Aiming at the problem of low information utilization rate of sentence dependency tree and poor feature extraction effect in relation extraction task, an Attention-guided Gate perceptual Graph Convolutional Network (Att-Gate-GCN) model was proposed. Firstly, a soft pruning strategy based on the attention mechanism was used to assign weights to the edges in the dependency tree through the attention mechanism, thus mining the effective information in the dependency tree and filtering the useless information at the same time. Secondly, a gate perceptual Graph Convolutional Network (GCN) structure was constructed, thus increasing the feature perception ability through the gating mechanism to obtain more robust relationship features, and combining the local and non-local dependency features in the dependency tree to further extract key information. Finally, the key information was input into the classifier, then the relationship category label was got. Experimental results indicate that, compared with the original graph convolutional network relation extraction model, the proposed model has the F1 score increased by 2.2 percentage points and 3.8 percentage points on SemEval2010-Task8 dataset and KBP37 dataset respectively, which makes full use of effective information, and improves the relation extraction ability of the model.

Concerning the problem that the research methods of relation extraction for distant supervision datasets have a lot of label noise, a dynamic label method applied to the hierarchical attention mechanism relation extraction model was proposed. Firstly, a concept of generating dynamic label based on the similarity of relation categories was proposed. Since the same relation labels contain similar feature information, calculating the similarity of relation categories of feature information is helpful to generate the dynamic label corresponding to the feature information. Secondly, the scoring function of the dynamic label was used to evaluate whether the distant supervision label was noise and to determine whether a new label was needed to generate to replace the distant supervision label, and the influence of label noise on the model was suppressed by adjusting the distant supervision label. Finally, according to the dynamic label, the hierarchical attention mechanism was updated to focus on the effective instances, the importance of each effective instance was relearned and key relation feature information was further extracted. The experimental results indicate that, compared with the original hierarchical attention mechanism relation extraction model, the proposed method has the Micro and Macro scores increased by 1.3 percentage points and 1.9 percentage points respectively, realizes the dynamic correction of the noise label, and improves the relation extraction ability of the model.

Concerning that Cuckoo Search algorithm for Multi-objective Optimization (MOCS) has slow speed in the late iteration and being easy to fall into the local optimum, a new MOCS based on Chaos Cloud Model (CCMMOCS) was proposed. In the evolutionary process, chaos theory was used to optimize the positions of general nests in order to avoid falling into the local optimum; then the cloud model was used to optimize the position of some better nests to improve the accuracy; finally the better value of them was chosen as the best value for optimization. The simulation experiments on five general test functions in error estimated value and diversity index show that CCMMOCS is much better than MOCS, Particle Swarm Optimization algorithm for Multi-objective Optimization (MOPSO) and NSGA-Ⅱ. Its Pareto fronts are closer to the ideal curve than those of other algorithms and the distribution is more uniform.

Concerning slow convergence speed in the later stage, low calculation accuracy and easily falling into the local optimum of basic Cuckoo Search (CS) algorithm, a Cuckoo Search based on Chaos theory (CCS) algorithm was proposed. Firstly, the chaos initialization was used to increase population diversity. Secondly, the chaos disturbance operator was introduced to the local optimal value to jump out of the premature convergence and improve the calculation accuracy. Finally, the global optimization was improved. Four single objective benchmark functions were tested. The simulation results in the best, the worst, average, median and standard deviation value show that CCS algorithm has faster convergence speed and higher convergence precision than CS algorithm. Harmonic is the vital cause of the distortion of current waveform and voltage instability. The analysis of harmonics in power quality analysis is a very important part in power system. The CCS algorithm was applied to harmonic estimation. The experimental results show that the CCS algorithm has better performance compared with the Particle Swarm Optimization (PSO) according to the analysis of harmonic current in mean value and standard deviation.

The data acquired by the satellite positioning receiver in the trajectory correction projectile must be filtered in real-time to predict the point. The calculation of traditional filtering method is time-consuming, and is difficult to meet the requirements of real-time filtering. A kind of extended Kalman filtering algorithm based on polynomial fitting was proposed. The data of projectile flight in the time interval was replaced by the fitting interpolation data. In this way the filter frequency could be reduced. Simulation results show that the computation time of the proposed method can be reduced by 7/8 compared to traditional extended Kalman filtering without reducing the filtering precision, and the real-time performance is improved. This method can provide important reference for the research of key technology of trajectory correction projectile. At the same time, the method can be applied to other filtering algorithms, and has a strong portability.

In Wireless Sensor Network (WSN) clustering routing algorithm, sensors energy consumption imbalance will result in "energy hole" phenomenon, and it will affect the network lifetime. For this problem, an energy-balanced unequal clustering routing protocol based on game theory named GBUC was put forward. In clustering stage, WSNs were divided into clusters of different sizes, the cluster radius was determined by the distance from cluster head to sink node and the residual energy. By adjusting the cluster head in the energy consumption of communication within the cluster and forwarding data to achieve energy balance. In inter-cluster communication phase, a game model was established by using the residual energy efficiency and link reliability as the benefit functions, using its Nash equilibrium solution to get joint energy balancing, optimal transmission path of link reliability, thereby improving network performance. The simulation results show that, compared with Energy-Efficient Uneven Clustering (EEUC) algorithm and Unequal Clustering Energy-Economical Routing (UCEER) algorithm, the GBUC algorithm has significantly improved the performance in balancing node energy consumption and prolonging the network lifetime.

In the real-time large data applications of Peer-to-Peer (P2P), to avoid free-riding behavior in the Video on Demand (VOD) system, a new incentive scheme of auction algorithm based on the discriminatory second price was proposed. The nodes obtained the video data block they needed using distributed dynamic auction between nodes. In auction, the bidding node firstly determined whether the budget was enough to bid based on discrimination rule, and set the upload bandwidth according to the number of bidding nodes. Secondly, the winner node was determined by the bid price. Finally, the bidding node paid the auction node according to the second highest price after it got the data block as its income. Analysis of the revenue of nodes, the bandwidth utilization and the proportion of selfless or selfish nodes indicate that the proposed scheme can effectively motivate nodes to take active part in sharing of video data blocks, and make efficient use of the upload bandwidth at the same time.

Concerning the poor performance in anti-noise and the limitations of the fixed partition of the phase-grouping method, an adaptive phase partitioning method was proposed. Firstly, this paper used the adaptive smoothing and Wallis filtering for image pre-processing and adopted the Canny method for edge detection; secondly, selected the optimal partitioning of the image based on the adaptive phase partitioning method to create the line supporting area; finally, fitted and merged the lines by the least squares method. The experimental results show that the proposed method can effectively solve the line breakage and mistaken extraction caused by the noise and fixed partition, and this method extracts line completely and precisely.

The slices of BPEL obtained are not complete if traditional program slicing technique is used. Therefore, a static BPEL program slicing technique based on BPEL program dependence graphs was proposed. This technique computed slices based on BPEL program dependence graphs, which were created according to the characteristics of BPEL. The results of case analysis prove that the technique based on BPEL program dependence graphs can obtain more complete slices, and thus giving software engineers more help to test, debug and maintain BPEL programs.