energy waste in smart parks.

This paper takes the load-intensive city demand aggregation modeling method as the research object, analyzes the raw data of the energy management platform of the smart park, extracts the energy consumption data that needs to be mined and processed, and processes the data from bad sources to improve the data usage efficiency. Based on the analysis of the data sources of the smart park energy management platform, energy consumption data is extracted and processed, and bad data is corrected. According to the cluster analysis method, the demand aggregation model of the park is established, and the results are used for the optimal energy dispatch of the park, which can improve the energy efficiency of the park.

Learning early-warning is of great importance to many educational domains, such as adaptive learning and personalized teaching, and has drawn numerous research attention in recent decades. In order to solve the problem of large prediction granularity in previous study. In this study, we seek to construct two novel features, including knowledge points and question types, and predict students' performance based on the two types of information. According to the predicted results, we divide the early-warning into 3 levels, and provide different levels of guidance and reminders for different warning levels of students. We did experiments based on the two types data of 141 students in Peking University. The result shows that our method has been significantly improved compared with Linear regression, RF and Adaboost. The experiment shows that the model's predicted grades and the real data Pearson correlation coefficient is 0.890568, and the accuracy of predicting warning levels is 85.81%.

In order to improve the success rate and efficiency of automatic identification of low-voltage distribution topology, this paper proposes a record-based topology identification method. The topology recognition is divided into consumer-transformer relation recognition and hierarchical recognition. High-speed power line carrier (HPLC) automatic networking was used to obtain and judge the equipment information to generate the consumer-transformer relationship in the low voltage area. The topology terminal unit and low-voltage equipment with topology recognition function generate records through specific topology signal exchanging information. Then, the intelligent distribution transformer terminal unit summarizes these records and combines the relationship between power consumers and transformer to generate the hierarchical relationship. This topology identification method has the advantages of simple control logic, fast identification speed and high identification accuracy, which can provide the important basic data for real-time fault location, power line loss analysis and area capacity estimation and other advanced applications of low-voltage distribution.

A multi-resolution control algorithm is adopted to minimize the negative effect caused by digital control delay to lower the total harmonic distortion (THD) of high frequency PWM rectifiers for aviation applications. The influence of digital control delay is analyzed first. The algorithm makes the best use of the limited digital signal processor (DSP) resources to increase the control frequency of the tasks that mainly affect the rectifier’s performance by executing the left tasks at a relatively low frequency. A 5kW prototype with full SiC devices is built to verify the effectiveness and advantages of the proposed algorithm and good experimental results have been obtained.

Facial expression recognition is one of the hotspots in the fields of computer vision and deep learning. It has very important applications in the domains of learning service recommendation, human–computer interaction and medical industry. Aiming at the problem that the traditional expression recognition method is not accurate, this paper proposes a method combining Gabor wavelet transform and convolutional neural network. Firstly, face positioning, cropping, histogram equalization and other preprocessing are performed on the expression image. Then we extract key frames of expression sequences. After that the Gabor wavelet transform is performed on the expression image to obtain magnitude and phase characteristics. Finally, we design a 2-channel CNN for training and classification. The experiment achieves an accuracy of 96.81% on the CK+ database and it has a certain improvement compared with the Gabor wavelet transform and the traditional CNN alone.