In order to capture more sentence semantic information from the extracted shared sentiment features for cross-domain sentiment analysis, a deep network model based on domain adversarial mechanism and BERT(bidirectional encoder representations from transformers)was proposed. The model firstly used BERT to obtain the semantic representation vectors of sentences, and then extracted the local features of sentences with a convolutional neural network. A domain adversarial neural network was designed to make the representations of features extracted from different domains to be as indistinguishable as possible, that was, the features extracted from source domain and target domain had much more similarities; and a sentiment classifier was trained on the source domain dataset with sentiment labels, and it was expected that the trained sentiment classifier would have good classification performance in the source domain, and in the target domain. The experimental results on Amazon product reviews dataset showed that the proposed method achieved the expectation and was competent for achieving cross-domain text sentiment classification.

Aiming at the fact that the background was dominant in the key frame extraction algorithm, in which the foreground target was too small and it was not easy to extract the features of moving targets in sports video, a key frame extraction algorithm for foreground moving target feature extraction based on background modeling algorithm was proposed, which was called visual background extractor(ViBe)algoritm. The foreground target detection of video sequence was firstly carried out using ViBe algorithm, afterwards the scale-invariant feature transformation(SIFT)features of the foreground moving target were extracted. Based on the similarity calculated from video frame series, the key frames of video were output according to the key frame discrimination method. The experimental results showed that the proposed algorithm could solve the problem of missed selection and misselection in traditional key frame extraction. Compared with the algorithm based on SIFT distribution histogram, the F₁ score was well improved. The algorithm based on ViBe could effectively identify key frames in sports video.

Most existing entity alignment methods typically relied on external information and required expensive manual feature construction to complete alignment. Knowledge graph-based methods used only semantic information and failed to use structural information. Therefore, this paper proposed a new entity alignment method based on adaptive attribute selection, training an entity alignment model based on the joint embedding of the two knowledge graphs, which combined the semantic and structural information. Also, this paper proposed the use of strong attribute constraint based on adaptive attribute selection, which could adaptively generate the most effective attribute category and weight, to improve the performance of entity alignment. Experiments on two realistic datasets showed that, compared with traditional methods, the precision of the proposed method was improved by 11%.

The traditional collaborative filtering recommendation algorithm failed to reflect short-term user interest. In order to reflect the short-term interests of users better, a collaborative filtering recommendation algorithm based on Gated Recurrent Unit(GRU)neural network with active learning was proposed. Based on the GRU neural network, the algorithm processed the data into time-series data to reflect the change of the users interest and used active learning to sample high-quality data dynamically for accelerating the training of GRU neural network. The result on MovieLens1M dataset showed that the GRU model with active learning could obtain higher short-term prediction success rate, recall rate, item coverage, and user coverage compared with the user-based collaborative filtering method(UCF), the markovian chain model-based collaborative filtering method(MC)and the matrix factory-based collaborative filtering method(LFM), so it could effectively predict the short-term interest of users, improve the accuracy, discover the long-tail items. Meanwhile, it could achieve the same effect with fewer iterations compared with the original GRU model.

To solve problems of slow detection and lack of behavioral safety analysis in intelligent nursing, a risk assessment method based on spatial hidden danger distribution and motion intention analysis was proposed. The k-means++ algorithm and background elimination method were used to enhance the object detection capability of YOLO(you only look once)v3, which led to the classification and localization of human targets as well as dangerous objects. The Kalman filter was used to predict the moving trajectory, where two parameters, namely the deviation angle of motion and the distance between the human target and danger, were used to construct the human movement patterns. The risk assessment model was established, and the degree of danger was evaluated according to different movement behavior patterns. Experimental results showed that for identifying different objects in the test set, both the detection precision and the recall rate of the enhanced YOLOv3 algorithm were over 95%. An increasement of IOU(intersection over union)at 7% was witnessed, and frames rate reached 31.3 frames/s. These results proved the real-time performance of the system. Since the proposed risk assessment model incorporated motion intentions of the human target, this method was expected to boost the performance in fitting the risk progression of different movement patterns, making the risk assessment more reasonable.

The problem of construction expansion online was studied for a class of nonaffine nonlinear interconnected large-scale systems. An adaptive neural decentralized connective stabilization method was proposed for expansion online of large-scale systems based on backstepping technique. By using neural network adaptive technique, the decentralized controller of the newly added subsystem was designed under the condition that the decentralized control laws and adaptive laws of the original system were kept to be unchanged. The interconnection parts caused the a new subsystem were dealt with in the controller of the new subsystem. An adaptive decentralized connective stabilization controller for the newly added subsystem was obtained. The controller could ensure that all the signals both in the newly added closed-loop nonaffine subsystem and the resultant expanded closed-loop large-scale system were uniformly ultimately connective bounded. The simulation results were given to verify the effectiveness of the proposed control method.

The current fieldbus control system usually only supports one or two kinds of fieldbus protocols. In view of the current situation that there were many standards of fieldbus protocols and many types of bus devices, a multi-protocol heterogeneous compatible fieldbus control system with three layers of monitoring level, control level and field equipment level was constructed for the purpose of supporting multiple protocols. By configuring gateway to solve the communication problems of heterogeneous fieldbus between PROFIBUS-DP and PROFIBUS-PA, HART, Modbus, DeviceNet, GPRS, etc., a fieldbus monitoring system with Siemens PCS7 as the upper monitor system and heterogeneous controller as the field control system was built, and the wireless and real-time embedded remote monitoring function of FCS and fieldbus intelligent instrument was realized by GPRS. The test results showed that the monitoring level could monitor and control heterogeneous fieldbus devices with different protocols remotely and real-time, and realize remote management and start-stop control of fieldbus devices. This system could be used as the feasibility verification of multi-protocol heterogeneous compatible FCS.

In view of the increasing dependence of energy planning on energy demand forecasting and the difficulty of energy demand forecasting, this paper analyzed various energy forecasting methods and discussed the direction of energy development. The article analyzed the current demand situation of energy development methods from the direction of global energy demand development in recent years. The existing main energy forecasting methods were summarized and compared. The advantages and disadvantages of the existing research methods and applicable occasions were summarized. Combined with the new direction of energy development, the future development prospects of energy forecasting were given. Furthermore, this paper applied the LEAP model to predict the energy demand of the African region, and analyzed the regional energy complementation effect and the role of “electricity substitution” in the development of energy demand.

In order to improve the high security of power network data transmission, an end-to-end security encryption scheme of NB-IoT(narrow band internet of things, NB-IoT)for smart grid based on physical unclonable function and domestic cipher algorithm SM3 was proposed in this paper. A self-controllable NB-IoT application layer security architecture was designed by introducing the SM3, extending the existing key derivation structure of LTE, and combining the physical unclonable function to ensure the generation of encryption keys between NB-IoT terminals and power grid business platforms. Analysis and experiment showed that the proposed scheme realized secure data transmission and bidirectional identity authentication between IoT devices and terminals. Its features included high compatibility, low communication costs, lightweight and flexible key update. In addition, the scheme also supported terminal authentication during key agreement, which furtherly enhanced the security of business systems in smart grid.

The central air conditioner had the potential to cope with the DC blocking fault through emergency control, ensure the safe and stable operation of the receiving-end grid after suffering from large power shortage. The basic principle of the central air conditioner was introduced. On this basis, the physical model of the central air conditioner which included central air conditioning unit, temperature change of frozen inflow and outflow water, heat exchange between the cooling water of the fan coil and the room, indoor average temperature change as well as the proportion of the room in the open state were established. An emergency control strategy for direct power cut and flexible recovery of central air conditioning was proposed. The feasibility of the emergency control of the central air conditioning system in response to stability control and under frequency/voltage load shedding were discussed respectively. The characteristics of the central air conditioning emergency control were simulated. The emergency control of central air conditioner was simulated after HVDC blocking fault occurs in Shandong power grid, verifying that the power grid frequency could be increased by 0.04 Hz when central air conditioners accounted for 1% of the total load in Shandong power grid.

Wind power ramp events(WPRE)could easily destroy the active power balance of the power system, as well as did harm to the frequency stability and power quality, which threatened the safe and stable operation of power grid. A novel imprecise conditional probability prediction approach was proposed based on the credal network(CN), which could provide the interval range of the occurrence probability of each WPRE state. The approach excavated the dependence relationships between WPRE and meteorological variables using the greedy search algorithm, and constructed a CN structure to express the relationships in an abstract way. The proposed approach extended the imprecise Dirichlet model(IDM)on hyperparameter settings to quantify the uncertain conditional dependences among the variables, thus to realize the parameter estimation of the CN. Based on the constructed CN model, a CN probability inference algorithm was employed to estimate the imprecise probability distribution of the multi-state WPRE. The case study with wind-farm operating measurements in Ningxia Province demonstrated that the proposed approach had excellent performance even under the prediction scenarios with insufficient samples.

With the improvement of public awareness of environmental protection, the discharge of industrial wastewater became a crucial issue in industrial production. The typical water quality models were based on static models which ignored the dynamic information in process variables, resulting in the reduction in the accuracy of model prediction and the generalization ability of the models. Considering the time-varying and dynamic characteristics of process variables, a time difference model embedded into canonical correlation analysis was proposed in this paper. The effect of the order of the time difference model on the prediction accuracy was also analyzed. Compared with the traditional canonical correlation analysis, the root mean square error values of effluent chemical oxygen demand and effluent total nitrogen were reduced from 1.502 8 to 0.564 5 and from 2.344 0 to 1.192 6, respectively. The correlation coefficient values were increased from 0.422 7 to 0.847 0 and from 0.405 9 to 0.793 6, respectively. The results indicated that the prediction accuracy and generalization ability of the model were both improved.

The development process and its applications of titanium-based coagulant were reviewed, including the mono-titanium coagulants and poly-titanium coagulants, and the single titanium coagulants and composite titanium coagulants, and the recent research process and the future development of titanium coagulants, which could provide guidances and references for the research and development of titanium-based coagulants.

In order to make up the deficiency of the traditional measuring method in monitoring the whole deformation and real-time monitoring of the bridge and overcome the defect of digital photography in monitoring the three-dimensional deformation of a bridge, this paper presented the image matching-time baseline parallax method: a digital camera was set up in the proper place, and a reference plane, consisting of reference points, was not far from the digital camera. Reference plane was perpendicular to the photographic direction, and the monitoring points were on the bridge. The two-dimensional deformation on the object plane of the deformation point was obtained based on image matching-time baseline parallax method, then the horizontal deformation on object plane was disassembled into the bridge direction and the direction perpendicular to bridge direction based on the principle of three-dimensional deformation monitoring, The three-dimensional deformation of the bridge was got. Results showed that measurement accuracy of the 山 东 大 学 学 报 (工 学 版)第50卷 - 第1期于承新,等:基于数字测量技术的桥梁监测及预警系统 \=-bridge was 0.72 mm and 1.16 mm in X and Z direction on object plane, respectively. It could meet accuracy requirements of deformation monitoring. As Phoenix mountain road-bridge showed good flexibility in X, Y, Z and comprehensive direction, and the absolute of the maximum positive and negative deflection was 7.57 mm, which was approximately equal to 1/8 of the allowable deflection of the bridge(L/1 000), and the bridge was in good health. Bridge health monitoring and warning system could achieve the integration of data acquisition, storage, processing and display. The deformation curves could show the deformation trend of the bridge and effectively warn the potential danger.

The liquid-liquid phase separation of Al₆₅Bi₂₈Cu₇ monotectic alloy melt cast in copper mould was investigated by resistivity method. The formation mechanism of core-shell structure of Al₆₅Bi₂₈Cu₇ monotectic alloy was discussed. The results showed that the anomalous changes in ρ-T curve confirmed the occurrence of liquid phase separation, monotectic reaction and eutectic reaction in the solidification of Al₆₅Bi₂₈Cu₇ monotectic alloy melt. The anomalous change above monotectic temperature in ρ-T curve should be attributed to the concentration fluctuation of melt. The core-shell structure of Al-rich core covered by Bi-rich could form in the Al₆₅Bi₂₈Cu₇ monotectic alloy.